Ph.D. Thesis
On he Empi ical Cha ac e iza ion o he Low
Vol age Dis ibu ion G id as a T ansmission
Medium o Na owband Powe Line
Communica ions (9-500 kHz)
Au ho : Jon González-Ramos
Supe iso s: D . I zia Angulo Pi a
D . Igo Fe nández Pé ez
2025
Ph.D. Thesis
On he Empi ical Cha ac e iza ion o he Low
Vol age Dis ibu ion G id as a T ansmission
Medium o Na owband Powe Line
Communica ions (9-500 kHz)
Au ho : Jon González-Ramos
Supe iso s: D . I zia Angulo Pi a
D . Igo Fe nández Pé ez
2025
(cc) 2025 Jon González-Ramos (cc by-nd 4.0)
i
ii
Abs ac
Na owband Powe Line Communica ion (NB-PLC) echnologies exploi he exis ing
Low Vol age (LV) dis ibu ion g id o ca ying da a signals, co e ing he 3-500 kHz
equency band. Al hough i o e s se e al ad an ages o e o he wi ed o wi eless
al e na i es, he equency- and ime-dependen ansmission medium can jeopa dize he
p ope pe o mance o NB-PLC sys ems.
In ecen yea s, a g ea deal o e o has been de o ed by he scien i ic communi y o
cha ac e ize he elec ical g id as a communica ion channel. Howe e , due o he elen less
e olu ion o he powe ne wo k, mainly ela ed o he expec ed inc ease in he numbe o
Elec ic Vehicles (EVs) and Dis ibu ed Ene gy Resou ces (DERs), he p opaga ion
medium is expec ed o ha e changed signi ican ly. The e o e, a p ope analysis o he
cu en and u u e cha ac e is ics o he LV g id as a ansmission medium is s ill needed.
In his con ex , his Doc o al Thesis aims o add ess wo main objec i es. Fi s ,
cha ac e ize he elec ical g id as a communica ion channel in e ms o he Non-In en ional
Emissions (NIEs), he g id access impedance, and he a enua ion in he 9-500 kHz
equency band and, second, e alua e NB-PLC echnologies acco ding o PRIME 1.4
s anda d unde di e en channel condi ions.
Conside ing he high pene a ion o EVs expec ed in he coming yea s, his Doc o al
Thesis speci ically ocuses on cha ac e izing he elec ical g id in he p esence o EV
Cha ging P ocesses (EVCPs). Conce ning he conduc ed emissions, his Doc o al Thesis
co e s h ee main aspec s. Fi s , he emissions gene a ed by EVCPs unde isola ed
condi ions, i.e., using a Line Impedance S abiliza ion Ne wo k (LISN), a e quali a i ely
cha ac e ized in bo h he equency and ime domains. Second, a no el p ocedu e o he
quan i a i e e alua ion o he NIEs gene a ed by EVCPs is p esen ed, in addi ion o
analyzing hei in e ac ion and p opaga ion h ough a con olled LV g id. Finally, he
emissions egis e ed du ing he cha ging o se e al EVs unde isola ed and on-line
condi ions (when measu ing di ec ly in he LV g id) a e compa ed by applying he
p e iously de ined me hod. Wi h ega d o he g id impedance, his Doc o al Thesis deals
wi h he sub-cycle, mean, and long- e m g id impedance a ia ions due o EVCPs by
means o wo measu emen campaigns ca ied ou in F ance and Aus ia.
Rega ding he e alua ion o he pe o mance o NB-PLC echnologies unde di e en
channel condi ions, his Doc o al Thesis co e s h ee main a eas o esea ch. Fi s , he
in luence o he spec al cha ac e is ics o he p e iously cha ac e ized conduc ed
emissions gene a ed by EVCPs is indi idually e alua ed h ough labo a o y ials. Simila ly,
second, he po en ial e ec s o impedance a ia ions a e add essed, conside ing bo h
equency-dependen and sub-cycle g id impedance a ia ions. Finally, in o de o e alua e
he pe o mance o PRIME 1.4 in a si ua ion close o eal g id condi ions, his Doc o al
Thesis in es iga es he po en ial deg ada ion o he communica ions in a con olled LV
g id in Aus ia.
In conclusion, his Doc o al Thesis deals wi h he cha ac e iza ion o he LV dis ibu ion
g id as a ansmission channel and he e alua ion o he pe o mance o PRIME 1.4
unde di e en channel condi ions, in o de o ake ad an age o he powe ne wo k o
NB-PLC.
iii
Labu pena
NB-PLC eknologiak behe- en sioko banake a-sa ea e abil zen du 3 kHz e a 500 kHz
bi a eko maiz asun-bandan da u-seinaleak ga aia zeko. Ha izko edo ha i gabeko bes e
auke a ba zuen aldean hainba aban aila di uen a en, maiz asuna en e a denbo a en
mendeko ansmisio-bideak zaildu egin dezake NB-PLC sis emen un zionamendu egokia.
Azken u eo an, komuni a e zien i ikoak ahalegin handia egin du BT banake a-sa ea
komunikazio-kanal gisa ka ak e iza zeko. Hala e e, sa e elek ikoa en e engabeko bilakae a
dela e a, ba ez e e ibilgailu elek ikoen e a bana u ako ene gia baliabideen kopu ua en
hazkundea ekin lo u a, komunikazio-kanala naba men alda uko dela espe o da. Be az,
beha ezkoa da behe en sio sa ea en egungo e a e o kizuneko ezauga iak beha bezala
az e zea, ansmisio-bide gisa.
Tes uingu u ho e an, Dok o ego Tesi honek bi helbu u nagusi jo a u nahi di u. Lehenik
e a behin, sa e elek ikoa komunikazio-kanal gisa ezauga i zea, nahigabeko emisioen,
sa ea en sa bide-inpedan zia en e a a enuazioa en e minoe an, 9-500 kHz-eko maiz asun-
bandan, e a, biga enik, NB-PLC eknologia ebalua zea PRIME 1.4 es anda a en
a abe a, kanal-baldin za desbe dine an.
Da ozen u ee an ibilgailu elek ikoen ba ne a ze handia espe o dela kon uan ha u a,
Dok o ego Tesi honek sa e elek ikoa en ka ak e izazioa ibilgailu elek ikoak ka ga zeko
p ozesue an oina i uko du. Emisioei dagokienez, Dok o ego Tesi honek hi u alde di
nagusi bil zen di u. Lehenik e a behin, baldin za isola ue an ibilgailu elek ikoak ka ga zeko
p ozesuek so u ako emisioak kuali a iboki ezauga i zen di a, hau da, LISN ba e abiliz,
bai maiz asuna en e emuan, bai denbo a en e emuan. Biga enik, p ozedu a be i zaile ba
au kez en da ibilgailu elek ikoak ka ga zeko p ozesuek so u ako nahigabeko emisioen
ebaluazio kuan i a ibo ako, e a, ho ez gain, haien in e akzioa e a hedapena az e zen di a
behe en sio sa e kon ola u ba en bidez. Azkenik, baldin za isola ue an e a behe en sio
sa ean zuzenean neu zean hainba IEk ka ga zean e egis a u ako emisioak alde a zen
di a, au ez zehaz u ako me odoa aplika uz. Sa ea en inpedan zia i dagokionez, Dok o ego
Tesi honek ibilgailu elek ikoak ka ga zeko p ozesuek e agindako inpedan zia en
azpizikloko, ba ezbes eko e a epe luzeko aldake ak jo a zen di u. Ho e a ako, bi
neu ke a-kanpaina egi en di a F an zian e a Aus ian.
NB-PLC eknologien e endimendua hainba kanal-baldin za an ebalua zea i dagokionez,
Dok o ego Tesi honek hi u ike ke a-a lo nagusi bil zen di u. Lehenik e a behin, au ez
ka ak e iza u ako ibilgailu elek ikoak ka ga zeko p ozesuek so u ako emisioen ezauga i
espek alen e agina ebalua zen da, labo a egiko saiakun zen bidez. E a be ean, biga enik,
inpedan zia-aldakun zen e ek uei helduko zaie, maiz asuna en mendeko inpedan zia-
aldakun zak zein azpizikloko sa eko inpedan zia-aldakun zak kon uan ha u a. Azkenik,
PRIME 1.4 en e endimendua sa ea en baldin za e eale a ik hu bil dagoen egoe a ba ean
ebalua zeko, Dok o ego Tesi honek komunikazioen deg adazio po en ziala ike zen du
behe en sioko sa e ba ean.
Ondo ioz, sa e elek ikoak NB-PLC ako eskain zen di uen auke ak maximiza zeko asmoz,
Dok o ego Tesi honek behe en sioko banake a-sa ea en ka ak e izazioa ansmisio-kanal
gisa e a PRIME 1.4 en e endimendua en ebaluazioa jo a zen di u, kanal-baldin za
desbe dine an.
i
Resumen
NB-PLC es una ecnología que u iliza la exis en e ed de dis ibución de baja ensión pa a
anspo a señales de da os en la banda de ecuencias de 3 kHz a 500 kHz. Aunque o ece
di e sas en ajas sob e o as al e na i as cableadas o inalámb icas, el medio de ansmisión,
dependien e de la ecuencia y el iempo, puede di icul a el co ec o uncionamien o de
los sis emas NB-PLC.
En los úl imos años, la comunidad cien í ica ha dedicado un g an es ue zo a ca ac e iza
la ed de dis ibución de baja ensión como canal de comunicación. Sin emba go, debido
a la incesan e e olución de la ed eléc ica, p incipalmen e elacionada con el aumen o del
núme o de ehículos eléc icos y ecu sos ene gé icos dis ibuidos, se espe a que el medio
de p opagación haya cambiado signi ica i amen e. Po lo an o, es necesa io un análisis
adecuado de las ca ac e ís icas ac uales y u u as de la ed de baja ensión como medio de
ansmisión.
En es e con ex o, es a Tesis Doc o al p e ende abo da dos obje i os p incipales. En
p ime luga , ca ac e iza la ed eléc ica como canal de comunicación en é minos de las
emisiones no in encionadas, la impedancia de acceso de la ed y la a enuación, en la banda
de ecuencias 9-500 kHz y, en segundo luga , e alua las ecnologías NB-PLC según el
es ánda PRIME 1.4 bajo di e en es condiciones de canal.
Teniendo en cuen a la al a pene ación de ehículos eléc icos que se espe a en los
p óximos años, es a Tesis Doc o al se cen a especí icamen e en la ca ac e ización de la
ed eléc ica en p esencia de p ocesos de ca ga de ehículos eléc icos. En cuan o a las
emisiones conducidas, es a Tesis Doc o al cub e es aspec os p incipales. En p ime luga ,
se ca ac e izan cuali a i amen e las emisiones gene adas po los p ocesos de ca ga de
ehículos eléc icos en condiciones aisladas, es deci , u ilizando una LISN, an o en el
dominio de la ecuencia como en el del iempo. En segundo luga , se p esen a un
no edoso p ocedimien o pa a la e aluación cuan i a i a de las emisiones no in encionadas
gene adas po los p ocesos de ca ga de ehículos eléc icos, además de analiza su
in e acción y p opagación a a és de una ed de baja ensión con olada. Po úl imo, se
compa an las emisiones egis adas du an e la ca ga de a ios ehículos eléc icos en
condiciones aisladas y al medi di ec amen e en la ed de baja ensión, aplicando el mé odo
p e iamen e de inido. En cuan o a la impedancia de ed, es a Tesis Doc o al abo da las
a iaciones sub-ciclo, media y a la go plazo de la impedancia debidas a los p ocesos de
ca ga de ehículos eléc icos. Pa a ello, se ealizan dos campañas de medida en F ancia y
Aus ia.
En cuan o a la e aluación del endimien o de las ecnologías NB-PLC en di e en es
condiciones de canal, es a Tesis Doc o al aba ca es á eas p incipales de in es igación. En
p ime luga , se e alúa indi idualmen e, median e ensayos de labo a o io, la in luencia de
las ca ac e ís icas espec ales de las emisiones conducidas gene adas po p ocesos de ca ga
de ehículos eléc icos p e iamen e ca ac e izadas. Asimismo, en segundo luga , se
abo dan los e ec os de las a iaciones de impedancia, conside ando an o las a iaciones
de impedancia dependien es de la ecuencia como las a iaciones sub-ciclo de impedancia.
Finalmen e, pa a e alua el endimien o de PRIME 1.4 en una si uación ce cana a las
condiciones eales de la ed, es a Tesis Doc o al in es iga la deg adación po encial de las
comunicaciones en una ed de baja ensión con olada en Aus ia.
En conclusión, con el in de maximiza las opo unidades o ecidas po la ed eléc ica
pa a NB-PLC, es a Tesis Doc o al abo da la ca ac e ización de la ed de dis ibución de
baja ensión como canal de ansmisión y la e aluación del endimien o de PRIME 1.4
bajo di e en es condiciones de canal.
i
Acknowledgmen s
Desde que empecé a edac a el documen o de esis, lle o pensando en que es e momen o
llega ía, el de esc ibi es os ag adecimien os y ecapi ula odo lo que ha supues o pa a mí
es a e apa de casi es años y medio. Espe o habe encon ado las palab as que desc iban
lo ag adecido que es oy po el apoyo que he sen ido du an e es e iempo.
En p ime luga , quie o da las g acias al equipo de la línea Txispas. T abaja con oso os
es un e dade o place y espe o segui haciéndolo en los p óximos años. Da id, g acias
po u ce canía, po siemp e es a dispues o a e isa un a ículo, incluso en us a os lib es
los ines de semana, y po us labo es de ges ión en la línea. Amaia, g acias po u buen
humo y us b omas, esos a os de co illeo en los que uno se en e a de odo y, sob e odo,
po es a siemp e abie a a echa me una mano. Quie o ag adece ambién a Alex, con quien
he podido compa i es e camino desde que empezamos en TSR. G acias po odas esas
con e saciones mañane as sob e nues o u u o (y esas con e saciones de Txisbec) y, po
supues o, po ayuda me con las in ini as ges iones que he enido que hace odos es os
años. A Ja ie , mucha sue e en u esis en ma cha, no engo dudas de que sald án g andes
esul ados de ella. A Mikel, g acias po odas esas con e saciones u bolís icas de los ca és
y po odo el iempo que has dedicado en las campañas de medidas sin las cuales es a esis
no exis i ía. Quie o ag adece ambién a Idu e, po su ce canía y po el buen abajo de
es os p ime os años de in es igación. Espe o que podamos segui abajando jun os en el
u u o. Y, po supues o, a mis di ec o es de esis. A Igo , g acias po el buen ollo que
siemp e ansmi es, po es a siemp e dispues o a echa una mano y, cómo no, po ese
ma a illo sis ema de medidas sin el cual es a esis no se ía posible. G acias a us clases de
TC, en é hace más de cinco años en TSR y es oy hoy a pun o de de ende es a esis
doc o al. Y, po úl imo, a I zia , g acias po enseña me u pasión po la in es igación y po
el abajo bien hecho. Sin us e isiones minuciosas y llenas de colo , es e documen o de
esis no se ía como es. Pe o, sob e odo, quie o da e las g acias po se siemp e un apoyo
pa a mí, po aconseja me y p eocupa e po mí, y po guia me en es e camino como lo has
hecho. G acias po odas esas con e saciones sob e la ida y el u u o, me han ayudado
mucho.
Quie o da las g acias ambién al es o de miemb os de TSR. Especialmen e, a Te esa, po
odas esas ho as de du o gimnasio y po esas in ini as con e saciones más allá del abajo.
G acias po es a siemp e ahí pa a mí, especialmen e en los momen os más di íciles.
G acias ambién a mis dos on os, Iñigo y Endika, po aleg a me los días en TSR. Sin
oso os mis cien nomb es no exis i ían. Endika, g acias po odos los buenos momen os
que hemos pasado jun os ue a y den o de TSR, es oy segu o de que oda ía nos quedan
muchos po i i . Iñigo, g acias po odas esas con e saciones sob e la ida de los úl imos
años. Espe o que nos queden muchos pali os po oma en los que eco da que “Hay un
loco en In e ne ”. Quie o ag adece ambién a Eneko, po las isas de los ca és y odas esas
con e saciones ue a de TSR. Espe o que podamos segui co illeando y “c i icando”
jun os muchos años. A JoM, po es a siemp e dispues o a ayuda y aleg a nos los ca és
con sus his o ias y sus memes. G acias ambién po esos lunes de pa xanga y, especialmen e,
po los kalimo xos e apéu icos p e ios, han sido un g an apoyo pa a mí. A Ma a, po el
buen ollo que siemp e c eas y po aleg a los ca és con esas pullas de las que nadie se
puede lib a . Al es o de doc o andos de TSR (E ick, O lando, D eyelian, Alejand o,
ii
Elizabe ), mucha sue e en lo que os queda de esis, es oy segu o de que ha éis un abajo
excelen e.
Me gus a ía ag adece ambién a Be nha d, po acoge me en su Uni e sidad y po da me
la opo unidad de conoce una ciudad como Viena. G acias po u ayuda en esa e apa de
es a esis, espe o que sigamos colabo ando en el u u o.
Quie o da ambién las g acias a mis amigos, que me han acompañado a lo la go de odos
es os años. P ime o, a mis elecos (Vicen e, Ad ián, C is ina, Jon Ma cos, Ainhoa, Rada,
Ixone, Xabi, And ea, Lau a, Izaskun, Andoni), po odos los buenos momen os que hemos
pasado du an e y después de la ca e a. Sin oso os, p obablemen e, no es a ía donde
es oy. Quie o da ambién las g acias a Sendoa, po odos esos ma es/mié coles de
Champions y sábados/domingos de P emie , y esas in ini as con e saciones u bolís icas con
pipas y pizzas. G acias ambién a mis amigos de San a Pola (Eli, Da id, Ignacio, Alex…)
po aleg a me las acaciones de e ano y hace que los meses de desconexión sean más
di e idos. A mis mad ileños (Paula, Beli, Rapo, Ne ea, Bea, Ka en y F an) y allisole ana
(Elisa) a o i os, g acias po esos iajes po ie as palen inas, en los que comenzó una
g an amis ad. Es oy segu o de que oda ía nos quedan muchos Villoldos (espe emos que
sin sus os) po delan e.
Quie o ag adece ambién a Josu, po se como un he mano pa a mí. Es oy segu o de que
los años no cambia án que sigamos compo ándonos como niños de cinco (😉) años.
G acias a Ma ian y Fe nan, po se siemp e un apoyo y hace me sen i uno más de la
amilia.
Po úl imo, quie o ag adece a mi amilia. A mis pad es, Rosa y Gonzalo, po apoya me
en odas las decisiones que he omado en la ida, habéis sido un pila undamen al. G acias
ambién po ansmi i me ues os alo es y po enseña me que es necesa io insis i pa a
consegui lo que uno busca. Sin oso os, hoy no se ía quien soy. Y a Malena, g acias po
se mi g an apoyo. Po aguan a me en mis días buenos y po sopo a me y aleg a mis días
malos. G acias po es a siemp e, po impulsa me a se mejo . C eo que no engo palab as
pa a desc ibi odo lo que signi icas pa a mí. Nunca e es a é lo su icien emen e ag adecido.
xi
Lis o Tables
Table I. Summa y o he jou nal pape s associa ed o he con ibu ions o his Doc o al
Thesis............................................................................................................................................... 24
Table II. Compa ison be ween cu en NB-PLC p o ocols and s anda ds [1], [56]. ........ 27
Table III. Compa ison be ween cu en BB-PLC p o ocols and s anda ds [47]. .............. 29
Table IV. Connec ed de ices a each house (T1, T2, T3, T4) in he con olled LV g id a
UASV. ............................................................................................................................................. 51
Table V. Summa y o he jou nal pape s associa ed o he con ibu ions o his Doc o al
Thesis............................................................................................................................................... 78
Table VI. Summa y o he in e na ional con e ence pape s associa ed o he con ibu ions
o his Doc o al Thesis. ................................................................................................................ 79
Table VII. Summa y o he na ional con e ence pape s associa ed o he con ibu ions o
his Doc o al Thesis. ..................................................................................................................... 79
x
Ac onyms
AC Al e na e Cu en
AMI Ad anced Me e ing In as uc u e
AMN A i icial Mains Ne wo k
ARIB Associa ion o Radio Indus ies and Businesses
AWGN Addi i e Whi e Gaussian Noise
BB-PLC B oadband Powe Line Communica ions
BER Bi E o Ra e
BPF Band Pass Fil e
CDF Cumula i e Densi y Func ion
CENELEC Eu opean Commi ee o Elec o echnical S anda diza ion
CFR Channel F equency Response
CL Compa ibili y Le el
DBPSK_C Di e en ial Bina y Phase Shi Keying wi h Fo wa d E o Co ec ion
DC Di ec Cu en
DQPSK_C Di e en ial Quad a u e Phase Shi Keying wi h Fo wa d E o
Co ec ion
DER Dis ibu ed Ene gy Resou ce
DSL Digi al Subsc ibe Lines
DSO Dis ibu ion Sys em Ope a o
EDF Élec ici e de F ance
EMC Elec omagne ic Compa ibili y
EUT Equipmen Unde Tes
EV Elec ic Vehicle
EVCP Elec ic Vehicle Cha ging P ocess
EVCS Elec ic Vehicle Cha ging S a ion
FCC Fede al Communica ions Commission
FEC Fo wa d E o Co ec ion
FER F ame E o Ra e
x i
FFT Fas Fou ie T ans o m
HDR High Da a Ra e
JCR Jou nal Ci a ion Repo s
LDR Low Da a Ra e
ICES Indus y Canada Equipmen S anda d o Digi al Equipmen
LISN Line Impedance S abiliza ion Ne wo k
LQI Link Quali y Indica o
LV Low Vol age
MTL Mul i-Conduc o T ansmission Line
MV Medium Vol age
NB-PLC Na owband Powe Line Communica ions
NIEs Non-In en ional Emissions
OFDM O hogonal F equency Di ision Mul iplexing
PFBL Pe cen age o he equency bins exceeding he limi s
PLC Powe Line Communica ions
POC Poin o Connec ion
PQ Powe Quali y
PWM Pulse Wid h Modula ion
QP Quasi-Peak
R_DBPSK Robus Bina y Quad a u e Phase Shi Keying
R_DQPSK Robus Di e en ial Quad a u e Phase Shi Keying
RCFMFD Random Ca ie F equency Modula ion Fixed Du y
RF Radio equency
RMS Roo Mean Squa e
RSSI Recei ed Signal S eng h Indica o
RTT Round T ip Time
SG Sma G id
SM Sma Me e
SNR Signal o Noise Ra io
x ii
SoC S a e o Cha ge
STFT Sho -Te m Fou ie T ans o m
SS Seconda y Subs a ion
TL T ansmission Line
TSHV To al Sup aha monic Vol age
2TL Two-Conduc o T ansmission Line
UASV Uni e si y o Applied Sciences Vienna
UNB-PLC Ul a Na owband Powe Line Communica ions
USRP Uni e sal So wa e Radio Pe iphe al
V2G Vehicle o G id
18
Chap e 1
Thesis syn hesis
19
1. O e iew
1.1. In oduc ion
The Sma G id (SG) concep is b inging a g ea e olu ion o he elec ici y ne wo k,
whe e digi al echnologies a e used in o de o be e ma ch he supply and demand o
elec ici y in eal ime while minimizing cos s and main aining he s abili y and eliabili y
o he g id. The changes a e mo e a - eaching in he dis ibu ion and cus ome domains,
due o he ac ha SGs allow consume s o be mo e in ol ed in he ope a ion o he
powe sys em [1]. The de elopmen o he SGs means he end o he classic
p oduc ion-dis ibu ion-consump ion scheme o he elec ical g id, e ol ing in o a mo e
dis ibu ed, lexible and e icien model. This new model pa es he way o he in eg a ion
o Dis ibu ed Ene gy Resou ces (DERs), he managemen o Elec ic Vehicle (EV)
cha ging, and he implemen a ion o ene gy demand esponse echnologies [2], [3], [4], [5].
Mo eo e , i also con ibu es o he educ ion o he losses and illegal usage o he
ansmission and dis ibu ion lines, as well as an easy ope a ion o he g id and au oma ed
moni o ing [3].
In his inexo able e olu ion o he elec ical g id, high-pe o mance communica ion
echnologies a e essen ial o ul ill he s ic equi emen s o new SG se ices [6]. Fo his
eason, some Dis ibu ion Sys em Ope a o s (DSOs) ha e u ned hei gaze o Powe Line
Communica ions (PLC) as an op ion o implemen he wo-way communica ion be ween
he u ili y p o ide and cus ome s in he Low Vol age (LV) dis ibu ion sec ion o he g id.
Fi s ly, as he elec ical g id is al eady deployed, co e age p oblems a e p e en ed e en in
a eas no co e ed by o he communica ion echnologies (nei he wi ed no wi eless) [7];
and, secondly, addi ional in as uc u e cos s a e a oided. Mo eo e , PLC echnologies
p o ide he DSOs he oppo uni y o ha e con ol o hei own da a ansmission wi hou
elying on hi d pa ies [8].
In ecen yea s, Na owband PLC (NB-PLC) echnologies, ope a ing in he 3-500 kHz
equency band, ha e been widely deployed o e he LV g id o applica ions such as
eleme y, moni o ing o he de ices connec ed o he powe g id, and signal quali y
analysis [9], [10], [11], [12]. Howe e , since he elec ical g id was no concei ed o
communica ions, he cha ac e is ics o he ansmission medium ha e been p o en o
conside ably deg ade he pe o mance o NB-PLC sys ems: noise due o Non-In en ional
Emissions (NIEs), impedance a ia ions, high a enua ion/ ansmission losses and he
ime- and equency-dependen channel esponse [7], [13], [14].
Fi s ly, wi h ega d o he conduc ed emissions, he conside able inc ease in he numbe
o elec onic de ices connec ed ia powe con e e s [15] has led o an inc ease in he
ampli ude o he NIEs in he equency ange o kHz [16], [17], [18], also known as
sup aha monics [19]. These high-ampli ude emissions, which a e inhe en o he ope a ion
o powe elec onics de ices [19], in addi ion o deg ading he quali y o NB-PLC
echnologies [20], [21], can also cause Powe Quali y (PQ) issues, such as equipmen
mal unc ion, alsi ica ion o ene gy/sma me e s, o o e ol age [22], [23], [24], [25], [26],
[27]. The ele ance o he e ec o he NIEs on he communica ions depends di ec ly on
he ampli ude, spec al o m, and ime- a ian beha io o he emission. Wi h he aim o
analyzing he in luence o NIEs on PLC echnologies unde labo a o y condi ions, he
20
ETSI TS 103 909 [28] de ined in 2012 a se o e e ence noises ep esen a i e o he LV
g id. Howe e , due o he la ge inc ease o DERs in ecen yea s, he ime and spec al
pa e ns o he emissions p esen in he g id a e expec ed o ha e a ied conside ably.
Mo eo e , he e ec o his kind o NIEs on PLC is s ill unknown.
Secondly, he g id access impedance is ano he majo aspec a ec ing NB-PLC, since i
signi ican ly condi ions he signal p opaga ion [29], [30]. I also plays a pi o al ole when
designing communica ion de ices connec ed o he g id [31], as he ans e ed powe o
he ansmi ed signals depends on he impedance ma ching be ween he ansmission
de ice and he e ec i e g id access impedance [32], [33]. Besides, e y low impedance
alues may gene a e nea sho ci cui si ua ions, which lead o o e cu en o elec onic
de ices di ec ly connec ed o he g id [31]. Finally, sho - e m impedance a ia ions, gi en
wi hin he undamen al pe iod o he mains (20 ms) and al eady epo ed in [13], [34],
could also ha e a nega i e e ec on PLC. As he impedance o he elec ical g id is a ec ed
by se e al ac o s, such as ha monics, in e ha monics, a ie y o elec onic de ices, and
he g id opology, among o he s, he need o ca ying ou ex ensi e measu emen
campaigns all a ound he wo ld, conside ing di e en g id con igu a ions, emains la en
[35], [36]. Mo eo e , due o he la ge numbe o exis ing opologies and loads, signi ican
a ia ions in he impedance can be obse ed om one coun y o ano he [35], u ning
his analysis e en mo e complica ed.
Finally, he ime [37] and equency a ia ions o he g id access impedance esul in o
conside able changes in he Channel F equency Response (CFR). The modulus o he CFR
is ela ed o he losses o he ansmi ed signal while p opaga ing h ough he g id, and i
can esul in PLC signals being no ably a enua ed [29], [30], [38]. This high a enua ion
can signi ican ly educe he maximum ange o he communica ions [39] and is, gene ally,
due o he dis ance and sudden changes in loads [14]. Fu he mo e, hese impedance
a ia ions do no only a ec he PLC signal bu also he NIEs p esen in he g id [40], [41].
In conclusion, in o de o ake ad an age o he po en ial o he LV dis ibu ion g id as a
communica ion in as uc u e o PLC communica ions, i is necessa y o cha ac e ize he
elec ical ne wo k in e ms o g id access impedance, channel esponse, and NIEs. In his
way, he indus y would be able o adap da a coding, modula ion, and mul iplexing
algo i hms o he cu en cha ac e is ics o he LV g id.
21
1.2. Mo i a ion
The cha ac e iza ion o he LV dis ibu ion g id as a communica ion channel has been
add essed by he scien i ic communi y o many yea s. Howe e , conside ing ha he
elen less e olu ion o he powe ne wo k is leading o conside able changes in he
cha ac e is ics o he g id, new e o s in o de o s udy he cu en and u u e
cha ac e is ics o he p opaga ion channel a e necessa y. Mo eo e , aking in o accoun
ha he majo i y o he con ibu ions published so a only ocus on he sup aha monic
ange (9-150 kHz), u he s udies co e ing he comple e NB-PLC equency band (up o
500 kHz) a e s ill necessa y.
The e a e se e al easons o ca y ou his cha ac e iza ion by means o empi ical ials.
Fi s , he e is s ill no heo e ical model o he LV dis ibu ion g id in he NB-PLC
equency ange (9-500 kHz). Second, al hough labo a o y ials allow o ep oducibili y
in he measu emen s, hey a e pe o med unde idealized condi ions and, hus, hey do no
ep esen eal g id condi ions. Thi d, since he manu ac u e s do no publicly sha e he
de ails o he implemen ed con e e s, he heo e ical analysis o he emissions is no
gene ally possible.
In his con ex , his Doc o al Thesis aims a cha ac e izing he elec ical g id as a
p opaga ion medium in he 9-500 kHz equency band in e ms o g id access impedance,
channel esponse, and NIEs, as well as empi ically e alua ing he pe o mance o NB-PLC
unde di e en channel condi ions.
22
2. Objec i es
This Doc o al Thesis has wo main objec i es (MOs):
- MO1. Empi ical cha ac e iza ion o he elec ical g id as a communica ion
channel in e ms o he g id access impedance, channel esponse o a enua ion,
and NIEs in he 9-500 kHz equency ange.
- MO2. E alua ion o NB-PLC echnologies unde di e en channel condi ions.
In o de o achie e hese wo MOs, a numbe o speci ic objec i es (SOs) ha e been
de ined.
- SO1. Cha ac e iza ion o he conduc ed emissions in he elec ical g id.
The conduc ed emissions should be cha ac e ized bo h in he equency and ime
domains in he 9-500 kHz equency ange. Conside ing he high pene a ion o
EVs in coming yea s, his Doc o al Thesis should speci ically add ess he
conduc ed emissions gene a ed by EV cha ging p ocesses (EVCPs). This
cha ac e iza ion should include NIEs gene a ed by a wide ange o EV models,
cha ging cu en s, and S a es o Cha ges (SoCs). Mo eo e , wi h he aim o
de e mining whe he isola ed measu emen s a e a good ep esen a ion o he eal
LV g id, di e en measu emen condi ions (isola ed/on-line) should be aken
in o conside a ion. Finally, in o de o ully de e mine he e ec s o he EVCPs
on he ampli ude o he emissions in he LV g id, he p opaga ion and in e ac ion
o hese NIEs should also be analyzed.
- SO2. Cha ac e iza ion o he g id access impedance. The sub-cycle, mean,
and long- e m g id access impedance should be cha ac e ized in he 9-500 kHz
equency ange. As in PO1, conside ing EVs as an example o mode n de ices
o be widely deployed in he ollowing yea s, he g id impedance in p esence o
EVCPs should be comple ely cha ac e ized. This cha ac e iza ion should include
di e en comme cial EV models, cha ging cu en s, and SoCs.
- SO3. E alua ion o he po en ial in luence o conduc ed emissions on
NB-PLC. This Doc o al Thesis should analyze he po en ial in luence o he
conduc ed emissions on NB-PLC. Fo his pu pose, NB-PLC should be
e alua ed in p esence o he p e iously cha ac e ized NIEs gene a ed by EVCPs.
- SO4. E alua ion o he po en ial in luence o impedance a ia ions on
NB-PLC. This Doc o al Thesis should e alua e he po en ial in luence o
impedance a ia ions on NB-PLC, aking in o conside a ion bo h
equency-dependen impedances s a ic o e ime and sub-cycle impedance
a ia ions.
In his ex , he e m cha ac e iza ion e e s o he calculus o ep esen a i e me ics and
pa ame e s o he signals and he iden i ica ion o hei pa e ns bo h in he equency and
ime domains.
A summa y o he main and speci ic objec i es o his Doc o al Thesis a e p esen ed in
Fig. 1.
23
Fig. 1. Summa y o he main objec i es (MOs) and speci ic objec i es (SOs) o his Doc o al Thesis.
Finally, in o de o ela e he objec i es o his Doc o al Thesis wi h he published pape s
ha make up he compendium, in Table V, a summa y o he jou nal pape s associa ed o
he con ibu ion o his Doc o al Thesis a e p esen ed, oge he wi h he objec i e o which
hey a e ela ed.
30
Al hough di e en BB-PLC echnologies ha e been de eloped o in-home and MV
channels, he e is s ill no BB-PLC sys em speci ically designed o be deployed o e he LV
dis ibu ion g id. Conside ing he enhanced pe o mance wi h espec o NB-PLC in e ms
o bandwid h, la ency, and secu i y equi emen s, BB-PLC is an al e na i e ha is being
conside ed by some DSOs o da a ansmission h ough he LV dis ibu ion g id, in o de
o ul ill he demanding equi emen s o new SG applica ions [76].
As he beha io o he ou doo channel is expec ed o be conside ably di e en om he
cha ac e is ics o he indoo en i onmen , DSOs a e aking in o conside a ion wo
di e en op ions. Fi s , he adap a ion o he exis ing indoo echnologies o he
cha ac e is ics o he LV g id; and, second, he de elopmen o a new s anda d speci ically
designed o he ou doo ansmission medium. In any case, he de elopmen o BB-PLC
echnologies would de ini ely be a solu ion o co e ing he needs o u u e SG
applica ions, including he in eg a ion o DERs o EV cha ging managemen , among
o he s [47].
31
3.2. Cha ac e iza ion o he elec ical g id as a
ansmission medium o NB-PLC echnologies
In his sec ion, he mos ele an aspec s o be conside ed o he empi ical
cha ac e iza ion o he elec ical g id as a ansmission medium a e add essed. Sec ion 3.2.1
deals wi h he conduc ed emissions, while he g id access impedance and he
a enua ion/channel esponse a e s udied in sec ion 3.2.2.
Th oughou he ex , he e m noise co esponds o an elec omagne ic phenomenon no
con eying in o ma ion and which is combined wi h a wan ed signal [77]. Since he
emissions e e ed o in his Doc o al Thesis a e always non-in en ional, emissions and
NIEs a e used as synonyms and a e de ined as unwan ed signals gene a ed by he powe
elec onics included in he ci cui y o connec ed elec onic de ices. The e ms in e e ence
and dis u bance, in u n, imply a deg ada ion o a sys em. In he case o an in e e ence,
he pe o mance o a communica ions sys em is jeopa dized, while a dis u bance is ela ed
o he deg ada ion o PQ, o he li e ime o a de ice, among o he s. Finally, he e m
dis o ion e e s o an undesi ed change in he wa e o m o a signal ha migh lead o he
appea ance o new equency componen s.
3.2.1. NIEs
Main sou ces o NIEs
Some ecen ly published a icles [78], [79], [80], [81], [82] show ha pho o ol aic in e e s
(PV), ba e y cha ge s, ene gy-e icien ligh ing, hyd opowe sys ems, wind u bines, o EV
cha ge s, among o he s, a e he main sou ces o he high-ampli ude conduc ed emissions
o he equency band be ween 3 kHz and 500 kHz. In se e al ield and labo a o y ials,
a equency and ime cha ac e iza ion o hese emissions has been ca ied ou , concluding
ha , as hese emissions occu a equencies assigned o NB-PLC, he quali y o
communica ions can be subs an ially a ec ed and deg aded [20], [50], [83], [84]. In some
ins ances, as o he EV cha ge s and PV panels, he sou ces o dis u bance a e loca ed
nea he sma me e , which migh imply an addi ional challenge o he co ec
pe o mance o NB-PLC [24]. O he non-in en ional emi ing de ices, such as mo o s [79],
ligh ing de ices [85], [86], [87], o elec onic ampli ie s [19], in oduce high-ampli ude
emissions wi h a wide ange o spec al pa e ns. A signi ican ime-dependen beha io
has also been obse ed o he p e iously men ioned sou ces [79]. I should also be no ed
ha high-ampli ude impulsi e dis u bances a e gene a ed by his equipmen when
commu ing be ween di e en s a es o wo king egimes [79], which may ha e an addi ional
nega i e e ec on communica ions.
No ma i e amewo k
Emission limi s
The maximum ampli ude o he emissions gene a ed by ce ain equipmen connec ed o
he elec ical g id has al eady been speci ied by he In e na ional Special Commi ee on
Radio In e e ence CISPR. CISPR15 (EN 55015) [85] de ines he limi s o he ligh ing
equipmen , whe eas CISPR11 (EN 55011) [88] add esses he maximum ampli udes o
32
induc ion cooking de ices. By con as , no speci ic limi s ha e been speci ied o many
o he sou ces o emissions, such as PVs, EVs, o hyd opowe sys ems. Fo hese de ices,
he ou -o -band limi s de ined o communica ions equipmen in EN 50065-1 [89] migh
be conside ed as a conse a i e c i e ion [90]. This echnical speci ica ion de ines limi s
adap ed o he equency band o ansmission (2-9 kHz, 9-150 kHz, o 150 kHz-30 MHz),
and he ype o de ec o ha should be used in he measu emen s, Roo Mean Squa e
(RMS), Quasi Peak (QP) o A e age [91], [92], [93]. These emission limi s a e de ined o
labo a o y condi ions and mus be e alua ed by means o a Line Impedance S abiliza ion
Ne wo k (LISN). A LISN is a s anda d load impedance allowing he epea abili y and
compa abili y o EMI measu emen s, which also p e en s he ials om being a ec ed
by ex e nal conduc ed emissions [94].
In he LV dis ibu ion g id, he Compa ibili y Le els (CLs) es ablish he maximum
ampli udes o he emissions ha canno be exceeded a a speci ic elec ical poin , as he
combina ion o he emissions gene a ed by all he equipmen connec ed o he ne wo k.
The g id ope a o should ensu e ha a leas in 95 % o he loca ions hese limi s a e no
exceeded. The Annex B o he IEC 61000-4-7 [95] speci ies he limi s o he equency
band om 2 kHz o 9 kHz in RMS alues, whe eas he IEC 61000-2-2 [96] de ines he
limi s o he 9-150 kHz equency ange in QP alues.
Measu emen me hods
In he bibliog aphy, he e m measu emen me hod e e s o he pos -p ocessing applied
o he eco ded signals o e alua e NIEs in he equency domain. Up o now, no
no ma i e measu emen me hod o he assessmen o he dis u bances in he LV
dis ibu ion g id has been de ined o equencies abo e 9 kHz. As p e iously men ioned,
he CLs in he 9-150 kHz equency ange ha a e included in IEC 61000-2-2 [96] a e
de ined o QP alues. Fo his eason, CISPR16 1-1 [91], a me hod based on a QP
de ec o , is he me hod ha migh be used o he e alua ion o he dis u bances in g id
measu emen s [97]. Ne e heless, his me hod p esen s se e al d awbacks: i s , i is no
in ended o g id measu emen s, bu a he o labo a o y condi ions; and, second, i has
high complexi y, compu a ional bu den, and memo y equi emen s.
Mo eo e , he e is no s anda dized p ocedu e o he quan i a i e cha ac e iza ion o he
emissions in he equency domain. The li e a u e only conside s he To al Sup aha monic
Vol age (TSHV) [98], [99], [100], a pa ame e ha gi es an insigh in o he o al ampli ude
o he emissions in he equency band unde analysis. Howe e , his pa ame e is highly
dependen on he measu emen me hod used o he pos -p ocessing o he co esponding
emission [101] and does no ake in o accoun he equency dis ibu ion o he ampli ude
o he emissions, which plays an essen ial ole in he p ope design o PLC echnologies.
Mo eo e , he concep s o na owband and b oadband emission ha e no ye been de ined
and, he e o e, he e is no no malized p ocedu e o hei p ope cha ac e iza ion.
Finally, i should be men ioned ha hese measu emen me hods a e only in ended o
cha ac e ize he emissions in he equency domain, wi hou conside ing hei
ime-dependen beha io . Since hese ime a ia ions can nega i ely a ec PLC, he
de elopmen o me hods ha add ess he ime cha ac e iza ion in conjunc ion wi h he
equency cha ac e iza ion is necessa y, o which he de ini ion o s anda dized
me hodologies in he ime domain is o upmos impo ance. A join ime- equency
33
domain cha ac e iza ion o he emissions in he equency ange 9-150 kHz can be ound
in [102].
Measu emen se ups
Exis ing s anda ds speci y measu emen se ups o he cha ac e iza ion o he emissions
om equipmen unde es . CISPR 16-1-2 [103] de ines a s anda dized measu emen se up
o he e alua ion o he emissions based on a LISN om 9 kHz o 109 MHz. Simila ly,
IEC 61000-4-7 [95] speci ies he use o an A i icial Mains Ne wo k (AMN) below 9 kHz.
In his way, a con olled and isola ed scena io is se and he measu emen s a e no a ec ed
by ex e nal NIEs [104].
Howe e , he measu emen campaigns ca ied ou so a lead o conclude ha he
e e ence impedances de ined in [95], [103] a e no a good ep esen a ion o he ac ual g id
access impedance alues o he LV dis ibu ion g id [41]. Fo his eason, he
cha ac e iza ion o he emissions should no only be based on measu emen s using AMN
o LISN, whe e isola ed e ec s can be e alua ed, bu also on ield ials, so ha
ep esen a i e dis u bance alues a e ob ained.
In his con ex , some au ho s [105] ha e op ed o conduc ing in es iga ions ela ed wi h
NIEs a econs uc ed acili ies. This ype o scena ios a oid he ideal condi ions o
labo a o y measu emen s as well as he uncon olled g id ac o s p esen in on- ield ials,
such as he a ie y o loads connec ed o he g id, he g id opology, he g id impedance,
o he elec ical cables, among o he s [78].
P opaga ion/in e ac ion o he emissions
Acco ding o he li e a u e, he emissions gene a ed by he de ices connec ed o he g id
p opaga e h ough he LV ne wo k, and may e en be ans e ed o he MV g id o e
dis ances o se e al kilome e s [106], [107], [108]. As a esul , hey may a ec he ope a ion
o ene gy me e s and PLC equipmen [109], [110]. As indi idual de ices ha e a g ea e
impac on highe equencies han on ha monics [40], he p edic ion o he NIEs in his
equency ange should conside he whole ins alla ion and no only indi idual de ices
[111].
In acco dance o [40], [41], [112], [113], [114], [115], he emissions gene a ed by a ce ain
de ice can be classi ied as p ima y and seconda y emissions. The p ima y emission
co esponds o he emission o igina ed inside he de ice, whe eas he seconda y emission
is gene a ed by o he elec onic equipmen o he g id i sel , and p opaga ed in o he
de ice. This p opaga ion highly depends on he impedance o neighbo ing de ices in
ela ion o he impedance o he elec ical g id [112], [116], [117]. Resonances a e a key
aspec in he p opaga ion o he dis u bances, as hey imply inc eases in he emission a
he swi ching equency [41], [112], [118]. In [40], i is s a ed ha a esonance esul s in an
inc ease in he seconda y emission, whe eas he p ima y emission is a enua ed.
Rega ding he in e ac ion, some wo ks ha e poin ed ou he exis ence o equency bea ing
and in e modula ion e ec s. Fo ins ance, in [17], se e al simula ions ca ied ou in
MATLAB led o conclude ha he cha ging o EVs o he same ype, due o sligh ly
di e en swi ching equencies ( 1, 1’), implies an emission a | 1- 1’|. In gene al, he
bea ing equency is in he o de o some Hz and does no a ec PLC. In e modula ion
34
dis o ion in he equency band up o 100 kHz, in u n, occu s due o he in e ac ion o
conside ably di e en swi ching equencies ( 1, 2) and is in he o de o ens o kHz [17].
Ano he simila analysis conce ning in e modula ion p oduc s can be ound in [119]. In
his a icle, he in e ac ion be ween a PV panel and an EV is epo ed, e idencing he
exis ence o in e modula ion p oduc s o up o 4 h o de .
The in e ac ion be ween end-use and PLC equipmen has also been s udied in he
li e a u e. Fi e ypes o in e ac ions be ween hese de ices a e desc ibed in [120],
concluding ha he pe o mance o communica ions is deg aded and he li e ime o he
end-use equipmen is conside ably educed.
Elec ic ehicles
The ising numbe o EVs in ol es conside able challenges o he elec ici y g id. Fo
example, EV Cha ging S a ions (EVCSs) gene a e highe ampli ude emissions han o he
elec onic de ices connec ed o he g id [121]. Fo his eason, he e is g ea in e es in he
scien i ic communi y in analyzing and cha ac e izing he dis u bances gene a ed by EVs
du ing he cha ging p ocess.
Since he manu ac u e s o he EVCSs do no publicly sha e he de ails o he implemen ed
con e e s, he heo e ical analysis o he dis u bances is no gene ally possible. Fo his
eason, as s a ed in [122], he cha ac e iza ion o hese emissions can only be ca ied ou
empi ically, by means o ex ensi e measu emen campaigns.
In ecen yea s, some labo a o y and ield ials ha e been de eloped o he analysis o
he dis u bances gene a ed by se e al EVCPs. Fo ins ance, in [105], he emissions
in oduced by a Bi-Di ec ional Vehicle o G id (V2G) EVCS a e s udied in a con olled
g id scena io up o 150 kHz. This wo k concludes ha na owband emissions a e p esen
a he swi ching equency and mul iples o i , caused by he Pulse Wid h Modula ion
(PWM), whe eas b oadband emissions occu a highe equencies due o he DC-DC
con e e when he Posi i e Tempe a u e Coe icien hea e is ac i a ed. Field ials
p esen ed in [123] show ha he high ampli ude emissions gene a ed by an elec ic bus do
no dec ease wi h equency in he band 2-150 kHz and ha hey could be as high as he
CL de ined in [96]. In [17], he p opaga ion and in e ac ion o he NIEs o ou EVs a e
analyzed up o 100 kHz in an isola ed g id scena io connec ed o he public elec ici y g id.
As he g id allows swi ching o mic og id mode, measu emen s in bo h con igu a ions
ha e been conside ed, ob aining di e en esul s in some cases. Re e ence [124] analyzes
he long- e m sup aha monic emissions o h ee EVs in he ime and equency domains
up o 100 kHz a h ee pa king ga ages. The p esen ed esul s lead o conclude ha he
powe g id dis u bance le els inc ease when he numbe o connec ed EVs ise. In [125],
he dis u bances gene a ed by di e en ypes o cha ge s a i e si es in China and
Ge many a e cha ac e ized in he equency domain. The a icle shows he dominan
emission equencies as well as hei ol age ampli ude up o 50 kHz. A simila analysis is
p esen ed in [126], [127], [128], whe e he swi ching equencies o wel e, en, and eigh
EVs, espec i ely, a e iden i ied. A huge a iabili y in he swi ching equency and i s
ampli ude can be obse ed in all h ee cases.
Howe e , in on-si e measu emen s, emissions a e a ec ed by se e al g id ac o s [129],
including he ime and equency-dependen g id access impedance. Thus, in some wo ks,
such as [104], he EVCS is di ec ly connec ed o a LISN in o de o isola e he se up om
35
he LV g id, a oiding ex e nal noise and ensu ing ha only he emissions gene a ed by he
EVCPs a e measu ed.
Measu emen campaigns o he cha ac e iza ion o he NIEs
Se e al au ho s ha e pe o med on- ield ials in di e en LV dis ibu ion g ids all a ound
he wo ld conside ing ce ain sou ces o dis u bances and g id opologies. As an example,
in [130], he NIEs a indus ial, esiden ial, and u al a eas in he Tu kish LV g id a e
cha ac e ized up o 100 kHz, ob aining emission ampli udes a ound 90 dBµV, 100 dBµV,
and 75 dBµV, espec i ely. A s a is ical cha ac e iza ion o he emissions in he LV g id
based on p obabilis ic unc ions is p esen ed in [25], co e ing he 2-150 kHz equency
band. Ano he s a is ical analysis can be ound in [131], whe e a long- e m cha ac e iza ion
o he emissions in he LV g id in Qa a is p esen ed conside ing mo e han 1.8 billion
samples a h ee di e en loca ions o e 10 days. The analysis is based on he s a iona i y,
au oco ela ion, and independence o he NIEs in he equency ange om 10 kHz o
490 kHz. A measu emen campaign ca ied ou in Spain [16], in u n, concludes ha he
p opaga ion channel in u al scena ios is ema kably noisy in low equencies, as he
highes NIEs occu a equencies lowe han 150 kHz, mainly lowe han 40 kHz.
Howe e , high-ampli ude emissions can be ound in he whole equency ange up o
500 kHz in u ban en i onmen s. In [12], a simila cha ac e iza ion o he NIEs in he
NB-PLC equency band is ga he ed o he LV g id in F ance. This a icle poin s ou ha
he noise is p edominan ly cyclos a iona y and concludes ha he backg ound noise is
s able o e pe iods o hou s.
Mo e in o ma ion abou he measu emen sys ems used o de eloping hese measu emen
campaigns can be ound in he jou nal pape (JP1), which can be ound in Appendix A.1.
3.2.2. G id access impedance
No ma i e amewo k
As he beha io o he g id access impedance in he equency band assigned o PLC
echnologies is s ill unknown [132], he egula o y amewo k in his ield is e y limi ed.
IEC 61000-4-7 de ines a e e ence g id impedance in he equency ange 2-9 kHz, bu
some au ho s ha e ex ended i s de ini ion up o 150 kHz [105], [133], [134]. Howe e , a
measu emen campaign ca ied ou in he LV g id in Aus ia, Swi ze land, Czech Republic,
and Ge many demons a es ha he e e ence g id impedance in IEC 61000-4-7
o e es ima es e en he highes alues o he impedances measu ed in he ac ual
dis ibu ion g id [133].
CISPR 16-1-2 desc ibes h ee di e en e e ence impedances o de ining speci ic LISNs
depending on he equency ange (9-150 kHz, 150 kHz-30 MHz, o
150 kHz-108 MHz) [103]. Acco ding o [134], he CISPR 16-1-2 e e ence impedance
de ined in he equency ange 9-150 kHz also exceeds he alues o he impedances
measu ed in indoo en i onmen s in Ge many, Uni ed Kingdom, and Spain.
36
Measu emen campaigns o he cha ac e iza ion o he
impedance
Only a ew measu emen campaigns o cha ac e ize he g id impedance ha e been
pe o med all a ound he wo ld. In [130], he g id access impedance in he LV g id in
Tu key is cha ac e ized up o 100 kHz, conside ing esiden ial, u al, and indus ial
scena ios. The pape concludes ha , ega dless o he a ea, impedance alues below 10 Ω
a e measu ed in he whole equency band. Ano he measu emen campaign co e ing he
same equency ange in Aus ia, Swi ze land, Czech Republic, and Ge many is p esen ed
in [133], showing ha he impedance inc eases wi h equency and ha emains below he
IEC 61000-4-7 e e ence impedance in he whole equency band. In [132], phase o
neu al impedance measu emen s a e ca ied ou a an EV cha ging plaza a he uni e si y
and ou esiden ial household ins alla ions in The Ne he lands in he 9-150 kHz equency
ange. This a icle concludes, i s , ha , due o he capaci o included in hei ci cui y,
bo h EVs and PVs imply a low-impedance pa h o communica ions signals, and, second,
ha household appliances ha e a conside able e ec on he g id impedance. Mo eo e ,
he impo ance o he opology on he impedance is also highligh ed. Al hough mos o
he published s udies ocus, exclusi ely, on he 9-150 kHz equency band, some
publica ions p o ide in o ma ion up o 500 kHz. Fo example, in [36], he access
impedance om 35 kHz o 500 kHz a h ee Seconda y Subs a ions (SSs) and a a se o
access poin s o he LV dis ibu ion g id in he Basque Coun y (Spain) has been
cha ac e ized. Fo his pu pose, wo u ban and a u al scena io ha e been conside ed. The
s udy p o ides an insigh in o he g ea di e ences be ween he impedances measu ed a
each elec ical poin and concludes ha he u ban dis ibu ion can be modeled as a
pa icula scena io o sho cable sec ions and nume ous homes. In [135], a simila
cha ac e iza ion in he equency band 30-500 kHz is pe o med, measu ing he g id access
impedance in an u ban/subu ban a ea in China wi h low- ise apa men buildings.
Howe e , hese esul s canno be ex apola ed o any dis ibu ion ne wo k, since he
pe o mance o he elec ical g id a he equencies associa ed o PLC changes d as ically
om one coun y o ano he , mainly due o he di e si y o loads and g id opologies [136],
[137], [138].
Mo e in o ma ion abou he measu emen sys ems used o de eloping hese measu emen
campaigns can be ound in he jou nal pape (JP1), which can be ound in Appendix A.1.
3.2.3. Channel esponse
As p e iously men ioned, he ime and equency a ia ions o he g id impedance imply
changes in he channel esponse [37]. Since he modulus o he CFR co esponds o he
a enua ion o he ansmi ed PLC signal, communica ions can su e om high
a enua ion [29], [30], [38]. This high a enua ion is one o he main ac o s a ec ing he
pe o mance o NB-PLC, since i can educe he ange o he communica ions [39].
The cha ac e iza ion o he PLC channel is ca ied ou acco ding o wo app oaches: he
op-down app oach, whe e a model is ob ained om a la ge numbe o ield ials, and he
bo om-up app oach, based on modeling he channel by applying T ansmission Line (TL)
heo y [139].
37
Rega ding he op-down app oach, he indoo PLC channel has been ex ensi ely analyzed
in e ms o a e age channel a enua ion o channel gain, delay sp ead, cohe ence
bandwid h, and channel capaci y in [136], [137], [140], [141], [142], [143], [144], [145], [146],
[147], [148], [149], [150]. In [151], he cha ac e iza ion o he indoo PLC channel based
on mul ipa h phenomenon can be ound. Conce ning he cha ac e iza ion o he ou doo
scena io, o ins ance, in [152], he g id access impedance, he channel esponse (in e ms
o delay sp ead, channel bandwid h, and a enua ion), and he NIEs in di e en u ban,
semiu ban, and u al scena ios in he LV dis ibu ion g id a e analyzed, as well as he
achie able da a a es. The esul s p esen ed in his pape only co e he CENELEC A
NB-PLC equency band (3-95 kHz). A simila analysis in he equency ange assigned o
BB-PLC (1.7-100 MHz) is ga he ed in [153], whe e he B azilian ou doo scena io is
cha ac e ized conside ing he a e age channel a enua ion, oo mean squa ed delay
sp ead, cohe ence bandwid h, cohe ence ime, and he achie able da a a e. In e e ence
[154], a cha ac e iza ion o he ou doo channel conside ing he mul ipa h signal
p opaga ion heo y is p esen ed. This model is e i ied by a se o channel esponses
ob ained by means o ield ials.
Since in he bo om-up app oach he ne wo k elemen s a e ma hema ically modeled, a
ho ough knowledge o he elec ical g id ( opology, cable cha ac e is ics, load impedances,
e c.) is needed. This app oach conside s he wo-conduc o TL (2TL) heo y [155], [156],
[157], [158], [159], [160], [161], used o powe ne wo ks connec ed wi h wo-conduc o
ansmission lines, and he mul i-conduc o TL (MTL) heo y [162], [163], [164], [165],
[166], [167], [168], [169], which is a gene aliza ion o he 2TL app oach. The TL heo y
has also been applied by some au ho s o he cha ac e iza ion o he ou doo PLC channel
[159], [164], [165], [170], [171]. In e e ence [172], bo h he LV ae ial and unde g ound
cable dis ibu ion lines a e model by means o he MTL heo y.
A combina ion o he op-down and bo om-up app oaches can be ound in [173], [174],
whe e a s a is ical analysis o he NB-PLC channel in he LV g id in Pakis an is p esen ed
up o 150 kHz and 500 kHz, espec i ely.
Measu emen campaigns o he cha ac e iza ion o he channel
esponse
The huge a ie y o causes ha in ol e a enua ions in PLC signals and he dependency o
he beha io o he elec ical g id on he geog aphical a ea [136], [137], [138], mainly
because o he he e ogenei y o g id opologies and connec ed loads, makes he p edic ion
o he PLC channel esponse a complex s udy ha should be add essed. In [130], o
example, he a enua ion be ween he SS and he SMs in he LV g id in Tu key up o
1 MHz is analyzed in esiden ial, indus ial, and u al a eas, e ealing a enua ions a ound
20 dB (dis ances om 50 m o 150 m), 30 dB (dis ances om 80 m o 270 m), and 40 dB
(dis ances om 150 m o 500 m) a each scena io, espec i ely. Re e ence [175] conside s
a se o ep esen a i e g id opologies, concluding, as in [176], ha an inc ease in dis ance
implies highe ansmission losses. Howe e , i is also s a ed ha he e is no a linea end
be ween he a enua ion and he dis ance be ween communica ions equipmen , due o he
g ea impac o he numbe o b anches in ee like opologies. In [177], apa om
measu ing he a enua ion in he equency band associa ed o NB-PLC, a cha ac e iza ion
o he ansmission losses measu ed om 500 kHz o 10 MHz in he LV g id in China is
p esen ed. A compa ison be ween he mean ansmission losses measu ed in u ban and
38
u al esiden ial a eas is ca ied ou , concluding ha he a enua ion wi h espec o
equency in he u ban scena ios is la e han wha is measu ed in u al a eas.
Mo e in o ma ion abou he measu emen sys ems used o de eloping hese measu emen
campaigns can be ound in he jou nal pape (JP1), which can be ound in Appendix A.1.
3.2.4. Conclusions and open issues
Since EVs will be massi ely deployed in coming yea s [178] and ha e been iden i ied as
one o he main sou ces o emissions in he g id [121], his Doc o al Thesis will ocus on
e alua ing he conduc ed emissions gene a ed by EVCPs, as well as hei impac on he
g id impedance. In any case, conside ing ha some o hese dis ibu ed ene gy esou ces
a e also based on in e e s, as i is he case o an EVCP, i is expec ed ha a leas pa o
he conclusions d awn in his hesis will emain alid in hose cases. Simila ly, i is expec ed
ha he esul s ela ed o p opaga ion and in e ac ion o emissions om EVCPs will also
apply o o he sou ces o NIEs o a simila na u e in his equency band.
Conside ing he s a e o he a p esen ed abo e, a se o esea ch gaps ha e been iden i ied
ega ding he cha ac e iza ion o he conduc ed emissions, which a e de ailed below
oge he wi h he main con ibu ions o his Doc o al Thesis
- Since he e a e nei he s anda dized me hods no me ics o he e alua ion and
quan i ica ion o he NIEs in he g id, his Doc o al Thesis will de ine a no el
p ocedu e o he e alua ion o he emissions in bo h he equency and ime
domains. This p ocedu e will no only analyze he ampli ude o he conduc ed
emissions, bu also hei spec al dis ibu ion and ime-dependen beha io .
- The majo i y o he exis ing s udies only co e he sup aha monic (9-150 kHz)
equency band. Fo his eason, his Doc o al Thesis will e alua e he conduc ed
emissions co e ing he whole NB-PLC equency band (9-500 kHz).
- Since mos s udies do no e alua e he emissions o pe iods longe han ew
minu es, his Doc o al Thesis will cha ac e ize he NIEs du ing 600 s, as la e
jus i ied in sec ion 4.2.2. The Doc o al Thesis will de ine a speci ic me hod o
he quan i ica ion o he ime a iabili y wi hin he eco ding ime.
- The li e a u e analyzes he p opaga ion and in e ac ion o he emissions based on
a p ima y/seconda y app oach. Howe e , his Doc o al Thesis will add ess he
p opaga ion and in e ac ion o he NIEs based on synch onized measu emen s
a di e en elec ical poin s in he LV dis ibu ion g id.
- Al hough he li e a u e poin s ou ha he emissions gene a ed by EVCPs unde
isola ed and on-line condi ions a e no simila , he e is a lack o s udies in which
he NIEs gene a ed by he same EVCPs unde bo h isola ed and LV g id
condi ions a e compa ed. Fo his eason, his Doc o al Thesis will analyze he
NIEs due o a wide ange o EV models and cha ging cu en s unde bo h
measu emen condi ions.
In o de o co e hese esea ch gaps, he ollowing c i e ia has been applied:
- The emission eco dings will be pe o med by means o a measu emen sys em
de eloped by he esea ch eam, which has been p o en o p o ide accu a e
alues in he equency band o in e es .
39
- Since no speci ic emission limi s ha e been de ined o EVCPs, his Doc o al
Thesis will use he PLC ou -o -band emission limi s de ined in EN 50065-1 [89]
o compa ison pu poses.
- Conce ning he measu emen me hods o he spec al cha ac e iza ion o he
emissions, he CISPR 16-1-1 s anda d [91] will be selec ed.
Conce ning he cha ac e iza ion o he g id impedance, he ollowing esea ch a enues
ha e been de e mined, as well as he con ibu ions o be made in his in es iga ion a ea:
- As p e iously men ioned o he conduc ed emissions, exis ing s udies only co e
he sup aha monic ange. This Doc o al Thesis will e alua e he g id impedance
in he whole NB-PLC equency ange (20-500 kHz).
- The li e a u e only add esses he cha ac e iza ion o he mean g id impedance in
he p esence o EVCPs. Fo his eason, his Doc o al Thesis will p esen a
comple e cha ac e iza ion o he impedance, conside ing sub-cycle (20 ms),
mean, and long- e m a ia ions (some hou s).
- This Doc o al Thesis will cha ac e ize he a ia ions o he CFR due o
impedance a ia ions ( equency-dependen and sub-cycle impedance a ia ions),
in o de o ela e bo h pa ame e s.
Wi h he aim o co e ing hese esea ch gaps, he ollowing c i e ia ha e been de e mined:
- The cha ac e iza ion o he g id impedance in he p esence o EVCPs is
pe o med in con olled LV g ids, since hey show wo main ad an ages. Fi s ,
since hey a e isola ed om he public LV g id, ex e nal ac o s ha may a ec
he impedance measu emen s a e a oided; and, second, he esemblance o eal
ne wo k condi ions is main ained, as hey a e ec ea ed based on he
cha ac e is ics o eal public LV g ids.
- The impedance and CFR measu emen s will be conduc ed by means o ad-hoc
measu emen sys ems designed by he esea ch eam, which ha e been p o en o
p o ide accu a e alues in he equency band o in e es .
46
61000-4-19:2014 [212] s anda ds. Mo eo e , i s accu acy has been also e alua ed in a
con olled and ep oducible labo a o y scena io and compa ed wi h espec o o he
me hods p o iding simila esul s [214].
Fig. 5. Measu emen sys em o he assessmen o he g id access impedance.
4.1.3. A enua ion
The a enua ion measu emen sys em, shown in Fig. 6, calcula es he ansmission losses,
in modulus and phase, be ween wo elec ical poin s, P1 and P2. Depending on he selec ed
con igu a ion, he mean o sub-cycle a enua ion is p o ided. The a enua ion is calcula ed
as he di e ence be ween he le el o he injec ed signal a P1 and he ecei ed signal a
P2.
This measu emen sys em is composed o ou ol age p obes: h ee o hem connec ed
a poin P1, and one a poin P2. One o he ac i e p obes connec ed a P1 is esponsible
o injec ing a sweep in he equency ange 5-530 kHz and o il e ing he 50 Hz signal,
while he wo passi e p obes (one a P1 and one a P2) measu e he ansmi ed and
ecei ed ol age, espec i ely.
The signal gene a o , con olled by means o a Ma lab sc ip , is esponsible o ansmi ing
bu s s o 90 ms no synch onized wi h he mains equency om 6 kHz o 507 kHz wi h
3 kHz s eps. A gap o 10 ms be ween wo consecu i e bu s s is used as a gua d pe iod.
F om each 90 ms bu s , only 20 ms a e e alua ed, dis ega ding he i s and he las 20 ms
and synch onizing he s a o hose 20 ms o be analyzed wi h a ze o c ossing o he mains
signal. This synch oniza ion, pe o med as desc ibed o he impedance measu emen
sys em in sec ion 4.1.2, is ca ied ou by means o one o he ac i e ol age p obes
connec ed a poin A. This ol age p obe is used o measu ing he 50 Hz e e ence signal,
in o de o be able o analyze he a enua ion in 20 ms in e als, synch onized o he mains
pe iod.
47
Fo he signal pos -p ocessing, a sliding Gaussian windowing wi h a 1/3000 s du a ion and
an o e lapping o 75 % be ween consecu i e windows is applied, so ha he equency
esolu ion is 3 kHz. The e o e, a ime esolu ion o sampling pe iod o he es ima ed
channel equency esponse o 83.33 µs is ob ained.
Fig. 6. Measu emen sys em o he assessmen o he channel equency esponse o a iable channel
condi ions.
48
4.2. Measu emen campaigns
4.2.1. Measu emen campaign o he cha ac e iza ion o he
emissions gene a ed by EVCPs unde isola ed condi ions
The on-si e measu emen s we e ca ied ou on a comme cial EVCS ope a ing in mode 3
acco ding o IEC 61851-1 [215]. All he measu emen s we e pe o med in monophasic
mode. Wi h he aim o implemen ing a con olled measu emen scena io, he se up was
isola ed om he LV g id by means o a LISN [216], a small ans o me (Polylux PD400
[217] wi h a a ed powe o 4000 VA), and a band-pass il e (BPF) adjus ed o he PLC
equency band. Bo h he LISN and he ans o me ope a ed in monophasic mode.
The EVCS was di ec ly connec ed o he Equipmen Unde Tes (EUT) po o he LISN.
The esul ing emissions we e measu ed om he Radio equency (RF) po wi h a digi al
oscilloscope, which was con olled by a lap op cha ging a a powe s a ion. As he
measu emen s a e ob ained om he RF po , a co ec ion ac o should be applied in he
pos -p ocessing s age o compensa e he inse ion loss o he LISN be ween he EUT and
RF po s. Typical alues o he inse ion loss o he LISN as a unc ion o equency can
be ound in [216]. In Fig. 7, a ep esen a ion o he measu emen se up used o he
eco ding o he NIEs using a LISN is shown.
Fig. 7. Measu emen scena io o he eco ding o he NIEs gene a ed by EVs du ing hei cha ging p ocess
using a LISN (isola ed condi ions).
The measu emen campaign conside s he eco ding o he emissions gene a ed by wo
EV models and wo di e en cha ging cu en s.
4.2.2. Measu emen campaign o he cha ac e iza ion,
p opaga ion, and in e ac ion o he emissions gene a ed by
EVCPs in a con olled LV g id
This measu emen campaign was ca ied ou in he “Concep G id” labo a o y o
Élec ici e de F ance (EDF), a unique es ing acili y ha goes beyond ideal condi ions o
labo a o y ials, bu a he same ime, a oids uncon olled backg ound dis o ion ha may
subs an ially a ec he esul s [105]. This es ing scena io simula es a LV dis ibu ion g id
composed o a Seconda y Subs a ion (SS) and i e houses (H1, H2, H3, H4 and H5), wi h
a h ee-phase ins alla ion, o which di e en elec onic de ices can be connec ed.
In his s udy, h ee di e en EVCSs a e analyzed. EVCS1 is ins alled a H2, EVCS2 a H3,
49
and EVCS3 a H5. The e a e h ee EV models a ailable, and each EV model can only be
cha ged a i s co esponding EVCS. A ep esen a ion o he measu emen scena io is
shown in Fig. 8, whe e he dis ance be ween he di e en houses and he loca ion o he
EVCSs a e indica ed.
Fig. 8. Measu emen scena io in he “Concep G id” labo a o y o EDF in Écuelles.
In o de o ha e a p edominan ly esis i e load a he POC, domes ic hea e s we e
connec ed o each house. All he measu emen s we e conduc ed in he same elec ical
phase (monophasic measu emen s).
The dis u bances we e measu ed a he Poin o Connec ion (POC) o each house o which
he EVCS unde s udy is ins alled. In o de o analyze he p opaga ion o he emissions
gene a ed by each EVCP indi idually, synch onized measu emen s we e ca ied ou a H2,
H3, and H5 when a single EV was cha ging a i s co esponding EVCS. The analysis o
he in e ac ion o he NIEs is based on measu emen s pe o med a H2, H3, and H5 when
he h ee EVs a e cha ging simul aneously.
Fo all he EVs unde s udy, eco dings o 600 s we e a ailable. The common ime
amewo ks used o PQ a e 20 ms, 3 s, and 1 day. Howe e , since a eco ding o 1 day
would esul in a la ge amoun o da a, implying an unmanageable ile, and 3 s would no
be enough ime o cha ac e ize he expec ed ime a ia ions, he measu emen ime was
con igu ed o 600 s. This measu emen ime co esponds o a ade-o be ween he ile
size and he ime needed o e alua e he a iabili y o he NIEs.
4.2.3. Measu emen campaign o he compa ison o he
emissions gene a ed by EVCPs unde isola ed and on-line
condi ions
The eco ding o he emissions gene a ed by EVCPs unde isola ed condi ions was
pe o med as desc ibed in sec ion 4.2.1.
The eco ding o he emissions gene a ed by EVCPs unde on-line condi ions was
pe o med a a pa king plaza, whe e he EVCS was di ec ly connec ed o he LV
50
dis ibu ion g id, as shown in Fig. 9. Fo his pu pose, he measu emen sys em o he
assessmen o he NIEs desc ibed in sec ion 4.1.1 was used.
Fig. 9. Measu emen scena io o he eco ding o he NIEs gene a ed by EVs du ing hei cha ging p ocess in
he LV g id (on-line condi ions).
This measu emen campaign conside s:
- The eco ding o he emissions gene a ed by se en comme cial EV models wi h
di e en cha ging cu en s unde isola ed condi ions. In he case o EV1, EV3,
and EV7, only wo cha ging cu en s we e possible (8A and 12A), whe eas h ee
cha ging cu en s we e possible o EV2, EV4, EV5, and EV6 (8A, 12A, and
16A).
- The eco ding o he emissions gene a ed by i e comme cial EV models wi h
di e en cha ging cu en s unde on-condi ions. In he case o EV10, only one
cha ging cu en was possible (16 A), while i e cha ging cu en s (8 A, 10 A,
12 A, 14 A, and 16 A) we e possible o he emaining EV models.
Fo all he EVs unde s udy, eco dings o 600 s we e a ailable, excep o EV6 wi h a
cha ging cu en o 8A, o which 150 s we e eco ded.
Fo he compa ison o he emissions gene a ed by a speci ic EV model unde isola ed and
on-line condi ions, EV1 and EV10 (16 A), EV2 and EV8 (8 A, 12 A, and 16 A), EV3 and
EV12 (12 A and 16 A), EV6 and EV11 (8 A, 12 A, and 16 A), and EV7 and EV9 (8 A,
12 A, and 16 A) a e conside ed.
4.2.4. Measu emen campaign o he cha ac e iza ion o he
sub-cycle and mean g id impedance in p esence o EVCPs
The measu emen campaign o he cha ac e iza ion o he sub-cycle and mean g id
impedance was ca ied ou in he same measu emen scena io desc ibed in sec ion 4.2.2.
The s udy conside s impedance measu emen s in he de aul si ua ion (no EV is cha ging
in he g id) and when each EV is cha ging. As in sec ion 4.2.2, each EV model can only
be cha ged a i s co esponding EVCS.
4.2.5. Measu emen campaign o he cha ac e iza ion o he
long- e m g id impedance a ia ions in p esence o EVCPs
The measu emen campaign o he e alua ion o he long- e m impedance a ia ions in
he p esence o EVCPs was ca ied ou in he con olled LV g id a he Uni e si y o
51
Applied Sciences Vienna (UASV) shown in Fig. 10. This scena io is composed o a SS and
ou houses (T1, T2, T3, T4) a which di e en elec onic de ices can be connec ed.
Monophasic measu emen s we e conduc ed a one o he phases o he ins alla ion (L1),
conside ing a s a opology.
Fig. 10. Rep esen a ion o he con olled LV g id a he UASV (Aus ia) [105].
In his measu emen campaign, he de aul si ua ion co esponds o he con igu a ion in
which he de ices ga he ed in Table IV a e connec ed. Apa om he measu emen
loca ion, he phase (L) o which each de ice is connec ed is also indica ed in Table IV.
L123 e e s o a de ice connec ed o he h ee phases ( ee-phased).
Table IV. Connec ed de ices a each house (T1, T2, T3, T4) in he con olled LV g id a UASV.
Measu emen loca ion
Connec ed de ices
T1
2 Sma Me e s (SMs) (L123)
T2
3 SMs (L123)
Pho o ol aic (PV) sys em (L2)
T3
2 SMs (L123)
PV sys em (L2)
1200 W load (L123)
T4
3 SMs (L123)
PV sys em (L123)
S o age sys em (L123)
In o de o e alua e he long- e m g id impedance a ia ions, ou EVs (EVA, EVB, EVC,
and EVD) we e indi idually cha ged a an EVCS connec ed o he POC o T2. The
connec ion o he EVCS o he g id includes a long cable o 30 me e s, which simula es
he ypical cable leng h in eal cha ging si ua ions.
52
4.2.6. In luence o conduc ed emissions gene a ed by EVCPs
on NB-PLC
The labo a o y se up o he e alua ion o NB-PLC acco ding o PRIME 1.4, which is
shown in Fig. 11, is composed o wo PL360G55CF-EK boa ds [218] supplied by a lap op,
a a iable a enua o con ollable by so wa e (A 1 in Fig. 11), a manual a iable
a enua o (A 2 in Fig. 11), and a Uni e sal So wa e Radio Pe iphe al (USRP) [219],
esponsible o ep oducing he eco ded NIEs. The a enua o con ollable by so wa e
in oduces a maximum a enua ion o 62.5 dB wi h s eps o 0.25 dB, while he manual
a enua o allows in oducing a enua ions up o 110 dB wi h inc emen s o 1 dB and
10 dB.
Fig. 11. Labo a o y scena io o he e alua ion o NB-PLC acco ding o PRIME 1.4 s anda d.
The measu emen campaign includes he conduc ed emissions gene a ed by he cha ging
p ocesses o he se en comme cial EV models wi h di e en cha ging cu en s de ailed in
4.2.3 unde isola ed condi ions. As p e iously men ioned, in he case o EV1, EV3, and
EV7, only wo cha ging cu en s we e possible (8A and 12A), whe eas h ee cha ging
cu en s we e possible o EV2, EV4, EV5, and EV6 (8A, 12A, and 16A). Fo all he EVs
unde s udy, eco dings o 600 s a e a ailable, excep o EV6 wi h a cha ging cu en o
8A, o which 150 s we e eco ded.
4.2.7. In luence o impedance a ia ions on NB-PLC
In o de o cha ac e ize he in luence o impedance a ia ions, a con olled labo a o y
scena io is de ined and implemen ed (see Fig. 12). I is composed o h ee LISNs,
esponsible o o e ing a ime-in a ian s anda dized impedance a he EUT po , so ha
he measu emen s a e no a ec ed by he a ia ions o he impedance o he g id. Besides,
LISNs pe o m a low-pass il e unc ion, wi h he aim o p e en ing unwan ed g id noise
om en e ing he EUT. In all he measu emen s, he cha ac e is ic impedance o he
LISNs is se o 2 Ω, acco ding o he PRIME speci ica ion [57]. A de ailed cha ac e iza ion
o he modulus o he impedance equency esponse o he LISNs can be ound in Fig.
13.
53
Fig. 12. Labo a o y scena io. Loads unde s udy a e connec ed a ei he poin s A, B, o C [57].
Fig. 13. Measu ed impedance modulus o he LISN model o he equency band o in e es .
In addi ion o he FER-SNR cu es ha e alua e he po en ial e ec o impedance
a ia ions on he quali y o he communica ions, in he case o he g id impedance, he
a enua ion due o he load unde es is also calcula ed. This a enua ion is ob ained as he
di e ence o he modulus o he channel equency esponse when he co esponding load
is connec ed o he se up and he modulus o he channel equency esponse when no load
is connec ed. Fo he e alua ion o he in luence o equency-dependen impedance
a ia ions, a se o Elec omagne ic Compa ibili y (EMC) il e s a e selec ed, while o he
sub-cycle impedance a ia ions a se o comme cial de ices a e e alua ed, as well as a
syn he ic load designed in he labo a o y.
NB-PLC acco ding o PRIME 1.4 s anda d a e es ablished by means o wo Mic ochip
PL360G55CF-EK e alua ion ki s [218], in such a way ha one is con igu ed o ac as a
ansmi e and he o he ac s as he ecei e . The ansmi e equipmen is connec ed o
he EUT po o he i s LISN, whe eas he ecei e is connec ed o he EUT po o he
hi d LISN. The second LISN allows connec ing di e en loads on he channel o analyze
54
hei e ec on communica ions. This way, he in luence o he load loca ion on he
a enua ion di e ence su e ed by he ansmi ed signal is pe o med by connec ing he
loads unde s udy o he EUT po o each o he h ee LISNs al e na ely (poin A, B, and
C in Fig. 12).
In o de o ensu e ha he measu emen s a e no a ec ed by he in e nal noise o he
ecei e and he noise in oduced by he load unde es in he case o he ac i e loads,
AWGN is injec ed by means o a signal gene a o connec ed o he RF po o he hi d
LISN. In his way, he noise o he measu emen se up is la o he whole equency band
o in e es and masks he po en ial e ec o he noise in oduced by he di e en
equipmen connec ed o he se up.
The e o e, aking in o accoun ha he noise o he sys em is gi en by a ixed le el o
AWGN, i is necessa y o modi y he powe o he ansmi ed signal, so ha he di e en
SNR alues ha make up he FER-SNR cu es a e ob ained. Fo ha pu pose, wo
a enua o s a e loca ed be ween he second and he hi d LISNs: a 110 dB a enua o wi h
1 dB and 10 dB s ep sizes and a 1 dB a enua o wi h 0.1 dB s ep size. A 110 dB a enua o
is also connec ed be ween he i s and he second LISN, in o de o ensu e a su icien
le el o a enua ion o ob ain he FER-SNR cu es. I s le el is ixed o 3 dB, excep o
he cases when i is no possible o ge a FER o 5 % jus by a ying he a iable a enua o s
loca ed be ween he second and hi d LISNs.
4.2.8. E alua ion o NB-PLC in a econs uc ed LV g id
The e alua ion o NB-PLC was ca ied ou in he econs uc ed LV g id a UASV
desc ibed in sec ion 4.2.5.
NB-PLC acco ding o PRIME 1.4 we e e alua ed by means o wo PL360G55CF-EK
boa ds, one ac ing as a ansmi e and connec ed o he SS and he o he ope a ing as a
ecei e and connec ed o he co esponding house (H1, H2, H3, and H4). In o de o
e alua e he pe o mance o PRIME 1.4, o each con igu a ion, he mean alue o he
SNR o he ansmi ed ames is ela ed o he FER.
55
5. Resul s and con ibu ions
This sec ion summa izes he main esul s and con ibu ions o his Doc o al Thesis. In
sec ion 5.1, he esul s ela ed o he cha ac e iza ion o he g id a e p esen ed, while
sec ion 0 add esses he e alua ion o NB-PLC unde di e en channel condi ions.
5.1. Cha ac e iza ion o he elec ical g id as a
ansmission medium
5.1.1. Cha ac e iza ion o he conduc ed emissions
Cha ac e iza ion o he conduc ed emissions gene a ed by
EVCPs unde isola ed condi ions
In his con ibu ion, a i s quali a i e spec al and ime cha ac e iza ion o he conduc ed
emissions gene a ed by a se o EVCPs using a LISN is p esen ed. The s udy, co e ing he
9-500 kHz equency band, conside s he emissions gene a ed by wo EV models and wo
cha ging cu en s du ing 5 s.
The esul s show ha bo h he spec al pa e n and ampli ude o he emissions depend on
he EV model and cha ging cu en unde s udy. In some cases, high-ampli ude onal o
na owband emissions a e eco ded a speci ic equencies, while, in o he , emissions in
he o m o colo ed noise dec easing wi h equency a e epo ed. An example o hese
spec al cha ac e is ics can be ound in Fig. 14.
Fig. 14. QP alues o he ampli ude o he emissions gene a ed by a ce ain EV model o cha ging cu en s o
9 A and 10 A.
Rega ding he ime analysis, he spec og ams o he eco dings show ha a signi ican
ime a iabili y is egis e ed in he equency band assigned o NB-PLC (see Fig. 15). In
62
Fig. 22. QP alues o he ampli ude o he emissions gene a ed by a ce ain EVCP in each pe iod o 50 s du ing
he eco ding ime (600 s) unde on-line condi ions.
Excep o he onal emissions wi h oscilla ing cen al equency gene a ed by a speci ic
EV model ega dless o he cha ging cu en , he onal and na owband emissions
gene a ed by he emaining EVCPs ollow he simpli ied equa ion o he model de ined in
he p e ious sec ion. Since he FFT componen a 100 Hz is he majo con ibu o o he
o al a iabili y in hese cases, i can be concluded ha he NIEs gene a ed by hese EVCPs
a y wi hin 10 ms, i.e., wo imes wi hin he mains cycle. This migh be due o he
posi i e-nega i e symme y o he powe con e sion s age. In he speci ic case o he onal
emissions wi h oscilla ing cen al equency, he FFT analysis shows a wide ange o
spec al componen s sp ead o e he whole FFT equency band, being he FFT
componen s a ound 41 Hz and i s mul iples (mainly 82 Hz and 123 Hz) he ones ha
p esen highe ampli ude (see Fig. 23). This implies ha hese emissions ollow a pe iodic
pa e n wi h a epe i ion a e o 24.4 ms.
63
Fig. 23. Modulus o he no malized FFT (dB) o he ime samples co esponding o a onal emission wi h
oscilla ing cen al equency.
Finally, i should be men ioned ha , only o h ee ou o he wel e EVCPs unde s udy,
he o al a iabili y di e s by mo e han 3 dB be ween bo h measu emen con igu a ions.
In hese cases, a highe a iabili y is epo ed i a LISN is used.
In summa y, his is he i s empi ical s udy ha p esen s a compa ison be ween he
emissions gene a ed by EVCPs unde isola ed and on-line condi ions conside ing a wide
ange o EV models and cha ging cu en s. The con ibu ion e eals ha , since he
equency and ime cha ac e is ics o he emissions depend on he measu emen
condi ions, his cha ac e iza ion canno be only based on ials conduc ed using a LISN.
Mo eo e , due o he di e en beha io o he emissions be ween he 9-150 kHz and
150-500 kHz equency bands, his s udy highligh s ha i is no possible o di ec ly
ex apola e he unde s anding o he NIEs in he 9–150 kHz equency ange, ypically
add essed in he li e a u e, o he 150–500 kHz band.
The jou nal pape ela ed o his con ibu ion (JP3) can be ound in Appendix A.3.
5.1.2. Cha ac e iza ion o he g id impedance
Cha ac e iza ion o he sub-cycle, mean, and long- e m
impedance a ia ions in he p esence o EVCPs
This con ibu ion aims a analyzing he in luence o EVCPs on he g id impedance in he
20-500 kHz equency band. Fo his pu pose, wo measu emen campaigns a e ca ied
ou in con olled LV dis ibu ion g ids in F ance and Aus ia. The i s measu emen
campaign aims o e alua e he sub-cycle and mean impedance due o EVCPs, while he
second analyzes he long- e m impedance a ia ions (some hou s) unde hese
ci cums ances.
64
Rega ding he i s measu emen campaign, he esul s show impedance alues lowe han
18 Ω in he equency band 20-500 kHz when measu ing a h ee loca ions unde s udy
(see Fig. 24). Some esonances a e obse ed a ce ain equencies in he modulus, which
imply ab up changes in he phase. These esonances, also epo ed in [135], [220], could
imply an inc ease in he ampli ude o he emissions gene a ed by a ce ain de ice [41] and,
hus, could endange he p ope ope a ion o PLC.
The obse ed alues go in line wi h he esul s p esen ed in [36], [135] o u ban scena ios
in he LV g id in Spain and China, espec i ely. By con as , hey conside ably di e om
he e e ence impedance alues included in he no ma i e cu en ly in o ce. As shown in
Fig. 24, in he 20-500 kHz equency ange, he modulus o he impedances de ined in
CISPR 16-1-2 anges om 7.3 Ω o 47.7 Ω.
(a) (b)
Fig. 24. Modulus and phase o he mean g id access impedance measu ed a he h ee loca ions unde s udy
when no EV is connec ed (de aul si ua ion), oge he wi h he e e ence impedance de ined in [103].
The s udy also analyzes he in luence o an EVCP on he mean g id access impedance by
compa ing he impedance measu ed in he de aul si ua ion (D) and when each EV is
cha ging, as shown in Fig. 25.
(a) (b)
Fig. 25. Modulus and phase o he mean g id access impedance measu ed a h ee loca ions unde s udy when
no EV is connec ed (de aul si ua ion) and when each EV is cha ging.
65
Fig. 25 shows di e ences o a ew ohms be ween he modulus o he impedance in he
de aul si ua ion and when each EV is cha ging a he same loca ion. Ne e heless, due o
he low alues o he g id impedance in he de aul si ua ion, hese di e ences in ampli ude
ep esen ela i e a ia ions ha migh be signi ican a speci ic equencies. As i can be
obse ed, in gene al, he spec al o m o he impedance, bo h he modulus and he phase,
is main ained in he whole equency band, excep o ce ain esonances occu ing in he
equency ange below 150 kHz. Mo eo e , he cha ging o an EV does no seem o imply
a displacemen o he p e iously exis ing esonances, as hey emain a he same
equencies as in he de aul si ua ion.
The s udy also add esses he in luence o an EVCP a a ce ain dis ance om he elec ical
poin whe e he EVCS is ins alled, concluding ha i seems ha an inc ease in he dis ance
does no educe he in luence on impedance. The e o e, hese esul s lead o conclude ha
he in luence o an EVCP does no only depend on he EVCP i sel , bu also on he g id
impedance in he de aul si ua ion.
Wi h espec o sub-cycle impedance a ia ions, as shown in Fig. 26, his s udy also leads
o conclude ha sub-cycle a ia ions, occu ing wi hin he 20 ms o he mains, occu when
an EV is cha ging, bu also in he de aul si ua ion. Mo e de ails abou bo h he mean and
sub-cycle a ia ions can be ound in he in e na ional con e ence pape (ICP2) shown in
Appendix B.1.2.
Fig. 26. Modulus and phase o he sub-cycle g id access impedance measu ed a a ce ain loca ion when no EV
is connec ed (de aul si ua ion).
Rega ding long- e m impedance a ia ions, his con ibu ion shows ha he spec al
cha ac e is ics (bo h ampli ude and spec al shape) o he g id impedance a y conside ably
du ing he cha ging p ocess o an EV (see Fig. 27). Speci ically, wo di e en impedance
s a es a e epo ed wi hin he 6070 s (impedance s a e 1 om 0 s o 3710 s and 5880 s o
5900 s, and impedance s a e 2 om 3710 s o 5880 s and om 5900 s o 6070 s). This
demons a es ha signi ican impedance a ia ions can occu du ing he cha ging p ocess
o an EV.
66
(a) (b)
Fig. 27. Modulus (a) and phase (b) o he g id impedance measu ed du ing 6070 s o he cha ging p ocess o a
ce ain EV wi h espec o ime ( e ical axis) and equency (ho izon al axis).
Al hough hey a e no clea ly no iceable in Fig. 27 due o he wide ange o alues shown
in he igu e, a ia ions wi hin each impedance s a e a e also gi en. An example o hese
a ia ions a e depic ed in Fig. 28. The a ia ions occu ing in he g id impedance o he
emaining EVCPs unde s udy can be ound in he jou nal pape (JP4) included in
Appendix A.4
(a) (b)
Fig. 28. Modulus (a) and phase (b) o he g id impedance measu ed du ing he cha ging p ocess o a ce ain EV
and when no EV is cha ging (de aul si ua ion).
Fig. 28Fig. 28 clea ly shows ha , ega dless o he impedance s a e and EVCP unde s udy,
no only a di e ence in he impedance ampli ude is epo ed, bu also he spec al shape
o he impedance a ies. Mo eo e , Fig. 28 also e eals a displacemen o he esonance
a ound 56 kHz occu ing in he de aul si ua ion o e en he appea ance o new esonances
a di e en equencies when each EV is cha ging. This in ol es ha he beha io o he
impedance migh depend on he measu emen scena io, which implies ha i is necessa y
o pe o m measu emen campaigns in di e en g id condi ions ( opology, ype o cable,
numbe o connec ed loads…) conside ing a wide ange o EV models and cha ging
con igu a ions.
67
I should also be no ed ha , a equencies below 150 kHz, low-ampli ude esonances
( alues lowe han 5 Ω) a e obse ed bo h in he de aul si ua ion and when each EV is
cha ging. These low impedances can esul in high a enua ion, which could be c i ical o
he p ope pe o mance o PLC [45], [120]. In he 150-500 kHz equency ange, simila
spec al ea u es a e egis e ed bo h in he de aul si ua ion and o all he EVCPs unde
s udy ( alues lowe han 15 Ω in he modulus and an induc i e beha io in he phase).
Thus, he beha io o he g id impedance seems o be mo e s able a equencies abo e
150 kHz (bo h he ampli ude and he spec al shape), which would acili a e he p ope
design o NB-PLC de ices acco ding o he cha ac e is ics o he communica ions channel
in his equency band.
In conclusion, his is he i s empi ical s udy ha analyzes he in luence o EVCPs on he
sub-cycle, mean, and long- e m g id impedance in he 20-500 kHz equency band. Since
he esul s show ha he EVCPs imply bo h sub-cycle and long- e m a ia ions, he
cha ac e iza ion o he impedance in p esence o EVs is a complex phenomenon ha
equi es u he analysis. Fo his eason, in o de o de elop a mo e gene al model, a la ge
numbe o measu emen campaigns should be ca ied ou all a ound he wo ld, aking in o
accoun di e en g id opologies, a eas wi h di e en popula ion densi ies, di e en EV
models, and cha ging condi ions. This would allow designing communica ion de ices
acco ding o he cha ac e is ics o he g id, leading o a be e pe o mance o PLC
sys ems.
The esul s ob ained in his s udy we e p esen ed in an in e na ional con e ence (ICP2),
which can be ound in Appendix B.1.2, as well as in he jou nal pape (JP4) included in
Appendix A.4.
68
5.2. E alua ion o NB-PLC unde di e en channel
condi ions
5.2.1. In luence o he spec al cha ac e is ics o he conduc ed
emissions on NB-PLC
This con ibu ion analyzes he in luence o he spec al cha ac e is ics o he conduc ed
emissions gene a ed by EVCPs on NB-PLC acco ding o PRIME 1.4 s anda d. Fi s , a
cha ac e iza ion o he emissions in he equency domain is p esen ed and, hen, he
pe o mance o PRIME 1.4 in p esence o hese emissions is e alua ed. Fo his pu pose,
a measu emen campaign is pe o med (desc ibed in 4.2.3), whe e he emissions gene a ed
by a se o EVs a e eco ded a an EVCS connec ed o a LISN, and hen ep oduced in a
con olled and p e iously well-cha ac e ized labo a o y scena io (desc ibed in sec ion
4.2.6).
In o de o e alua e he pe o mance o Fo wa d E o Co ec ion (FEC) and epe i ion
codes in he p esence o sub-cycle a ia ions, he s udy conside s Di e en ial Quad a u e
Shi Keying (DQPSK) wi h FEC (DQPSK_C), Di e en ial Bina y PSK wi h FEC
(DBPSK_C), Robus DQPSK (R_DQPSK), and R_DBPSK. In all he measu emen s,
1000 Type B ames a e ansmi ed conside ing a ixed 256-by e leng h message and
equency channels 1 and 3-8 o PRIME 1.4.
The esul s ela ed o he cha ac e iza ion o he emissions lead o conclude ha he
spec al pa e n o he emissions is highly dependen on he equency band unde analysis.
In channel 1 (42-89 kHz), se e al high-ampli ude onal o na owband emissions a
di e en equencies a e obse ed. By con as , in channels 3-8 (151-471 kHz), la
backg ound noise wi h low-ampli ude onal emissions a e egis e ed ega dless o he EV
model and cha ging cu en . The onal emissions in he 42-89 kHz equency ange
co espond, in mos cases, o ha monics o he swi ching equency o he in e e
included in he ci cui y o he EV cha ge [126]. The spec al ea u es o hese NIEs can
be ound in he con e ence pape s ela ed o his con ibu ion (ICP3 and NCP1), which
a e included in Appendix B.1.3 and Appendix B.2.1, espec i ely.
The e alua ion esul s show ha , due o he di e en spec al shape o he emissions
depending on he equency ange, he h esholds ha se he quali y o communica ions
a y conside ably om one equency channel o ano he . By con as , no di e ences in
he minimum SNR needed o achie e a FER o 5 % a e epo ed o a gi en EV model
and di e en cha ging cu en s, since simila spec al ea u es a e ob ained in hese cases.
Mo eo e , a be e pe o mance o modula ions including epe i ion codes (R_DQPSK
and R_DBPSK) han modula ions using only FEC (DQPSK_C, DBPSK_C) is shown. In
gene al, he ou modula ions a e capable o co ec ing onal o na owband emissions
occu ing a ce ain equencies, p o iding a minimum SNR o achie e a FER o 5 % lowe
han he h eshold co esponding o AWGN. Howe e , when backg ound noise wi h
mul iple low-ampli ude emissions is measu ed, as p esen ed in channels 3-8 o all EVCPs
unde s udy, he h esholds a e p ac ically iden ical o he AWGN case. The only excep ion
o his is EV1 wi h cha ging cu en s o 8 A and 12 A using DQPSK_C and DBPSK_C
modula ions, whe e he e is a la ge numbe o emissions wi h small equency gaps
69
be ween hem, which esul s in o addi ional deg ada ion wi h espec o he AWGN case.
The e o e, he e ec o onal emissions on NB-PLC is de e mined by he numbe o
emissions, hei ampli ude wi h espec o he backg ound noise and he equency gap
be ween consecu i e peaks.
The FER-SNR cu es p esen ed in his con ibu ion ake in o conside a ion he spec al
pa e n o he emissions and no hei ac ual ampli ude. Howe e , since he emissions in
channel 1 a e signi ican ly highe han he emissions in channels 3-8, he minimum ecei ed
signal le el needed o ob ain a FER o 5 % in channel 1 could be highe han he
co esponding o channels 3-8. In o de o demons a e his beha io , as an example, he
RSSI needed o achie ing a FER o 5 % is compa ed o EV2 wi h a cha ging cu en o
8 A wi h DBPSK_C modula ion in channels 1 and 3. While in channel 1 his pa ame e
akes a alue o 83 dBµV, only 60 dBµV a e equi ed o ob ain he same alue o FER in
channel 3. Howe e , due o he spec al pa e n o he emissions, he SNR equi ed o
ob aining a FER o 5 % is highe in channel 3 (5.9 dB in channel 3 compa ed o 4.9 dB in
channel 1).
The e o e, wi h he aim o comple ely e alua ing he in luence o he conduc ed emissions
gene a ed by EVCPs in e ms o hei spec al pa e n, ac ual ampli ude o he NIEs and
hei ime-dependen beha io , u u e wo k should combine he minimum SNR and RSSI
equi ed o ob aining a FER o 5 %.
The con e ence pape s ela ed o his con ibu ion (ICP3 and NCP1) can be ound in
Appendix B.1.3 and Appendix B.2.1.
5.2.2. In luence o he g id impedance a ia ions
In luence o equency-dependen impedance a ia ions
This con ibu ion add esses he e alua ion o he in luence o he equency-dependen
g id impedance on NB-PLC acco ding o PRIME 1.4. The s udy is ca ied ou in he
labo a o y se up desc ibed in 4.2.7. In his s udy, equency channels 1 and 3-8 de ined in
PRIME 1.4 a e e alua ed by ansmi ing 1000 ames wi h a 256-by e leng h using a
DBPSK modula ion.
In o de o e alua e he in luence o equency-dependen impedance a ia ions on
PRIME 1.4, ou EMC il e s designed acco ding o IEC 60939 a e conside ed, which
show impo an equency-dependen impedance a ia ions ha emain cons an o e
ime. These il e s, which a e used o elec omagne ic in e e ence supp ession, pe o m
a double unc ion. Fi s , he il e p o ec s an elec onic con ol ci cui om ol age spikes
in he mains supply, which may be gene a ed, o example, by elec omechanical swi ches
and elays. Simul aneously, he same il e also ac s in he opposi e di ec ion, a enua ing
he in e e ence gene a ed in he uni owa ds he powe supply line. Howe e , he use o
EMC il e s migh also lead o an undesi ed e ec on NB-PLC signals. Due o he pa allel
and/o se ies esonances o EMC il e s, no ching e ec s can a ec communica ion
signals, which a e mo e likely o be dis u bed when he ou pu impedance o he il e is
low in compa ison wi h he impedance o he g id.
This con ibu ion conside s bo h impedance a ia ions due o he EMC il e s in an
open-ci cui con igu a ion, i.e., when no ex e nal load is connec ed, and in a loaded
70
con igu a ion, when a 50 Ω esis o in se ies wi h a 560 nF capaci o is connec ed o he
il e . Fig. 29 and Fig. 30 show he modulus o he impedance measu ed when he ou
EMC il e s a e connec ed o he se up wi hou and wi h a load connec ed in L’-N’ po s,
espec i ely.
Fig. 29. Impedance modulus o he ou EMC il e s connec ed o he se up wi hou connec ing a load in L’-N’
po s.
Fig. 30. Impedance modulus o he ou EMC il e s connec ed o he se up when a load is connec ed in L’-N’
po s.
Fig. 29 and Fig. 30 e eal ha hese il e s p esen a wide ange o impedance alues along
he equency ange om 10 kHz o 500 kHz, mainly due o he esonance e ec s hey
cause a speci ic equencies. The a ia ions wi h espec o he impedance o he LISN
71
shown in Fig. 12 a e he esul o he combina ion o he impedance o each EMC il e
and he componen s o he LISN. The analysis o he schema ics o he il e s lead o
conclude ha he impedance a ia ions in oduced by he EMC il e s a e mainly due o
he capaci i e in e ace o he il e s wi h he ne wo k.
Rega ding he e alua ion o he pe o mance o PRIME 1.4 in he p esence o
equency-dependen impedance a ia ions, as p e iously desc ibed in sec ion 4.2.7, he
s udy is based on he calcula ion o wo pa ame e s: he a enua ion due o he EMC il e s
unde es and he po en ial deg ada ion o he FER-SNR cu es ha se he quali y o he
communica ions.
Conce ning he signal a enua ion, he esul s p esen ed in his s udy show ha he
impedance a ia ions due o he connec ion o he EMC il e s esul in high a enua ion
in he equency band assigned o NB-PLC (see Fig. 31). Acco ding o he pe o med ials,
he highes a enua ion is epo ed in hose equency channels in which he modulus o
he g id impedance is low and main ained in he equency channel (low s anda d
de ia ion) and in which he phase p esen s ab up a ia ions. These a enua ions could
cause communica ions o ail i he noise le el is high o i he ecei ed signal powe is
close o he sensi i i y limi o he ecei ing equipmen .
Fig. 31. A enua ion due o he EMC il e s unde s udy in open-ci cui con igu a ion. Blue lines indica e ha
he il e s a e connec ed a poin A, ed lines indica e connec ion poin B, and g een lines indica e connec ion
poin C.
Howe e , he esul s also show ha he a ia ions in he channel equency esponse due
o he impedance a ia ions because o he EMC il e s a e no selec i e enough o deg ade
NB-PLC in e ms o he SNR h esholds ha se he quali y o he communica ions (see
Fig. 32). This in ol es ha PRIME 1.4 pe o ms p ope ly unde hese ci cums ances, no
being a ec ed by he channel cha ac e is ics caused by he in oduc ion o he EMC il e s.
78
6. Impac
The esul s o his Doc o al Thesis ha e been published in scien i ic jou nals indexed in
he Jou nal Ci a ion Repo s (JCR), as well as in in e na ional and na ional con e ences. In
Table V, he jou nal pape s associa ed o he con ibu ions o his Doc o al Thesis a e
ga he ed. Six o hese publica ions ha e al eady been published, while one o hem is unde
e iew a he ime o publica ion o his hesis.
Table V. Summa y o he jou nal pape s associa ed o he con ibu ions o his Doc o al Thesis.
Publica ion
code
Ti le
Publishe
Jou nal
Yea
IF/Qua ile
JP1 [42]
A Re iew on he Empi ical
Cha ac e iza ion o he Low
Vol age Dis ibu ion G id as
a Communica ion Channel
o Powe Line
Communica ions
Else ie
Sus ainable
Ene gy, G ids
and Ne wo ks
2023
4.8/Q1
JP2 [43]
Emissions Gene a ed by
Elec ic Vehicles in he
9-500 kHz Band:
Cha ac e iza ion,
P opaga ion, and In e ac ion
Else ie
Elec ic Powe
Sys ems
Resea ch
2024
3.3/Q2
JP3 [44]
Compa ison o Conduc ed
Emissions Due o Elec ic
Vehicle Cha ging P ocesses
unde Isola ed and On-Line
Condi ions in he 9-500 kHz
F equency Range
Else ie
Sus ainable
Ene gy, G ids
and Ne wo ks
2024
4.8/Q1
JP4
Cha ac e iza ion o he
Long-Te m Impedance
Va ia ions due o Elec ic
Vehicle Cha ging om
20 kHz o 500 kHz
IEEE
IEEE Open
Jou nal o
Powe and
Ene gy
Unde
Re iew
3.3/Q2
JP5 [45]
Cha ac e iza ion o he
Po en ial E ec s o EMC
Fil e s o Powe Con e e s
on Na owband Powe Line
Communica ions
MDPI
Elec onics
2021
2.69/Q3
JP6 [46]
Cha ac e iza ion o he
Po en ial E ec s o
Sub-Cycle Impedance
Va ia ions on PRIME 1.4
Else ie
Enginee ing
Science and
Technology, an
In e na ional
Jou nal
2024
5.1/Q1
JP7 [47]
Upg ading he Powe G id
Func ionali ies wi h
B oadband Powe Line
Communica ions: Basis,
Applica ions, Cu en
T ends and Challenges
MDPI
Senso s
2022
3.9/Q2
In Table VI and Table VII, a summa y o he pape s p esen ed in in e na ional and na ional
con e ences du ing his Doc o al Thesis a e de ailed.
79
Table VI. Summa y o he in e na ional con e ence pape s associa ed o he con ibu ions o his Doc o al
Thesis.
Publica ion
code
Ti le
Con e ence
Yea
ICP1 [224]
Empi ical cha ac e iza ion o he conduc ed
dis u bances gene a ed by he elec ic ehicles
du ing he cha ging p ocess
CIRED Po o
Wo kshop 2022
2022
ICP2 [225]
In luence o Elec ic Vehicle Cha ging on he G id
Access Impedance om 20 kHz o 500 kHz
2023 In e na ional
Con e ence on
Sma Ene gy
Sys ems and
Technologies
(SEST)
2023
ICP3 [226]
In luence o he Spec al Pa e n o he Conduc ed
Emissions Gene a ed by Elec ic Vehicle Cha ging
on PRIME 1.4
IEEE
In e na ional
Con e ence on
Communica ions,
Con ol, and
Compu ing
Technologies o
Sma G ids
(Sma G idComm)
2024
ICP4
Cha ac e iza ion o he Po en ial E ec s o EMC
Fil e s on Powe Line Communica ions
CIGRE
Cen ennial Session
Pa is
2021
ICP5
E alua ion o PRIME 1.4 in a Recons uc ed Low
Vol age G id
14 h Wo kshop o
Powe line
Communica ions
(WSPLC)
2023
ICP6
BB-PLC o e LV ne wo ks – a s ep o wa d
owa ds Sma G id implemen a ion
ICSC-CITIES: V
Ibe o-Ame ican
Cong ess o Sma
Ci ies
2022
ICP7 [220]
Cha ac e iza ion o he LV dis ibu ion g id o he
deploymen o a pilo BB-PLC ne wo k
2023 In e na ional
Symposium on
Powe Line
Communica ions
and i s
Applica ions
(ISPLC)
2023
Table VII. Summa y o he na ional con e ence pape s associa ed o he con ibu ions o his Doc o al Thesis.
Publica ion
code
Ti le
Con e ence
Yea
NCP1
Análisis del impac o de las emisiones conducidas
gene adas po la ca ga del ehículo eléc ico en las
ecnologías NB-PLC
II Cong eso de
Redes In eligen es
Fu uRed
2023
NCP2
Ca ac e ización de la ed eléc ica como medio de
ansmisión de se icios de banda ancha (BB-PLC)
II Cong eso de
Redes In eligen es
Fu uRed
2023
I should also be no ed ha he ollowing con ibu ions o his Doc o al Thesis ha e been
included in he 4 h S udy Repo o he WG CLC/TC57-219/WG11 o CENELEC:
- Emissions om EV cha ge s.
- P opaga ion o he emissions gene a ed du ing EV cha ging.
- Impac o he EV cha ging p ocess on he impedance o he dis ibu ion ne wo k.
80
- Impedance cha ac e is ics o EMI il e s.
- A enua ion o MCE signals caused by EMI il e s.
- Sub-cycle a ia ions o he impedance.
Finally, his Doc o al Thesis has con ibu ed o he chap e “EMC wi h communica ions
sys ems” o he echnical b ochu e o he CIGRE WG C4.68 “Elec omagne ic
Compa ibili y issues in mode n and u u e powe sys ems”, ha will be published du ing
2025.
81
7. Bibliog aphy
[1] M. Seijo Simó, G. López, J. Ma anza, and J. Mo eno, “Planning and Pe o mance
Challenges in Powe Line Communica ions Ne wo ks o Sma G ids,” In J Dis ib
Sens Ne w, ol. 2016, Ma . 2016, doi: 10.1155/2016/8924081.
[2] P. Mlýnek, M. Rusz, L. Benešl, J. Sláčik, and P. Musil, “Possibili ies o B oadband
Powe Line Communica ions o Sma Home and Sma Building Applica ions,”
Senso s, ol. 21, no. 1, 2021, doi: 10.3390/s21010240.
[3] I. Colak, “In oduc ion o sma g id,” in 2016 In e na ional Sma G id Wo kshop and
Ce i ica e P og am (ISGWCP), 2016, pp. 1–5. doi: 10.1109/ISGWCP.2016.7548265.
[4] H. Gha a i and R. Gha u ian, “Sma G id: The Elec ic Ene gy Sys em o he
Fu u e [Scanning he Issue],” P oceedings o he IEEE, ol. 99, no. 6, pp. 917–921, 2011,
doi: 10.1109/JPROC.2011.2124210.
[5] G. Hallak and G. Bumille , “Time a ian ol age and cu en modeling
co esponding o access impedance measu emen s,” in 2018 IEEE In e na ional
Symposium on Powe Line Communica ions and i s Applica ions (ISPLC), 2018, pp. 1–6. doi:
10.1109/ISPLC.2018.8360226.
[6] D. Baimel, S. Tapuchi, and N. Baimel, “Sma g id communica ion echnologies-
o e iew, esea ch challenges and oppo uni ies,” in 2016 In e na ional Symposium on
Powe Elec onics, Elec ical D i es, Au oma ion and Mo ion (SPEEDAM), 2016, pp. 116–
120. doi: 10.1109/SPEEDAM.2016.7526014.
[7] A. Majumde and J. J. Ca e y, “Powe line communica ions,” IEEE Po en ials, ol.
23, no. 4, pp. 4–8, 2004, doi: 10.1109/MP.2004.1343222.
[8] A. Sendin e al., “Adap a ion o Powe line Communica ions-Based Sma Me e ing
Deploymen s o he Requi emen s o Sma G ids,” Ene gies (Basel), ol. 8, no. 12, pp.
13481–13507, 2015, doi: 10.3390/en81212372.
[9] J. Slacik, P. Mlynek, M. Rusz, P. Musil, L. Benesl, and M. P acek, “B oadband Powe
Line Communica ion o In eg a ion o Ene gy Senso s wi hin a Sma Ci y
Ecosys em,” Senso s, ol. 21, no. 10, 2021, doi: 10.3390/s21103402.
[10] P. Mlynek, J. Misu ec, Z. Kolka, J. Slacik, and R. Fujdiak, “Na owband Powe Line
Communica ion o Sma Me e ing and S ee Ligh ing Con ol,” IFAC-
Pape sOnLine, ol. 48, no. 4, pp. 215–219, 2015, doi:
h ps://doi.o g/10.1016/j.i acol.2015.07.035.
[11] A. Llano, D. De La Vega, I. Angulo, and L. Ma on, “Impac o Channel
Dis u bances on Cu en Na owband Powe Line Communica ions and Lessons o
Be Lea n o he Fu u e Technologies,” IEEE Access, ol. 7, pp. 83797–83811, 2019,
doi: 10.1109/ACCESS.2019.2924806.
[12] I. El eki, S. Jacques, I. Aouichak, T. Doligez, Y. Raingeaud, and J.-C. Le Bune el,
“Cha ac e iza ion o Na owband Noise and Channel Capaci y o Powe line
82
Communica ion in F ance,” Ene gies (Basel), ol. 11, no. 11, 2018, doi:
10.3390/en11113022.
[13] M. An oniali and A. M. Tonello, “Measu emen and Cha ac e iza ion o Load
Impedances in Home Powe Line G ids,” IEEE T ans Ins um Meas, ol. 63, no. 3,
pp. 548–556, 2014, doi: 10.1109/TIM.2013.2280490.
[14] J. Ana o y, N. Thee hayi, R. Tho appillil, M. M. Kissaka, and N. H. M ungi, “The
In luence o Load Impedance, Line Leng h, and B anches on Unde g ound Cable
Powe -Line Communica ions (PLC) Sys ems,” IEEE T ansac ions on Powe Deli e y,
ol. 23, no. 1, pp. 180–187, 2008, doi: 10.1109/TPWRD.2007.911020.
[15] D. Thomas, “Conduc ed emissions in dis ibu ion sys ems (1 kHz–1 MHz),” IEEE
Elec omagn Compa Mag, ol. 2, no. 2, pp. 101–104, 2013, doi:
10.1109/MEMC.2013.6550941.
[16] I. Fe nández, D. de la Vega, A. A inda, I. Angulo, N. U ibe-Pé ez, and A. Llano,
“Field T ials o he Cha ac e iza ion o Non-In en ional Emissions a Low-Vol age
G id in he F equency Range Assigned o NB-PLC Technologies,” Elec onics (Basel),
ol. 8, no. 9, 2019, doi: 10.3390/elec onics8091044.
[17] T. Slangen, T. an Wijk, V. Ćuk, and S. Cobben, “The P opaga ion and In e ac ion
o Sup aha monics om Elec ic Vehicle Cha ge s in a Low-Vol age G id,” Ene gies
(Basel), ol. 13, no. 15, 2020, doi: 10.3390/en13153865.
[18] T. S eubel, C. Ka mann, A. Eisenmann, and K. Rudion, “De ec ion and Moni o ing
o Sup aha monic Anomalies o an Elec ic Vehicle Cha ging S a ion,” in 2019
IEEE Milan Powe Tech, 2019, pp. 1–5. doi: 10.1109/PTC.2019.8810596.
[19] “CENELEC SC 205A. CLC/TR 50669. In es iga ion esul s on elec omagne ic
in e e ence in he equency ange below 150 kHz,” 2017.
[20] N. U ibe-Pé ez, I. Angulo, L. He nández-Callejo, T. A zuaga, D. la Vega, and A.
A inda, “S udy o Unwan ed Emissions in he CENELEC-A Band Gene a ed by
Dis ibu ed Ene gy Resou ces and Thei In luence o e Na ow Band Powe Line
Communica ions,” Ene gies (Basel), ol. 9, no. 12, 2016, doi: 10.3390/en9121007.
[21] P. T. Jensen and P. Da a i, “Powe Con e e Impedance and Emission
Cha ac e iza ion Below 150 kHz,” in 2021 IEEE In e na ional Join EMC/SI/PI and
EMC Eu ope Symposium, 2021, pp. 255–260. doi:
10.1109/EMC/SI/PI/EMCEu ope52599.2021.9559177.
[22] M. Yeshalem, B. Khan, and O. Mahela, “Conduc ed elec omagne ic emissions o
compac luo escen lamps and elec onic ballas modeling,” AIMS Elec onics and
Elec ical Enginee ing, ol. 6, pp. 178–187, Ap . 2022, doi: 10.3934/elec eng.2022011.
[23] J. Meye , S. Haehle, and P. Schegne , “Impac o highe equency emission abo e
2kHz on elec onic mass-ma ke equipmen ,” in 22nd In e na ional Con e ence and
Exhibi ion on Elec ici y Dis ibu ion (CIRED 2013), 2013, pp. 1–4. doi:
10.1049/cp.2013.1027.
83
[24] P. Ko sampopoulos e al., “EMC Issues in he In e ac ion Be ween Sma Me e s and
Powe -Elec onic In e aces,” IEEE T ansac ions on Powe Deli e y, ol. 32, no. 2, pp.
822–831, 2017, doi: 10.1109/TPWRD.2016.2561238.
[25] P. De Falco and P. Va ilone, “S a is ical Cha ac e iza ion o Sup aha monics in Low-
Vol age Dis ibu ion Ne wo ks,” Applied Sciences, ol. 11, no. 8, 2021, doi:
10.3390/app11083574.
[26] S. T. Y. Al alahi e al., “Sup aha monics in Powe G id: Iden i ica ion, S anda ds, and
Measu emen Techniques,” IEEE Access, ol. 9, pp. 103677–103690, 2021, doi:
10.1109/ACCESS.2021.3099013.
[27] A. A. Alkah ani e al., “Powe Quali y in Mic og ids Including Sup aha monics:
Issues, S anda ds, and Mi iga ions,” IEEE Access, ol. 8, pp. 127104–127122, 2020,
doi: 10.1109/ACCESS.2020.3008042.
[28] “ETSI TS 103 909 V1.1.1. Powe Line Telecommunica ions (PLT) Na ow band
anscei e s in he ange 9 kHz o 500 kHz Powe Line Pe o mance Tes Me hod
Guide. ,” Dec. 2012.
[29] G. Hallak and G. Bumille , “Impedance measu emen o elec ical equipmen loads
on he powe line ne wo k,” in 2017 IEEE In e na ional Symposium on Powe Line
Communica ions and i s Applica ions (ISPLC), 2017, pp. 1–6. doi:
10.1109/ISPLC.2017.7897099.
[30] G. Hallak, G. Bumille , and C. Nieß, “Accu a e access impedance measu emen s on
he powe line wi h op imized calib a ion p ocedu es,” in 2017 IEEE In e na ional
Ins umen a ion and Measu emen Technology Con e ence (I2MTC), 2017, pp. 1–6. doi:
10.1109/I2MTC.2017.7969803.
[31] C. Nieß, J.-P. Ki zig, and G. Bumille , “Measu emen Sys em o Time Va iable
Subcycle Impedance on Powe Lines,” in 2021 IEEE In e na ional Ins umen a ion and
Measu emen Technology Con e ence (I2MTC), 2021, pp. 1–6. doi:
10.1109/I2MTC50364.2021.9459875.
[32] I. Fe nández, M. Albe o, J. Mon albán, A. A inda, I. Angulo, and D. de la Vega, “A
new ol age p obe wi h imp o ed pe o mance a he 10 kHz–500 kHz equency
ange o ield measu emen s in LV ne wo ks,” Measu emen , ol. 145, pp. 519–524,
2019, doi: h ps://doi.o g/10.1016/j.measu emen .2019.05.106.
[33] G. Hallak, C. Nieß, and G. Bumille , “Accu a e Low Access Impedance
Measu emen s Wi h Sepa a ed Load Impedance Measu emen s on he Powe -Line
Ne wo k,” IEEE T ans Ins um Meas, ol. 67, no. 10, pp. 2282–2293, 2018, doi:
10.1109/TIM.2018.2814138.
[34] F. J. C. Co ipio, J. A. C. A abal, L. D. del Rio, and J. T. E. Munoz, “Analysis o he
cyclic sho - e m a ia ion o indoo powe line channels,” IEEE Jou nal on Selec ed
A eas in Communica ions, ol. 24, no. 7, pp. 1327–1338, 2006, doi:
10.1109/JSAC.2006.874402.
84
[35] L. G. da S. Cos a e al., “Access impedance in B azilian in-home, b oadband and low-
ol age elec ic powe g ids,” Elec ic Powe Sys ems Resea ch, ol. 171, pp. 141–149,
2019, doi: h ps://doi.o g/10.1016/j.eps .2019.02.015.
[36] I. Fe nández, A. A inda, I. Angulo, D. D. La Vega, N. U ibe-Pé ez, and A. Llano,
“Field T ials o he Empi ical Cha ac e iza ion o he Low Vol age G id Access
Impedance F om 35 kHz o 500 kHz,” IEEE Access, ol. 7, pp. 85786–85795, 2019,
doi: 10.1109/ACCESS.2019.2924253.
[37] P. A. C. Lopes, J. M. M. Pin o, and J. B. Ge ald, “Dealing Wi h Unknown Impedance
and Impulsi e Noise in he Powe -Line Communica ions Channel,” IEEE
T ansac ions on Powe Deli e y, ol. 28, no. 1, pp. 58–66, 2013, doi:
10.1109/TPWRD.2012.2214065.
[38] S. Rönnbe g, M. Lundma k, M. Wahlbe g, M. Ande sson, A. La sson, and M. Bollen,
“A enua ion and Noise Le el - Po en ial P oblems wi h Communica ion ia he
Powe G id,” in 19 h In e na ional Con e ence on Elec ici y Dis ibu ion, Vienna, May 2007.
[39] M. An oniali, A. Tonello, and F. Ve sola o, “A S udy on he Op imal Recei e
Impedance o SNR Maximiza ion in B oadband PLC,” Jou nal o Elec ical and
Compu e Enginee ing, ol. 2013, Ma . 2013, doi: 10.1155/2013/635086.
[40] S. Cassano, F. Sil es o, E. D. Jaege , and C. Le oi, “Modeling o ha monic
p opaga ion o as DC EV cha ging s a ion in a Low Vol age ne wo k,” in 2019
IEEE Milan Powe Tech, 2019, pp. 1–6. doi: 10.1109/PTC.2019.8810969.
[41] S. K. Ronnbe g, M. H. J. Bollen, A. La sson, and M. Lundma k, “An o e iew o he
o igin and p opaga ion o Sup aha monics (2-150 kHz),” 2014.
[42] J. González-Ramos, A. Galla e a, I. Angulo, I. Fe nández, A. A inda, and D. de la
Vega, “A e iew on he empi ical cha ac e iza ion o he low ol age dis ibu ion g id
as a communica ion channel o powe line communica ions,” Sus ainable Ene gy,
G ids and Ne wo ks, ol. 36, p. 101217, 2023, doi:
h ps://doi.o g/10.1016/j.segan.2023.101217.
[43] J. González-Ramos, A. Galla e a, I. Fe nández, I. Angulo, D. de la Vega, and A.
A inda, “Emissions gene a ed by elec ic ehicles in he 9-500 kHz band:
Cha ac e iza ion, p opaga ion, and in e ac ion,” Elec ic Powe Sys ems Resea ch, ol.
231, p. 110289, 2024, doi: h ps://doi.o g/10.1016/j.eps .2024.110289.
[44] J. González-Ramos, A. Galla e a, I. Fe nández, I. Angulo, D. de la Vega, and A.
A inda, “Compa ison o conduc ed emissions due o elec ic ehicle cha ging
p ocesses unde isola ed and on-line condi ions in he 9–500 kHz equency ange,”
Sus ainable Ene gy, G ids and Ne wo ks, ol. 38, p. 101333, 2024, doi:
h ps://doi.o g/10.1016/j.segan.2024.101333.
[45] J. González-Ramos, I. Angulo, I. Fe nández, A. A inda, and D. de la Vega,
“Cha ac e iza ion o he Po en ial E ec s o EMC Fil e s o Powe Con e e s on
Na owband Powe Line Communica ions,” Elec onics (Basel), ol. 10, no. 2, 2021,
doi: 10.3390/elec onics10020152.
85
[46] J. González–Ramos, I. Angulo, I. Fe nández, A. Galla e a, D. de la Vega, and A.
A inda, “Cha ac e iza ion o he po en ial e ec s o sub-cycle impedance a ia ions
on PRIME 1.4,” Enginee ing Science and Technology, an In e na ional Jou nal, ol. 56, p.
101775, 2024, doi: h ps://doi.o g/10.1016/j.jes ch.2024.101775.
[47] J. González-Ramos e al., “Upg ading he Powe G id Func ionali ies wi h
B oadband Powe Line Communica ions: Basis, Applica ions, Cu en T ends and
Challenges,” Senso s, ol. 22, no. 12, 2022, doi: 10.3390/s22124348.
[48] S. Galli, A. Scaglione, and Z. Wang, “Powe Line Communica ions and he Sma
G id,” in 2010 Fi s IEEE In e na ional Con e ence on Sma G id Communica ions, 2010,
pp. 303–308. doi: 10.1109/SMARTGRID.2010.5622060.
[49] A. R. Ndjiongue and H. Fe ei a, “Powe Line Communica ions (PLC) Technology:
Mo e Than 20 Yea s o In ense Resea ch,” T ansac ions on Eme ging Telecommunica ions
Technologies, pp. 1–20, Jul. 2019, doi: 10.1002/e .3575.
[50] G. López e al., “The Role o Powe Line Communica ions in he Sma G id
Re isi ed: Applica ions, Challenges, and Resea ch Ini ia i es,” IEEE Access, ol. 7,
pp. 117346–117368, 2019, doi: 10.1109/ACCESS.2019.2928391.
[51] L. Be ge , A. Schwage , and J. J. Escude o-Ga zas, “Powe Line Communica ions
o Sma G id Applica ions,” Jou nal o Elec ical and Compu e Enginee ing, ol. 2013,
Jan. 2013, doi: 10.1155/2013/712376.
[52] P. Chadwick, “CENELEC s anda ds o occupa onal EMF exposu e.” Accessed:
Sep. 18, 2023. [Online]. A ailable:
h ps://www.icni p.o g/cms/upload/p esen a ions/Join /ChadwickP.pd
[53] C. Cano, A. Pi olo, D. Malone, L. Lampe, A. M. Tonello, and A. G. Dabak, “S a e
o he A in Powe Line Communica ions: F om he Applica ions o he Medium,”
IEEE Jou nal on Selec ed A eas in Communica ions, ol. 34, no. 7, pp. 1935–1952, 2016,
doi: 10.1109/JSAC.2016.2566018.
[54] M. Wasowski e al., “Sou ces o Non-In en ional Sup aha monics in LV Ne wo k
and I s Impac on OSGP PLC Communica ion – Expe imen al S udy,” IEEE
T ansac ions on Powe Deli e y, ol. 37, no. 6, pp. 5244–5254, 2022, doi:
10.1109/TPWRD.2022.3175090.
[55] B. Masood and S. Baig, “S anda diza ion and deploymen scena io o nex gene a ion
NB-PLC echnologies,” Renewable and Sus ainable Ene gy Re iews, ol. 65, pp. 1033–
1047, 2016, doi: h ps://doi.o g/10.1016/j. se .2016.07.060.
[56] M. Sanz, J. I. Mo eno, G. López, J. Ma anza, and J. Be ocal, “Web-Based Toolki
o Pe o mance Simula ion and Analysis o Powe Line Communica ion Ne wo ks,”
Ene gies (Basel), ol. 14, no. 20, 2021, doi: 10.3390/en14206475.
[57] “PRIME Alliance, ‘PRIME 1.4 Whi e Pape ’, PRIME Alliance,” B ussels, 2014.
Accessed: Sep. 18, 2023. [Online]. A ailable: h ps://www.p ime-
alliance.o g/media/2020/04/whi ePape P imeV1p4_ inal.pd
86
[58] “ITU-T Rec. G9904; Na owband o hogonal equency di ision mul iplexing powe
line communica ion anscei e s o PRIME ne wo ks,” 2012.
[59] “G3 PLC Alliance, ‘Na owband OFDM PLC speci ica ions o G3 PLC ne wo ks’,
G3 PLC Alliance,” Pa is, 2021.
[60] “ITU-T Rec. G9903; Na owband o hogonal equency di ision mul iplexing powe
line communica ion anscei e s o G3 PLC ne wo ks,” 2012.
[61] “CLC/TS 50568 4:2015; Elec ici y me e ing da a exchange Pa 4: Lowe laye PLC
p o ile using SMITP B PSK modula ion,” 2015.
[62] “CLC/TS 50590:2015; Elec ici y me e ing da a exchange. Lowe laye PLC p o ile
using Adap a i e Mul i Ca ie Sp ead Spec um (AMC SS) modula ion,” 2015.
[63] “ISO/IEC 14908 1:2012; In o ma ion echnology – Con ol ne wo k p o ocol – Pa
1: P o ocol s ack,” 2012.
[64] “ITU T Rec. G9902; Na owband o hogonal equency di ision mul iplexing powe
line communica ion anscei e s o ITU T G,hnem ne wo ks,” 2012.
[65] “‘IEEE S anda d o Low-F equency (less han 500 kHz) Na owband Powe Line
Communica ions o Sma G id Applica ions,’ in IEEE S d 1901.2-2013, ol., no.,
pp.1-269, 6 Dec. 2013, doi: 10.1109/IEEESTD.2013.6679210.”
[66] “PRIME Alliance. Deploymen s & pilo s.” Accessed: Oc . 03, 2023. [Online].
A ailable: h ps://www.p ime-alliance.o g/alliance/deploymen s/
[67] M. Ca lsson, Sma Me e in No h Ame ica. Accessed: Oc . 03, 2023. [Online].
A ailable: h ps://media.be ginsigh .com/2022/12/23180016/bi-smna5-ps.pd
[68] L. Lampe, A. Tonello, and T. Swa , Powe line communica ions: P inciples, s anda ds and
applica ions om mul imedia o sma g id: Second edi ion. 2016. doi:
10.1002/9781118676684.
[69] S. Galli, A. Scaglione, and Z. Wang, “Fo he G id and Th ough he G id: The Role
o Powe Line Communica ions in he Sma G id,” P oceedings o he IEEE, ol. 99,
no. 6, pp. 998–1027, 2011, doi: 10.1109/JPROC.2011.2109670.
[70] J. G anado, A. To alba, and C. Ál a ez-A oyo, “Pa ial discha ge de ec ion using
PLC ecei e s in MV cables: A heo e ical amewo k,” Elec ic Powe Sys ems Resea ch,
ol. 164, pp. 61–69, 2018, doi: h ps://doi.o g/10.1016/j.eps .2018.07.030.
[71] “IEEE S anda d o B oadband o e Powe Line Ne wo ks: Medium Access Con ol
and Physical Laye Speci ica ions,” IEEE S d 1901-2020 (Re ision o IEEE S d 1901-
2010), pp. 1–1622, 2021, doi: 10.1109/IEEESTD.2021.9329263.
[72] “Open PLC Eu opean Resea ch Alliance o New Gene a ion PLC In eg a ed
Ne wo k Phase 2.” Accessed: Ap . 10, 2023. [Online]. A ailable: h ps://co dis.
eu opa.eu/p ojec /id/026920/es
[73] “ITU-T Rec. G9960; Uni ied High-Speed Wi e-Line Based Home Ne wo king
T anscei e s—Sys em A chi ec u e and Physcial Laye Speci ica ion. 2018. .”
87
Accessed: Ap . 10, 2023. [Online]. A ailable: h ps://www.i u.in / ec/T-REC-
G.9960-201811-I/en
[74] “HD-PLC Speci ica ions. 4 h Gene a ion HD-PLC Qua o Co e O e iew, HD-
PLC Alliance. .” Accessed: Ap . 10, 2023. [Online]. A ailable: h ps://hd-plc.o g/
wp-con en /uploads/2021/01/HDPD-P0010E_Whi e-Pape -HD-PLC4-6.pd
[75] “HomePlug AV Speci ica ion Ve sion 2.0, HomePlug Powe line Alliance. 2013. .”
Accessed: Ap . 10, 2023. [Online]. A ailable: h ps://web.a chi e.o g/web/20
121103131751/h p://www.homeplug.o g/ ech/whi epape s/HomePlug_AV2_W
hi e_Pape _ 1.0.pd
[76] “Segalo o, J.-F. Towa ds a T uly Sma G id—The Case o B oadband o e Powe
Line. IDC Whi e Pape , PRIME Alliance. ,” 2021. Accessed: Sep. 18, 2023. [Online].
A ailable: h ps://www.p ime-alliance.o g/media/2021/09/TOWARDS-A-
TRULY-SMART-GRID-Summa y.pd
[77] In e na ional Elec o echnical Vocabula y (IEV) - Pa 161: Elec omagne ic
Compa ibili y, “IEC 60050-161:1990.” Accessed: Oc . 04, 2023. [Online]. A ailable:
h ps://webs o e.iec.ch/publica ion/68718
[78] M. Kla e al., “Emission le els abo e 2 kHz — Labo a o y esul s and su ey
measu emen s in public low ol age g ids,” in 22nd In e na ional Con e ence and
Exhibi ion on Elec ici y Dis ibu ion (CIRED 2013), 2013, pp. 1–4. doi:
10.1049/cp.2013.1102.
[79] I. Fe nandez e al., “Cha ac e iza ion o non-in en ional emissions om dis ibu ed
ene gy esou ces up o 500 kHz: A case s udy in Spain,” In e na ional Jou nal o Elec ical
Powe & Ene gy Sys ems, ol. 105, pp. 549–563, 2019, doi:
h ps://doi.o g/10.1016/j.ijepes.2018.08.048.
[80] T. Yalcin, M. Özdemi , P. Kos yla, and Z. Leonowicz, “Analysis o sup a-ha monics
in sma g ids,” in 2017 IEEE In e na ional Con e ence on En i onmen and Elec ical
Enginee ing and 2017 IEEE Indus ial and Comme cial Powe Sys ems Eu ope (EEEIC /
I&CPS Eu ope), 2017, pp. 1–4. doi: 10.1109/EEEIC.2017.7977812.
[81] M. Bollen, M. Olo sson, A. La sson, S. Rönnbe g, and M. Lundma k, “S anda ds o
sup aha monics (2 o 150 kHz),” IEEE Elec omagn Compa Mag, ol. 3, no. 1, pp.
114–119, 2014, doi: 10.1109/MEMC.2014.6798813.
[82] E. O. A. La sson, M. H. J. Bollen, M. G. Wahlbe g, C. M. Lundma k, and S. K.
Rönnbe g, “Measu emen s o High-F equency (2–150 kHz) Dis o ion in Low-
Vol age Ne wo ks,” IEEE T ansac ions on Powe Deli e y, ol. 25, no. 3, pp. 1749–1757,
2010, doi: 10.1109/TPWRD.2010.2041371.
[83] A. Sendin, I. Be ganza, A. A zuaga, A. Pulkkinen, and I. H. Kim, “Pe o mance
esul s om 100,000+ PRIME sma me e s deploymen in Spain,” in 2012 IEEE
Thi d In e na ional Con e ence on Sma G id Communica ions (Sma G idComm), 2012, pp.
145–150. doi: 10.1109/Sma G idComm.2012.6485974.
[84] G. López, J. I. Mo eno, E. Sánchez, C. Ma ínez, and F. Ma ín, “Noise Sou ces,
E ec s and Coun e measu es in Na owband Powe -Line Communica ions
94
[153] A. A. M. Pico one, T. R. de Oli ei a, R. Sampaio-Ne o, M. Khos a y, and M. V
Ribei o, “Channel cha ac e iza ion o low ol age elec ic powe dis ibu ion
ne wo ks o PLC applica ions based on measu emen campaign,” In e na ional Jou nal
o Elec ical Powe & Ene gy Sys ems, ol. 116, p. 105554, 2020, doi:
h ps://doi.o g/10.1016/j.ijepes.2019.105554.
[154] M. Zimme mann and K. Dos e , “A mul ipa h model o he powe line channel,”
IEEE T ansac ions on Communica ions, ol. 50, no. 4, pp. 553–559, 2002, doi:
10.1109/26.996069.
[155] A. M. Tonello and F. Ve sola o, “Bo om-Up S a is ical PLC Channel Modeling—
Pa I: Random Topology Model and E icien T ans e Func ion Compu a ion,”
IEEE T ansac ions on Powe Deli e y, ol. 26, no. 2, pp. 891–898, 2011, doi:
10.1109/TPWRD.2010.2096518.
[156] A. M. Tonello and F. Ve sola o, “Bo om-Up S a is ical PLC Channel Modeling—
Pa II: In e ing he S a is ics,” IEEE T ansac ions on Powe Deli e y, ol. 25, no. 4, pp.
2356–2363, 2010, doi: 10.1109/TPWRD.2010.2053561.
[157] T. Esmailian, F. Kschischang, and P. Gulak, “In-building powe lines as high-speed
communica ion channels: Channel cha ac e iza ion and a es channel ensemble,” In .
J. Communica ion Sys ems, ol. 16, pp. 381–400, Jun. 2003, doi: 10.1002/dac.596.
[158] D. Anas asiadou and T. An onakopoulos, “Mul ipa h cha ac e iza ion o indoo
powe -line ne wo ks,” IEEE T ansac ions on Powe Deli e y, ol. 20, no. 1, pp. 90–99,
2005, doi: 10.1109/TPWRD.2004.832373.
[159] J. Ana o y, N. Thee hayi, and R. Tho appillil, “Powe -Line Communica ion Channel
Model o In e connec ed Ne wo ks—Pa I: Two-Conduc o Sys em,” IEEE
T ansac ions on Powe Deli e y, ol. 24, no. 1, pp. 118–123, 2009, doi:
10.1109/TPWRD.2008.2005679.
[160] T. Bos oen and O. Van de Wiel, “Modelling he low- ol age powe dis ibu ion
ne wo k in he equency band om 0.5 MHz o 30 MHz o b oadband powe line
communica ions (PLC),” in 2000 In e na ional Zu ich Semina on B oadband
Communica ions. Accessing, T ansmission, Ne wo king. P oceedings (Ca . No.00TH8475),
2000, pp. 171–178. doi: 10.1109/IZSBC.2000.829248.
[161] F. J. Cane e, J. A. Co és, L. Díez, and J. T. En ambasaguas, “A channel model
p oposal o indoo powe line communica ions,” IEEE Communica ions Magazine,
ol. 49, no. 12, pp. 166–174, 2011, doi: 10.1109/MCOM.2011.6094022.
[162] S. Galli and T. C. Banwell, “A de e minis ic equency-domain model o he indoo
powe line ans e unc ion,” IEEE Jou nal on Selec ed A eas in Communica ions, ol.
24, no. 7, pp. 1304–1316, 2006, doi: 10.1109/JSAC.2006.874428.
[163] F. Ve sola o and A. M. Tonello, “An MTL Theo y App oach o he Simula ion o
MIMO Powe -Line Communica ion Channels,” IEEE T ansac ions on Powe Deli e y,
ol. 26, no. 3, pp. 1710–1717, 2011, doi: 10.1109/TPWRD.2011.2126608.
[164] J. Ana o y, N. Thee hayi, and R. Tho appillil, “Powe -Line Communica ion Channel
Model o In e connec ed Ne wo ks—Pa II: Mul iconduc o Sys em,” IEEE
95
T ansac ions on Powe Deli e y, ol. 24, no. 1, pp. 124–128, 2009, doi:
10.1109/TPWRD.2008.2005681.
[165] T. Sa enae and P. Delogne, “De e minis ic modeling o he (shielded) ou doo
powe line channel based on he mul iconduc o ansmission line equa ions,” IEEE
Jou nal on Selec ed A eas in Communica ions, ol. 24, no. 7, pp. 1277–1291, 2006, doi:
10.1109/JSAC.2006.874423.
[166] H. Meng e al., “Modeling o ans e Cha ac e is ics o he b oadband powe line
communica ion channel,” IEEE T ansac ions on Powe Deli e y, ol. 19, no. 3, pp.
1057–1064, 2004, doi: 10.1109/TPWRD.2004.824430.
[167] E. G. Bakhoum, “${ m S}$-Pa ame e s Model o Da a Communica ions O e 3-
Phase T ansmission Lines,” IEEE T ans Sma G id, ol. 2, no. 4, pp. 615–623, 2011,
doi: 10.1109/TSG.2011.2168613.
[168] T. Banwell and S. Galli, “A no el app oach o he modeling o he indoo powe line
channel pa I: ci cui analysis and companion model,” IEEE T ansac ions on Powe
Deli e y, ol. 20, no. 2, pp. 655–663, 2005, doi: 10.1109/TPWRD.2005.844326.
[169] S. Galli and T. Banwell, “A no el app oach o he modeling o he indoo powe line
channel-Pa II: ans e unc ion and i s p ope ies,” IEEE T ansac ions on Powe
Deli e y, ol. 20, no. 3, pp. 1869–1878, 2005, doi: 10.1109/TPWRD.2005.848732.
[170] R. Alaya and R. A ia, “Cha ac e iza ion o low ol age access ne wo k o
na owband powe line communica ions,” in 2017 25 h In e na ional Con e ence on
So wa e, Telecommunica ions and Compu e Ne wo ks (So COM), 2017, pp. 1–6. doi:
10.23919/SOFTCOM.2017.8115514.
[171] A.-I. Chiu a and N. Secă eanu, “Theo e ical pos ula ion o PLC channel model,”
Jou nal o Elec ical and Elec onics Enginee ing, ol. 2, May 2009.
[172] A. Laza opoulos, “Towa ds Modal In eg a ion o O e head and Unde g ound Low-
Vol age and Medium-Vol age Powe Line Communica ion Channels in he Sma
G id Landscape: Model Expansion, B oadband Signal T ansmission Cha ac e is ics,
and S a is ical Pe o mance Me ics (In i ed Pape ),” ISRN Signal P ocessing, ol.
2012, Oc . 2012, doi: 10.5402/2012/121628.
[173] B. Masood, G. Song, S. Baig, M. Rasheed, and J. Hou, “Measu emen s and
Cha ac e iza ion o Low and Medium Vol age Residen ial, Comme cial and
Indus ial NB-PLC Ne wo ks o AMI,” IET Gene a ion, T ansmission and Dis ibu ion,
ol. 14, Dec. 2020, doi: 10.1049/ie -g d.2020.1233.
[174] B. Masood, S. Guobing, R. A. Naq i, M. B. Rasheed, J. Hou, and A. U. Rehman,
“Measu emen s and channel modeling o low and medium ol age NB-PLC
ne wo ks o sma me e ing,” IET Gene a ion, T ansmission & Dis ibu ion, ol. 15, no.
2, pp. 321–338, Jan. 2021, doi: h ps://doi.o g/10.1049/g d2.12023.
[175] I. Fe nández e al., “Cha ac e iza ion o he equency-dependen ansmission losses
o he g id up o 500 kHz,” in 25 h In e na ional Con e ence on Elec ici y Dis ibu ion,
Mad id, Jun. 2019.
96
[176] T. Maenou and M. Ka ayama, “S udy on Signal A enua ion Cha ac e is ics in Powe
Line Communica ions,” in 2006 IEEE In e na ional Symposium on Powe Line
Communica ions and I s Applica ions, 2006, pp. 217–221. doi:
10.1109/ISPLC.2006.247464.
[177] D. Shao, Q. Wang, Y. Lu, Y. Shu, C. Lai, and K. Zhang, “Analyses and Modeling o
Powe Line Channel A enua ion Cha ac e is ics o Low Vol age Access Ne wo k
in China,” in 2014 IEEE 80 h Vehicula Technology Con e ence (VTC2014-Fall), 2014,
pp. 1–5. doi: 10.1109/VTCFall.2014.6965873.
[178] I. Tsi opoulos, P. Siskos, and P. Cap os, “The cos o echa ging in as uc u e o
elec ic ehicles in he EU in a clima e neu ali y con ex : Fac o s in luencing
in es men s in 2030 and 2050,” Appl Ene gy, ol. 322, p. 119446, 2022, doi:
h ps://doi.o g/10.1016/j.apene gy.2022.119446.
[179] J. Ma anza, S. Alexand es, and C. Rod iguez-Mo cillo, “Pe o mance e alua ion o
wo na owband PLC sys ems: PRIME and G3,” Compu S and In e aces, ol. 36, no.
1, pp. 198–208, 2013, doi: h ps://doi.o g/10.1016/j.csi.2013.05.001.
[180] K. Razazian, M. Uma i, A. Kamalizad, V. Logino , and M. Na id, “G3-PLC
speci ica ion o powe line communica ion: O e iew, sys em simula ion and ield
ial esul s,” in ISPLC2010, 2010, pp. 313–318. doi: 10.1109/ISPLC.2010.5479881.
[181] A. Sanz, D. Sancho, C. Guemes, and J. A. Co és, “A physical laye model o G3-
PLC ne wo ks simula ion,” in 2017 IEEE In e na ional Symposium on Powe Line
Communica ions and i s Applica ions (ISPLC), 2017, pp. 1–6. doi:
10.1109/ISPLC.2017.7897116.
[182] A. Van Lae e, C. Waw zyniak, S. Be e, and V. Moeyae , “De elopmen , alida ion
and u iliza ion o an ITU-T G.9903 PHY simula o o communica ion pe o mance
e alua ion,” in 2016 In e na ional Symposium on Powe Line Communica ions and i s
Applica ions (ISPLC), 2016, pp. 167–172. doi: 10.1109/ISPLC.2016.7476276.
[183] B. Reynde s, E. Van Lil, S. Pollin, and S. an den Be gh, “Compa ison o he
di e en ial mode agains he cohe en mode in G3-PLC,” 2014. [Online]. A ailable:
h ps://api.seman icschola .o g/Co pusID:57866622
[184] K. Razazian, M. Uma i, and A. Kamalizad, “E o co ec ion mechanism in he new
G3-PLC speci ica ion o powe line communica ion,” in ISPLC2010, 2010, pp. 50–
55. doi: 10.1109/ISPLC.2010.5479944.
[185] L. Di Be , S. D’Alessand o, and A. M. Tonello, “A G3-PLC simula o o access
ne wo ks,” in 18 h IEEE In e na ional Symposium on Powe Line Communica ions and I s
Applica ions, 2014, pp. 99–104. doi: 10.1109/ISPLC.2014.6812329.
[186] A. A aye o, A. Ala ishe, and Y. I ano , “Powe line communica ion echnologies:
Modeling and simula ion o PRIME physical laye ,” ol. 2, pp. 931–936, Oc . 2012.
[187] P. Chebo aye , D. U azaye , I. Pospelo a, A. Ka assenko, D. Zyko , and A.
Shelupano , “Analysis o he le el o SNR in he PLC channel based on he esul s
o ma hema ical and physical modelling,” J Phys Con Se , ol. 1145, no. 1, p. 12013,
Jan. 2019, doi: 10.1088/1742-6596/1145/1/012013.
97
[188] K. Razazian, A. Kamalizad, M. Uma i, Q. Qu, V. Logino , and M. Na id, “G3-PLC
ield ials in U.S. dis ibu ion g id: Ini ial esul s and equi emen s,” in 2011 IEEE
In e na ional Symposium on Powe Line Communica ions and I s Applica ions, 2011, pp. 153–
158. doi: 10.1109/ISPLC.2011.5764382.
[189] A. E. Dulay, R. I. R. As ille o, P. G. P. La Rosa, R. J. L. L. Tan, A. D. Tapang, and
A. San os, “Pe o mance e alua ion o G3 na owband PLC s anda d o
ansmission h ough single and pai ed ans o me ,” in 2016 3 d MEC In e na ional
Con e ence on Big Da a and Sma Ci y (ICBDSC), 2016, pp. 1–6. doi:
10.1109/ICBDSC.2016.7460333.
[190] C. Males, V. Popa, A. La ic, and I. Finis, “Pe o mance e alua ion o Powe Line
Communica ions o e powe ans o me s,” in 2012 20 h Telecommunica ions Fo um
(TELFOR), 2012, pp. 627–630. doi: 10.1109/TELFOR.2012.6419288.
[191] S. Us un E can, “Powe line Communica ion: Re olu ionizing da a ans e o e
elec ical dis ibu ion ne wo ks,” Enginee ing Science and Technology, an In e na ional
Jou nal, ol. 52, p. 101680, 2024, doi: h ps://doi.o g/10.1016/j.jes ch.2024.101680.
[192] V. Blazek e al., “E o Analysis o Na owband Powe -Line Communica ion in he
O -G id Elec ical Sys em,” Senso s, ol. 22, no. 6, 2022, doi: 10.3390/s22062265.
[193] M. A. Wibisono, N. Moonen, and F. Le e ink, “In e e ence o LED Lamps on
Na owband Powe Line Communica ion,” in 2020 IEEE In e na ional Symposium on
Elec omagne ic Compa ibili y & Signal/Powe In eg i y (EMCSI), 2020, pp. 219–221. doi:
10.1109/EMCSI38923.2020.9191485.
[194] A. H. Beshi e al., “E ec s o he Swi ching F equency o Random Modula ed Powe
Con e e on he G3 Powe Line Communica ion Sys em,” in 2022 In e na ional
Symposium on Elec omagne ic Compa ibili y – EMC Eu ope, 2022, pp. 778–782. doi:
10.1109/EMCEu ope51680.2022.9901074.
[195] W. El Sayed, P. C o e i, N. Moonen, P. Lezynski, R. Smolenski, and F. Le e ink,
“Elec omagne ic In e e ence o Sp ead-Spec um Modula ed Powe Con e e s in
G3-PLC Powe Line Communica ion Sys ems,” IEEE Le e s on Elec omagne ic
Compa ibili y P ac ice and Applica ions, ol. 3, no. 4, pp. 118–122, 2021, doi:
10.1109/LEMCPA.2021.3121091.
[196] W. El Sayed, H. Loschi, M. A. Wibisono, N. Moone , P. Lezynski, and R. Smolenski,
“The In luence o Sp ead-Spec um Modula ion on he G3-PLC Pe o mance,” in
2021 Asia-Paci ic In e na ional Symposium on Elec omagne ic Compa ibili y (APEMC), 2021,
pp. 1–4. doi: 10.1109/APEMC49932.2021.9597060.
[197] W. Elsayed, P. Lezynski, R. Smolenski, A. Madi, M. Paze a, and A. Kempski,
“De e minis ic s. Random Modula ed In e e ence on G3 Powe Line
Communica ion,” Ene gies (Basel), ol. 14, Jun. 2021, doi: 10.3390/en14113257.
[198] L. Da Rocha Fa ias, L. F. Mon ei o, M. O. Leme, and S. L. S e an, “Empi ical
Analysis o he Communica ion in Indus ial En i onmen Based on G3-Powe Line
Communica ion and In luences om Elec ical G id,” Elec onics (Basel), ol. 7, no. 9,
2018, doi: 10.3390/elec onics7090194.
98
[199] A. Sendin, I. Be ganza, A. A zuaga, A. Pulkkinen, and I. H. Kim, “Pe o mance
esul s om 100,000+ PRIME sma me e s deploymen in Spain,” in 2012 IEEE
Thi d In e na ional Con e ence on Sma G id Communica ions (Sma G idComm), 2012, pp.
145–150. doi: 10.1109/Sma G idComm.2012.6485974.
[200] A. Sendin, I. Be ganza, A. A zuaga, X. Oso io, I. U u ia, and P. Anguei a,
“Enhanced Ope a ion o Elec ici y Dis ibu ion G ids Th ough Sma Me e ing
PLC Ne wo k Moni o ing, Analysis and G id Condi ioning,” Ene gies (Basel), ol. 6,
no. 1, pp. 539–556, 2013, doi: 10.3390/en6010539.
[201] I. A echalde, M. Cas o, I. Ga cía-Bo egue o, A. Sendín, I. U u ia, and A.
Fe nandez, “Pe o mance o PLC communica ions in equency bands om 150
kHz o 500 kHz,” in 2017 IEEE In e na ional Symposium on Powe Line Communica ions
and i s Applica ions (ISPLC), 2017, pp. 1–5. doi: 10.1109/ISPLC.2017.7897123.
[202] A. Mengi, M. Waech e , and M. Koch, “500 kHz G3-PLC access echnology o he
oll-ou s in Ge many,” in 18 h IEEE In e na ional Symposium on Powe Line
Communica ions and I s Applica ions, 2014, pp. 179–183. doi:
10.1109/ISPLC.2014.6812358.
[203] P. S. Sausen, A. Sausen, M. De Campos, L. F. Sau hie , A. C. Oli ei a, and R. R.
Emmel Júnio , “Powe Line Communica ion Applied in a Typical B azilian U ban
Powe Ne wo k,” IEEE Access, ol. 9, pp. 72844–72856, 2021, doi:
10.1109/ACCESS.2021.3078697.
[204] A. A uzuaga, I. Be ganza, A. Sendin, M. Sha ma, and B. Va ada ajan, “PRIME
In e ope abili y Tes s and Resul s om Field,” in 2010 Fi s IEEE In e na ional
Con e ence on Sma G id Communica ions, 2010, pp. 126–130. doi:
10.1109/SMARTGRID.2010.5622029.
[205] P. Mlynek, M. Kou ny, J. Misu ec, and Z. Kolka, “Measu emen s and e alua ion o
PLC modem wi h G3 and PRIME s anda ds o S ee Ligh ing Con ol,” in 18 h
IEEE In e na ional Symposium on Powe Line Communica ions and I s Applica ions, 2014,
pp. 238–243. doi: 10.1109/ISPLC.2014.6812318.
[206] J. Slacik, P. Mlynek, R. Fujdiak, J. Misu ec, and P. Silha y, “Pe o mance e alua ion
o mul i-ca ie powe line communica ion sys ems in eal condi ion o sma g id
neighbo hood a ea ne wo ks,” in 2017 9 h In e na ional Cong ess on Ul a Mode n
Telecommunica ions and Con ol Sys ems and Wo kshops (ICUMT), 2017, pp. 391–397. doi:
10.1109/ICUMT.2017.8255157.
[207] N. U ibe-Pé ez, I. Angulo, D. de la Vega, T. A zuaga, A. A inda, and I. Fe nández,
“On- ield e alua ion o he pe o mance o IP-based da a ansmission o e
na owband PLC o sma g id applica ions,” In e na ional Jou nal o Elec ical Powe &
Ene gy Sys ems, ol. 100, pp. 350–364, 2018, doi:
h ps://doi.o g/10.1016/j.ijepes.2018.02.030.
[208] I. Be ganza, A. Sendin, R. Ayala, and J. S. Gomez, “Low ol age as he inal on ie
o b oadband o e powe line,” in 27 h In e na ional Con e ence on Elec ici y Dis ibu ion
(CIRED 2023), 2023, pp. 3804–3810. doi: 10.1049/icp.2023.0717.
99
[209] A. Sendin, P. Losada-Sanisid o, J. Ga cía-Rod íguez, P. González-Méndez, and I.
Be ganza, “B oadband O e Powe Line Communica ion P o o ype De elopmen
o Nex Gene a ion Sma Me e s: Valida ion in Access Elec ic Powe Dis ibu ion
Ne wo ks,” IEEE Access, ol. 12, pp. 30191–30208, 2024, doi:
10.1109/ACCESS.2024.3367987.
[210] “ETSI Technical Commi ee. Powe Line Telecommunica ions (PLT) Na ow Band
T anscei e s in he Range 9 kHz o500 kHz Powe Line Pe o mance Tes Me hod
Guide ETSI Technical Commi ee Powe line Communica ions
TechnicalSpeci ica ion; ETSI TS 103 909 V1.1.1 ,” Sophia An ipolis, Dec. 2012.
[211] “In e na ional Elec o echnical Commission, ‘IEC 61557-3:2007. Elec ical sa e y in
low ol age dis ibu ion sys ems up o 1000 Va.c. and 1500 Vd.c. - Equipmen o
es ing, measu ing o moni o ing o p o ec i e measu es - Pa 3: Loop impedance,’”
2007.
[212] “In e na ional Elec o echnical Commission, ‘IEC 61000-4-19:2015.
Elec omagne ic compa ibili y (EMC). Pa 4-19: Tes ing and measu emen
echniques: Tes o immuni y o conduc ed, di e en ial mode dis u bances and
signaling in he equency ange 2 kHz o 150 kHz a a.c. powe po s,’” 2015.
[213] I. Fe nández e al., “Measu emen Sys em o he Mean and Sub-cycle LV G id Access
Impedance om 20 kHz o 10 MHz,” IEEE T ansac ions on Powe Deli e y, pp. 1–9,
2023, doi: 10.1109/TPWRD.2023.3238647.
[214] I. Fe nández e al., “Compa ison o Measu emen Me hods o LV G id Access
Impedance in he F equency Range Assigned o Nb-Plc Technologies,” Elec onics
(Basel), ol. 8, no. 10, 2019, doi: 10.3390/elec onics8101155.
[215] “In e na ional Elec o echnical Commission, ‘IEC 61851-1:2017. Elec ic ehicle
conduc i e cha ging sys em - Pa 1: Gene al equi emen s’, ,” 2017.
[216] HAMEG Ins umen s. A Rohde & Schwa z Company, “HM6050-2 Line Impedance
S abiliza ion Ne wo k. Technical Da a/Inse ion Loss.” Accessed: Jun. 26, 2023.
[Online]. A ailable: h ps://scdn. ohde-
schwa z.com/u /pws/dl_downloads/dl_common_lib a y/dl_manuals/gb_1/h/h
m6050_2/HM6050-2_Use Manual_de_en_04.pd
[217] “Polylux PD4000 speci ica ions.” Accessed: Jan. 22, 2024. [Online]. A ailable:
h ps://polylux.com/images/ iles/es/pd4000.pd
[218] “Mic ochip PL360G55CF E alua ion Boa d.” Accessed: Jun. 26, 2023. [Online].
A ailable: h ps://www.mic ochip.com/en-us/de elopmen ool/ PL360G55CF-
EK#
[219] E us Resea ch, “USRP N200/N210 Ne wo ked Se ies.” Accessed: Jun. 26, 2023.
[Online]. A ailable: h ps://www.e us.com/wp-
con en /uploads/2019/01/07495_E us_N200-210_DS_Flye _HR_1.pd
[220] J. González-Ramos e al., “Cha ac e iza ion o he LV dis ibu ion g id o he
deploymen o a pilo BB-PLC ne wo k,” in 2023 IEEE In e na ional Symposium on
100
Powe Line Communica ions and i s Applica ions (ISPLC), 2023, pp. 19–24. doi:
10.1109/ISPLC57122.2023.10104185.
[221] D. Chak a o y, J. Meye , P. Schegne , S. Yanchenko, and M. Schocke, “Impac o
Mode n Elec onic Equipmen on he Assessmen o Ne wo k Ha monic
Impedance,” IEEE T ans Sma G id, ol. 8, no. 1, pp. 382–390, 2017, doi:
10.1109/TSG.2016.2587120.
[222] A. Id is, M. Sab y, N. I. Abdul Razak, and A. L. Yuso , “Fas ime- a ying channels
in MIMO-OFDM sys em using di e en di e si y echnique,” in 2013 IEEE
Symposium on Wi eless Technology & Applica ions (ISWTA), 2013, pp. 138–141. doi:
10.1109/ISWTA.2013.6688756.
[223] J. Mon albán, M. M. Velez, G. P ie o, I. Eizmendi, G. Be jon-E iz, and J. L. O diales,
“On app oaching o gene ic channel equaliza ion echniques o OFDM based
sys ems in ime a ian channels,” in 2011 IEEE In e na ional Symposium on B oadband
Mul imedia Sys ems and B oadcas ing (BMSB), 2011, pp. 1–6. doi:
10.1109/BMSB.2011.5954940.
[224] J. González- amos, I. Fe nández, I. Angulo, A. Galla e a, D. D. La Vega, and A.
A inda, “Empi ical cha ac e iza ion o he conduc ed dis u bances gene a ed by he
elec ic ehicles du ing he cha ging p ocess,” in CIRED Po o Wo kshop 2022: E-
mobili y and powe dis ibu ion sys ems, 2022, pp. 319–323. doi: 10.1049/icp.2022.0719.
[225] J. González-Ramos, I. Angulo, I. Fe nández, A. Galla e a, A. A inda, and D. de la
Vega, “In luence o Elec ic Vehicle Cha ging on he G id Access Impedance om
20 kHz o 500 kHz,” in 2023 In e na ional Con e ence on Sma Ene gy Sys ems and
Technologies (SEST), 2023, pp. 1–6. doi: 10.1109/SEST57387.2023.10257356.
[226] J. González-Ramos e al., “In luence o he Spec al Pa e n o he Conduc ed
Emissions Gene a ed by Elec ic Vehicle Cha ging on PRIME 1.4,” in 2024 IEEE
In e na ional Con e ence on Communica ions, Con ol, and Compu ing Technologies o Sma
G ids (Sma G idComm), 2024, pp. 167–173. doi:
10.1109/Sma G idComm60555.2024.10738085.
101
Chap e 2
Conclusions and Fu u e Wo k
102
1. Conclusions
This Doc o al Thesis has ocused, i s , on cha ac e izing he elec ical g id as a
ansmission medium in e ms o he g id access impedance, channel esponse, and NIEs
co e ing he 9-500 kHz equency band. Second, NB-PLC echnologies unde di e en
channel condi ions ha e been e alua ed by means o labo a o y measu emen s.
Cha ac e iza ion o he elec ical g id
NIEs
Cha ac e iza ion o he NIEs gene a ed by EVs du ing hei cha ging p ocess. A no el me hodology
o he e alua ion o any conduc ed emission in he LV g id in he equency and ime
domains has been p oposed. The spec al analysis, which allows e alua ing no only he
ampli ude o he emissions, bu also he dis ibu ion o he emissions o e he whole
equency band, shows ha he emissions depend on he EV model, cha ging cu en , and
SoC. The ime analysis, in u n, e eals ha , excep o he onal emissions whose cen al
equency oscilla es wi h ime, he NIEs gene a ed by EVCPs ollow a sub-cycle pe iodic
pa e n (20 ms) in he equency band o in e es . In he speci ic case o he onal emissions
wi h a ime-oscilla ing cen al equency, a ime-dependen beha io wi hin 24.4 ms is
epo ed.
P opaga ion and in e ac ion o he emissions gene a ed by EVCPs. Conce ning he p opaga ion o
he p e iously cha ac e ized emissions, his Doc o al Thesis has p o en ha hey
p opaga e se e al me e s h ough he LV g id. Al hough in mos cases he dis u bances
a e a enua ed wi h dis ance, he e migh be esonances ha lead o highe ampli udes a
an elec ical poin dis an om he sou ce o he emissions. Mo eo e , he in luence o he
simul aneous cha ging o se e al EVs has been s udied, showing ha , in gene al, he
ampli udes co espond o he supe posi ion o he indi idual emissions, in addi ion o
in e modula ion p oduc s due o he swi ching equencies o he in e e s.
Compa ison o he emissions gene a ed by EVPCs using a LISN (isola ed condi ions) and when
measu ing di ec ly in he LV g id (on-line condi ions). The esul s show ha he emissions
measu ed unde on-line condi ions gene ally exceed he ones measu ed using a LISN, e en
hough he highes -ampli ude onal emissions a e se e al dB lowe o on-line condi ions.
The s udy also e eals ha he high-ampli ude emissions a e concen a ed in a ew
equency bins a equencies up o 150 kHz when conside ing isola ed condi ions. Unde
on-line condi ions, in u n, high-ampli ude emissions exceeding he PLC ou -o -band
emission limi s a e epo ed in he whole 9-500 kHz equency band o ce ain EVCPs.
The e o e, since he equency and ime cha ac e is ics o he emissions depend on he
measu emen condi ions, his cha ac e iza ion canno be only based on ials conduc ed
using a LISN.
G id impedance
Cha ac e iza ion o he mean g id impedance in he p esence o EVCPs. This Doc o al Thesis has
e ealed ha he measu ed impedances, ega dless o whe he an EV is cha ging in he
ne wo k o no , a e conside ably lowe han he e e ence impedances de ined in he
CISPR 16-1-2 s anda d. Besides, he in luence on he g id impedance o he dis ance
be ween he measu emen poin and he POC whe e he EVCSs a e ins alled has been
103
e alua ed, concluding, i s , ha he highes impedance a ia ions do no necessa ily occu
a he elec ical poin whe e he EV is connec ed; and, second, ha he in luence o a
ce ain EVCP depends no only on he EVCP i sel , bu also on he impedance in he
de aul si ua ion (when no EV is connec ed).
Cha ac e iza ion o he sub-cycle impedance a ia ions in he p esence o EVCPs. This Doc o al Thesis
has also epo ed a ime-dependen beha io o he g id impedance wi hin he undamen al
pe iod o he mains (20 ms) a speci ic equencies bo h in he absence o EVs and when
each EV is cha ging.
Cha ac e iza ion o he long- e m impedance a ia ions in he p esence o EVCPs. This Doc o al
Thesis has also deal wi h he long- e m a ia ions (some hou s) o he g id impedance,
concluding ha bo h he ampli ude and he spec al cha ac e is ics a e modi ied du ing he
cha ging p ocess o an EV. Speci ically, di e en impedance s a es (di e en spec al
pa e ns and ampli udes) ha e been epo ed o all he EVCPs unde analysis.
Addi ionally, a ia ions wi hin an impedance s a e ha e been also egis e ed, which show
signi ican di e ences in ampli ude and phase a he high-ampli ude esonances occu ing
a speci ic equencies.
E alua ion o NB-PLC unde di e en channel condi ions
NIEs
In luence o he spec al pa e n o he emissions gene a ed by EVCPs on PRIME 1.4. The esul s
show ha he ou modula ions used in his s udy (DQPSK_C, DBPSK_C, R_DQPSK,
and R_DBPSK) a e gene ally capable o co ec ing onal o na owband emissions
occu ing a ce ain equencies (in all cases i modula ions wi h epe i ion codes a e used;
o some EVCPs, also when modula ions including FEC a e used), since hey only a ec
a limi ed numbe o subca ie s. AWGN, in u n, a ec s all subca ie s equally and is
c i ical o OFDM communica ions. Conside ing ha he emissions show high-ampli ude
onal and na owband emissions in channel 1 and a spec al pa e n simila o AWGN in
channels 3-8, his Doc o al Thesis leads o conclude ha he spec al pa e n o he
emissions in channels 3-8 is mo e c i ical o communica ions han in channel 1 in e ms
o FER-SNR cu es. Thus, he esul s p esen ed in his documen p o ide e idence ha
i is necessa y o NB-PLC de ices o use modula ions wi h FEC (wi h epe i ion codes i
possible), in o de o a oid he nega i e e ec s o he conduc ed emissions gene a ed by
EVs du ing hei cha ging p ocess.
G id impedance
In luence o he equency-dependen impedance a ia ions on PRIME 1.4. The esul s e eal ha
he highes signal a enua ion is epo ed in hose equency channels in which he modulus
o he impedance is low and main ained wi hin he equency channel (low s anda d
de ia ion) and in which he phase p esen s ab up a ia ions. Rega ding he e alua ion o
he quali y o he communica ions, no ching e ec s in equency-dependen impedances
ha e been demons a ed o deg ade FER-SNR cu es.
In luence o he sub-cycle impedance a ia ions on PRIME 1.4. This Doc o al Thesis has also deal
wi h sub-cycle impedance a ia ions, concluding ha , due o he di e ence be ween he
impedance equency esponses in he ON and OFF s a es, a highe SNR is equi ed o
ob ain a FER o 5 %. Finally, i , in addi ion o he sub-cycle impedance a ia ions, he
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
3
e sion).
Al hough di e en BB-PLC echnologies ha e been de eloped o in-
home and MV channels, he e is s ill no BB-PLC sys em speci ically
designed o be deployed o e he LV dis ibu ion g id. Conside ing he
enhanced pe o mance wi h espec o NB-PLC in e ms o bandwid h,
la ency, and secu i y equi emen s, BB-PLC is an al e na i e ha is being
conside ed by some dis ibu ion sys em ope a o s (DSOs) o da a
ansmission h ough he LV dis ibu ion g id, in o de o ul ill he
demanding equi emen s o new SG applica ions [41].
As he beha io o he ou doo channel is expec ed o be conside ably
di e en om he cha ac e is ics o he indoo en i onmen , DSOs a e
aking in o conside a ion wo di e en op ions. Fi s , he adap a ion o
he exis ing indoo echnologies o he cha ac e is ics o he LV g id;
and, second, he de elopmen o a new s anda d based on he ou doo
ansmission medium. In any case, he de elopmen o BB-PLC ech-
nologies would de ini ely be a solu ion o co e ing he needs o u u e
SG applica ions, including he in eg a ion o dis ibu ed ene gy e-
sou ces o EV cha ging managemen , among o he s [7].
Wi h he aim o p o iding a isual o e iew o he e olu ion o PLC,
Fig. 1 shows he pas , p esen , and u u e o he echnology
de elopmen .
3. Main pa ame e s o be measu ed
In his pape , he mos ele an aspec s o be conside ed o he
empi ical cha ac e iza ion o he elec ical g id as a ansmission me-
dium a e add essed: NIEs, g id access impedance and a enua ion/
channel esponse. Th oughou he ex , he e m noise co esponds o an
elec omagne ic phenomenon no con eying in o ma ion and which is
combined wi h a wan ed signal [42]. Since he emissions e e ed o in
his pape a e always non-in en ional, emissions and NIEs a e used as
synonyms and a e de ined as unwan ed signals gene a ed by he powe
elec onics included in he ci cui y o connec ed elec onic de ices. The
e ms in e e ence and dis u bance, in u n, imply a deg ada ion o a
sys em. In he case o an in e e ence, he pe o mance o a communi-
ca ions sys em is jeopa dized, while a dis u bance is ela ed o he
deg ada ion o Powe Quali y (PQ), o he li e ime o a de ice, among
o he s. Finally, he e m dis o ion e e s o an undesi ed change in he
wa e o m o a signal ha migh lead o he appea ance o new equency
componen s.
3.1. Non-in en ional emissions
3.1.1. Backg ound
In he las yea s, he e has been a conside able inc ease in he
numbe o elec onic de ices connec ed o he elec ical g id ia powe
con e e s [43]. These de ices, as hey a e based on in e e s wi h
swi ching equencies abo e 10 kHz, ha e a signi ican impac on he
le els o he NIEs gene a ed in he equency ange o kHz [44–46], also
known as sup aha monics [47]. These high-ampli ude emissions, which
a e inhe en o he ope a ion o powe elec onics de ices [47], can
cause PQ issues, such as equipmen mal unc ion, alsi ica ion o ene -
gy/sma me e s, o o e ol age [48–53], in addi ion o deg ading he
quali y o PLC echnologies [54,55]. As he e m sup aha monics only
e e s o he dis u bances in he 2–150 kHz equency band, i is no
su icien o add ess EMC issues ha may a ec NB-PLC up o 500 kHz,
nei he BB-PLC in he MHz bands.
The ele ance o he e ec o he NIEs depends di ec ly on he
ampli ude, spec al o m, and ime- a ian beha io o he emission [7].
Wi h he aim o analyzing hei in luence on communica ions unde
labo a o y condi ions, he ETSI TS 103 909 [56] de ined a se o e e -
ence noises ep esen a i e o he LV g id. Howe e , due o he la ge
inc ease o Dis ibu ed Ene gy Resou ces (DERs) in ecen yea s, he
ime and spec al pa e ns o he dis u bances p esen in he g id a e
expec ed o ha e a ied conside ably and, hus, hei e ec on com-
munica ions is s ill unknown.
Fo his eason, he cha ac e iza ion o he dis u bances p esen in
he LV dis ibu ion g id, in bo h he ime and equency domains, e-
qui es u he s udy and becomes essen ial o he u he de elopmen
o PLC echnologies. To do so, i is necessa y o ca y ou ex ensi e ield
measu emen campaigns, in which di e en con igu a ions and g id
opologies a e conside ed, so ha he emissions o he g id a e co ec ly
cha ac e ized.
3.1.2. Classi ica ion o noise and main sou ces o NIEs
The noise o he elec ical g id can be classi ied as backg ound noise,
which a ies o e long pe iods o ime, and impulsi e noise, which
shows as ime- a ying beha io . The backg ound noise, in u n, can be
di ided in o colo ed noise, which is mainly caused by esiden ial elec-
onic equipmen , such as compu e s, dimme s, o hai d ye s [57], and
has a low powe spec al densi y, and na owband emissions caused by
b oadcas ing emissions. The connec ion o disconnec ion o elec onic
de ices can imply impulsi e noise ha is ape iodic wi h he mains e-
quency. In o he cases, due o he ec i ie s included in he powe sup-
plies, pe iodic impulsi e noise synch onous wi h he undamen al
equency can be ound in he g id, wi h epe i ion a es mul iple o
50/60 Hz. Asynch onous pe iodic impulsi e noise is also p esen in he
g id, showing epe i ion a es o 50–200 kHz [7,57–61]. In Fig. 2, a
summa y o he classi ica ion o he noise p esen in he LV g id is
shown. Some au ho s ha e s a is ically cha ac e ized he impulsi e
[62–65] and backg ound [66,67] noise in indoo channels in he e-
quency band assigned o PLC, which can se e as a basis o he
Fig. 1. E olu ion o PLC echnology: pas , p esen , and u u e.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
4
cha ac e iza ion o he noise in ou doo en i onmen s.
In compliance wi h [68–71], he noise o he elec ical g id can be
pa e ned as he sum o he p e iously men ioned ypes o noises. In
[72], i is s a ed ha he cyclos a iona y noise, caused by de ices
showing an impedance changing in he sho - e m [73], is p edominan
in he elec ical g id and has a signi ican in luence on NB-PLC, as i
educes conside ably he da a a e o he OFDM-based echnologies
[74].
Some ecen ly published a icles [75–79] show ha pho o ol aic
in e e s (PV), ba e y cha ge s, ene gy-e icien ligh ing, hyd opowe
sys ems, wind u bines, o EV cha ge s, among o he s, a e he main
sou ces o he high-ampli ude conduc ed emissions. In se e al ield and
labo a o y ials, a equency and ime cha ac e iza ion o hese emis-
sions has been ca ied ou , concluding ha , as hey occu a equencies
assigned o NB-PLC, he quali y o communica ions can be subs an ially
a ec ed and deg aded [30,54,80,81]. In some ins ances, as o he EV
cha ge s and PV panels, he sou ces o dis u bance a e loca ed nea he
sma me e , which migh imply an addi ional challenge o he co ec
pe o mance o NB-PLC [50]. O he non-in en ional emi ing de ices,
such as mo o s [76], ligh ing de ices [82–84], o elec onic ampli ie s
[47], in oduce high-ampli ude emissions wi h a wide ange o spec al
pa e ns. A signi ican ime-dependen beha io has also been obse ed
o he p e iously men ioned sou ces [76]. I should also be no ed ha
high-ampli ude impulsi e dis u bances a e gene a ed by his equipmen
when commu ing be ween di e en s a es o wo king egimes [76],
which may ha e an addi ional nega i e e ec on communica ions. In he
BB-PLC equency band, in u n, conduc ed emissions a e no expec ed
o be ha high. Fig. 3 shows he main sou ces o he emissions p esen in
he LV dis ibu ion g id, as well as he publica ions add essing hei
cha ac e iza ion.
3.1.3. No ma i e amewo k
3.1.3.1. Emission limi s. The e is g ea in e es om he s anda diza ion
and egula ion o ganiza ions in he cha ac e iza ion o NIEs in he e-
quency and ime domains o equencies abo e 2 kHz. Fo example, he
o me SC 205 Wo king G oup 11 o CENELEC, cu en ly TC219, is in
cha ge o s udying he dis u bances in he elec ical g id and de e -
mining he immuni y le els o communica ions [47]. I should be
men ioned ha his wo king g oup has ecen ly eques ed da a o in-
clusion in a epo abou he p opaga ion o conduc ed emissions up o
500 kHz. Mo eo e , h ough he wo king g oups TC77A and CISPR
SC/H, he In e na ional Elec o echnical Commission (IEC) de ines he
equi emen s o egula ing emissions so ha he compa ibili y o elec-
ical p oduc s in he equency ange up o 500 kHz is ensu ed.
The maximum ampli ude o he emissions gene a ed by ce ain
equipmen connec ed o he elec ical g id has al eady been speci ied by
he In e na ional Special Commi ee on Radio In e e ence CISPR.
CISPR15 (EN 55015) [82] de ines he limi s o he ligh ing equipmen ,
whe eas CISPR11 (EN 55011) [85] add esses he maximum ampli udes
Fig. 2. Classi ica ion o he noise o he LV dis ibu ion g id.
Fig. 3. Main sou ces o emissions in he LV g id and summa y o he publica ions add essing hei cha ac e iza ion.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
5
o induc ion cooking de ices. By con as , no speci ic limi s ha e been
speci ied o many o he sou ces o emissions, such as PVs, EVs, o hy-
d opowe sys ems. Fo hese de ices, he ou -o -band limi s de ined o
communica ions equipmen in EN 50065–1 [86] migh be conside ed as
a conse a i e c i e ion [57]. This echnical speci ica ion de ines limi s
adap ed o he equency band o ansmission (2–9 kHz, 9–150 kHz, o
150 kHz-30 MHz), and he ype o de ec o ha should be used in he
measu emen s, Roo Mean Squa e (RMS), Quasi Peak (QP) o A e age
[87–89]. These emission limi s a e de ined o labo a o y condi ions and
mus be e alua ed by means o a Line Impedance S abiliza ion Ne wo k
(LISN). A LISN is a s anda d load impedance allowing he epea abili y
and compa abili y o EMI measu emen s, which also p e en s he ials
om being a ec ed by ex e nal conduc ed emissions [90].
In he LV dis ibu ion g id, he Compa ibili y Le els (CLs) es ablish
he maximum ampli udes o he emissions ha canno be exceeded a a
speci ic elec ical poin , as he combina ion o he emissions gene a ed
by all he equipmen connec ed o he ne wo k. The g id ope a o should
ensu e ha a leas in 95% o he loca ions hese limi s a e no exceeded.
The Annex B o he IEC 61000–4–7 [91] speci ies he limi s o he
equency band om 2 kHz o 9 kHz in RMS alues, whe eas he IEC
61000–2–2 [92] de ines he limi s o he 9–150 kHz equency ange in
QP alues.
3.1.3.2. Measu emen se ups. Exis ing s anda ds speci y measu emen
se ups o he cha ac e iza ion o he emissions om equipmen unde
es . CISPR 16–1–2 [93] de ines a s anda dized measu emen se up o
he e alua ion o he emissions based on a LISN om 9 kHz o 109 MHz.
Simila ly, IEC 61000–4–7 [91] speci ies he use o an A i icial Mains
Ne wo k (AMN) below 9 kHz. In his way, a con olled and isola ed
scena io is se and he measu emen s a e no a ec ed by ex e nal NIEs
[94].
Howe e , he measu emen campaigns ca ied ou so a lead o
conclude ha he e e ence impedances de ined in [91,93] a e no a
good ep esen a ion o he ac ual g id access impedance alues o he LV
dis ibu ion g id [95]. Fo his eason, he cha ac e iza ion o he
emissions should no only be based on measu emen s using AMN o
LISN, whe e isola ed e ec s can be e alua ed, bu also on ield ials, so
ha ep esen a i e dis u bance alues a e ob ained.
In his con ex , some au ho s [96,97] ha e op ed o conduc ing in-
es iga ions ela ed wi h NIEs a econs uc ed acili ies. This ype o
scena ios a oid he ideal condi ions o labo a o y measu emen s as well
as he uncon olled g id ac o s p esen in on- ield ials, such as he
a ie y o loads connec ed o he g id, he g id opology, he g id
impedance, o he elec ical cables, among o he s [75,98].
3.1.3.3. Measu emen me hods. In he bibliog aphy, he e m measu e-
men me hod e e s o he pos -p ocessing applied o he eco ded sig-
nals o e alua e NIEs in he equency domain. Up o now, no no ma i e
measu emen me hod o he assessmen o he dis u bances in he LV
dis ibu ion g id has been de ined o equencies abo e 9 kHz. As p e-
iously men ioned, he compa ibili y le els in he 9–150 kHz equency
ange ha a e included in IEC 61000–2–2 [92] a e de ined o QP alues
[92]. Fo his eason, CISPR16 1–1 [87], a me hod based on a QP de-
ec o , is he me hod ha migh be used o he e alua ion o he dis-
u bances in g id measu emen s [99]. Ne e heless, his me hod
p esen s se e al d awbacks: i s , i is no in ended o g id measu e-
men s, bu a he o labo a o y condi ions; and, second, i has high
complexi y, compu a ional bu den, and memo y equi emen s.
Acco dingly, he s anda diza ion ins i u ion IEC SC77A/WG9 is
cu en ly wo king on he de elopmen o a no ma i e me hod o he
e alua ion o NIEs in he LV dis ibu ion g id. In his con ex , se e al
au ho s ha e p oposed new me hods and compa ed hei pe o mance
based on simple syn he ic signals o es signals eco ded in he LV g id
[100–104]. In Table 2, he main cha ac e is ics o he measu emen
me hods (in o ma i e o no ma i e) p oposed in he s anda ds o
ecen ly conside ed o i s inclusion in a s anda d a e ga he ed. In o de
o show he di e en esul s p o ided by each me hod, Fig. 4 p esen s
he spec um o a pa icula emission acco ding o IEC 61000–4–7 (max,
RMS), IEC 61000–4–30 (max, RMS), CISPR 16–1–1 (QP), and Ligh QP
(QP) me hods in he 2–150 kHz equency band.
Once he spec a a e ob ained, he e is no s anda dized p ocedu e o
he quan i a i e cha ac e iza ion o he emissions in he equency
domain. The li e a u e only conside s he To al Sup aha monic Vol age
(TSHV) [110–112], a pa ame e ha gi es an insigh in o he o al
ampli ude o he emissions in he equency band unde analysis.
Howe e , his pa ame e is highly dependen on he measu emen
me hod used o he pos -p ocessing o he co esponding emission
[102] and does no ake in o accoun he equency dis ibu ion o he
ampli ude o he emissions, which plays an essen ial ole in he p ope
design o PLC echnologies. Mo eo e , he concep s o na owband and
b oadband emission ha e no ye been de ined and, he e o e, he e is no
no malized p ocedu e o hei p ope cha ac e iza ion.
Table 2
Compa ison o he measu emen me hods o he assessmen o he conduc ed emissions in he LV g id.
Me hod F equency
band
Time window Window
leng h
O e lapping F equency s ep size Gaps Ou pu
me ics
CISPR 16–1–1[87] 9–150 kHz Gaussian/Kaise /Lanczos,
e c.
20 ms ≥75% ≤100 Hz No QP
150 kHz-30 MHz 0.5 ms ≤4.5 kHz
IEC 61000–4–7[91] 2–9 kHz Rec angula 200 ms 0% 200 Hz No RMS, max
IEC 61000–4–30[105] 9–150 kHz Rec angula 0.5 ms - 2 kHz Yes RMS, max
Subsampling app oach[106] ≤150 kHz Rec angula 5 ms 0% - No RMS, max
OMP comp essi e sensing[107] 2–150 kHz Rec angula 0.5 ms 0% 200 Hz No RMS, max
Bayesian comp essi e sensing
[108]
2–150 kHz Rec angula 0.5 ms 0% 200 Hz No RMS, max
Wa ele app oach[109] 2–150 kHz Rec angula 200 ms 0% 200 Hz No RMS, max
Ligh QP[99] 9–150 kHz Rec angula 20 ms 0% 100 Hz No QP, RMS, max
Fig. 4. Spec um o a pa icula emission gene a ed by he cha ging p ocess o
an EV acco ding o IEC 61000–4–7 (max, RMS), IEC 61000–4–30 (max, RMS),
CISPR-16–1–1 (QP), and Ligh QP (QP) me hods in he 2–150 kHz e-
quency band.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
6
Finally, i should be men ioned ha he measu emen me hods
p esen ed in his sec ion a e only in ended o cha ac e ize he emissions
in he equency domain, wi hou conside ing hei ime-dependen
beha io . Since hese ime a ia ions can nega i ely a ec PLC, he
de elopmen o me hods ha add ess he ime cha ac e iza ion in
conjunc ion wi h he equency cha ac e iza ion is necessa y, o which
he de ini ion o s anda dized me hodologies in he ime domain is o
upmos impo ance. A join ime- equency domain cha ac e iza ion o
he emissions in he equency ange 9–150 kHz can be ound in [113].
3.1.4. P opaga ion/in e ac ion o he emissions
Acco ding o he li e a u e, he dis u bances gene a ed by he de-
ices connec ed o he g id p opaga e h ough he LV ne wo k, and may
e en be ans e ed o he MV g id o e dis ances o se e al kilome e s
[97,114–116]. As a esul , hey may a ec he ope a ion o ene gy
me e s and PLC equipmen [117,118]. As indi idual de ices ha e a
g ea e impac on highe equencies han on ha monics [119], he
p edic ion o he dis u bances in his equency ange should conside
he whole ins alla ion and no only indi idual de ices [120].
In acco dance o [95,119,121–124], he dis u bances gene a ed by a
ce ain de ice can be classi ied as p ima y and seconda y emissions. The
p ima y emission co esponds o he emission o igina ed inside he de-
ice, whe eas he seconda y emission is gene a ed by o he elec onic
equipmen o he g id i sel , and p opaga ed in o he de ice. This
p opaga ion highly depends on he impedance o neighbo ing de ices in
ela ion o he impedance o he elec ical g id [121,125,126]. Reso-
nances a e a key aspec in he p opaga ion o he dis u bances, as hey
imply inc eases in he emission a he swi ching equency [95,121,
127]. In [119], i is s a ed ha a esonance esul s in an inc ease in he
seconda y emission, whe eas he p ima y emission is a enua ed.
Rega ding he in e ac ion, some wo ks ha e poin ed ou he exis-
ence o equency bea ing and in e modula ion e ec s. Fo ins ance, in
[45], se e al simula ions ca ied ou in MATLAB led o conclude ha he
cha ging o EVs o he same ype, due o sligh ly di e en swi ching
equencies (
1
,
1
’), implies an emission a |
1
-
1
’|. In gene al, he
bea ing equency is in he o de o some Hz and does no a ec PLC.
In e modula ion dis o ion in he equency band up o 100 kHz, in u n,
occu s due o he in e ac ion o conside ably di e en swi ching e-
quencies (
1
,
2
) and is in he o de o ens o kHz [45]. This e ec is also
epo ed in [97], whe e he in e ac ion be ween h ee Elec ic Vehicle
Cha ging P ocesses (EVCPs) in a con olled and isola ed LV g id is
analyzed in he equency ange up o 500 kHz. Mul iple onal emissions
no egis e ed when measu ing he dis u bances gene a ed by each
EVCP indi idually a e shown when connec ing se e al EVCPs simul a-
neously. The a icle concludes ha hese emissions a e linea combi-
na ions o he undamen al equencies o he in e e s (20 • 2− 1,
20 • 2+ 1, 22 • 2− 1…), i.e., in e modula ion p oduc s. Ano he
simila analysis conce ning in e modula ion p oduc s can be ound in
[128]. In his a icle, he in e ac ion be ween a PV panel and an EV is
epo ed, e idencing he exis ence o in e modula ion p oduc s o up o
4 h o de .
The in e ac ion be ween end-use and PLC equipmen has also been
s udied in he li e a u e. 5 ypes o in e ac ions be ween hese de ices
a e desc ibed in [129], concluding ha he pe o mance o communi-
ca ions is deg aded and he li e ime o he end-use equipmen is
conside ably educed.
3.1.5. Elec ic ehicles (EVs)
The high pene a ion o EVs in he ollowing yea s will signi ican ly
con ibu e o each he main objec i es o he SGs, allowing, among
o he s, he educ ion o cos s and en i onmen al impac s [81,130].
Howe e , he ising numbe o EVs also in ol es conside able chal-
lenges. Fo example, EV Cha ging S a ions (EVCSs) gene a e highe
ampli ude emissions han o he elec onic de ices connec ed o he g id
[131], as hey a e based on ecen ly de eloped echnologies, aimed a
minimizing he cha ging ime and inc easing he cha ging e iciency.
Fo his eason, he e is g ea in e es in he scien i ic communi y in
analyzing and cha ac e izing he dis u bances gene a ed by EVs du ing
he cha ging p ocess.
Since he manu ac u e s o he EVCSs do no publicly sha e he de-
ails o he implemen ed con e e s, he heo e ical analysis o he dis-
u bances is no gene ally possible. Fo his eason, as s a ed in [132],
he cha ac e iza ion o hese emissions can only be ca ied ou empi i-
cally, by means o ex ensi e measu emen campaigns.
In ecen yea s, some labo a o y and ield ials ha e been de eloped
o he analysis o he dis u bances gene a ed by se e al EVCPs. Fo
ins ance, in [96], he emissions in oduced by a Bi-Di ec ional Vehicle o
G id (V2G) EVCS a e s udied in a con olled g id scena io up o 150 kHz.
This wo k concludes ha na owband emissions a e p esen a he
swi ching equency and mul iples o i , caused by he Pulse Wid h
Modula ion (PWM), whe eas b oadband emissions occu a highe e-
quencies due o he DC-DC con e e when he Posi i e Tempe a u e
Coe icien hea e is ac i a ed. Field ials p esen ed in [133] show ha
he high ampli ude emissions gene a ed by an elec ic bus do no
dec ease wi h equency in he band 2–150 kHz and ha hey could be
as high as he CL de ined in [92]. An isola ed g id scena io, connec ed o
he public elec ici y g id, is also used in [45], whe e he p opaga ion
and in e ac ion o he NIEs o ou EVs a e analyzed up o 100 kHz. As
he g id allows swi ching o mic og id mode, measu emen s in bo h
con igu a ions ha e been conside ed, ob aining di e en esul s in some
cases. In [97], a p ocedu e o he equency and ime cha ac e iza ion
o he emissions in a con olled LV g id is p esen ed. The s udy, based on
he calcula ion o se e al pa ame e s in he equency domain and a FFT
analysis, e eals high-ampli ude emissions a ying wi hin he unda-
men al pe iod o he mains in he equency ange 9–500 kHz. Re e ence
[134] analyzes he long- e m sup aha monic emissions o h ee EVs in
he ime and equency domains up o 100 kHz a h ee pa king ga ages.
The p esen ed esul s lead o conclude ha he powe g id dis u bance
le els inc ease when he numbe o connec ed EVs ise. In [135], he
dis u bances gene a ed by di e en ypes o cha ge s a i e si es in
China and Ge many a e cha ac e ized in he equency domain. The
a icle shows he dominan emission equencies as well as hei ol age
ampli ude up o 50 kHz. A simila analysis is p esen ed in [136–138],
whe e he swi ching equencies o wel e, en, and eigh EVs, espec-
i ely, a e iden i ied. A huge a iabili y in he swi ching equency and
i s ampli ude can be obse ed in all h ee cases.
Howe e , in on-si e measu emen s emissions a e a ec ed by se e al
g id ac o s [139], including he ime and equency-dependen g id
access impedance. Thus, in some wo ks, such as [94], he EVCS is
di ec ly connec ed o a LISN in o de o isola e he se up om he LV
g id, a oiding ex e nal noise and ensu ing ha only he emissions
gene a ed by he EVCPs a e measu ed. In [140], a ime and equency
analysis o he dis u bances in such a con olled en i onmen is p e-
sen ed. Howe e , i only conside s wo EV models and does no p o ide
an in-dep h s udy o he ime- a ian beha io o he dis u bances.
3.2. Impedance and channel esponse
3.2.1. Backg ound
The g id access impedance is one o he main ac o s a ec ing PLC
sys ems. The ime [141] and equency a ia ions o he g id access
impedance [73,142] esul s, in u n, in o conside able changes in he
channel equency esponse (CFR). As he modulus o he CFR co e-
sponds o he losses o he ansmi ed signal, PLC can be g ea ly
a enua ed [143–146]. This high a enua ion can signi ican ly educe
he maximum ange o he communica ions [147] and is, gene ally, due
o sudden changes in loads [6]. Howe e , hese impedance a ia ions do
no only a ec he PLC signal bu also he NIEs p esen in he g id [95,
119]. Hence, he CFR may ha e a double e ec on he quali y o he
communica ions.
Addi ionally, as he e is no impedance ma ching be ween he
impedance o he communica ions equipmen and he g id access
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
7
impedance, he maximum powe ans e is no achie ed and, hus, he
ansmi ed signal migh be subs an ially a enua ed [148].
3.2.2. Main ac o s a ec ing impedance and channel esponse
Load a ia ions on he powe lines, day- ime ene gy o nigh - ime
ene gy demands, among o he s, a e he main causes o he ime-
dependen beha io o he impedance [149]. T adi ionally, only he
long- e m impedance a ia ions, which a e caused by he sudden
connec ion o disconnec ion o elec ic equipmen connec ed o he g id,
ha e been conside ed. Acco ding o he li e a u e, some measu emen
campaigns o he cha ac e iza ion o he g id access impedance, such as
[150] o [151], ha e been ca ied ou all a ound he wo ld, concluding
ha no signi ican long- e m a ia ions we e obse ed.
Ne e heless, in he las yea s, some wo ks highligh he exis ence o
sho - e m a ia ions, pe iodical and synch onous wi h he mains signal,
caused by ce ain elec onic equipmen [5,73,146]. No mally, hese
sub-cycle a ia ions a e due o he swi ched componen s (powe sup-
plies, AC/DC con e e s, and in e e s) included in he ci cui y o
ce ain elec onic equipmen , which esul in wo di e en ia ed
impedance s a es [5,73,152,153]. An example o he modulus and phase
o a ime- a ying impedance wi hin he 20 ms pe iod in 50 Hz g ids is
shown in Fig. 5. An analysis o he in luence o hese sub-cycle a ia ions
on NB-PLC PRIME can be ound in [152]. The s udy concludes ha hese
a ia ions, no only due o he di e ence be ween he wo impedance
s a es wi hin he cycle, bu also due o he ime ins an s in which he
a ia ions occu , ha e a signi ican nega i e e ec on he channel
esponse es ima ion and he channel equaliza ion p ocesses de ined by
NB-PLC echnologies. This occu s since he ecei e o he communi-
ca ion sys em uses he symbols o he heade o es ima e he channel
equency esponse and, when applying ha es ima ion o he payload,
he CFR has apidly changed due o he sub-cycle impedance a ia ions.
The e o e, he equaliza ion is pe o med by using an inco ec es ima-
ion o he channel. I should be men ioned ha , as s a ed in [152], he
leng h o he ansmi ed ame, as well as he modula ion used, ha e an
impac on he sys em pe o mance unde hese ci cums ances.
Acco ding o he li e a u e, he a enua ion measu ed a low e-
quencies, 50 Hz o 60 Hz, canno be ex apola ed o hose p esen ed in
he powe g id a equencies whe e PLC communica ions a e es ab-
lished [33]. As i is de ailed and analyzed in [154], a hese highe
equencies, high a enua ions can be due o di e en causes: g id o-
pology, mul ipa h, dis ance, equency o ype o elec ical cables. Ac-
co ding o [155] he e lec ion caused by he impedance misma ching
has a signi ican impac on he signal loss. In addi ion, he in luence o
dis ance and numbe o b anches in he a enua ion su e ed by he PLC
signal is s udied. The a icle concludes ha he g ea e he dis ance
be ween he communica ions equipmen and he numbe o b anches,
he highe he a enua ion measu ed. Simila esul s a e p esen ed in
[156], whe e he addi ional high losses gene a ed by he b anches in he
ee-like opologies a e highligh ed. Re e ence [57] poin s ou he
impac o dis ance in a enua ion, especially when ansmi ing a high
equencies. I is wo h men ioning ha [155] also conside s he s udy o
wo di e en lines wi h he same numbe o b anches and he same
dis ance be ween ansmi e and ecei e , ob aining di e en esul s o
a enua ion. I concludes ha e en hough, as men ioned be o e, he
ansmission losses a e di ec ly p opo ional o dis ance and numbe o
b anches, o he ac o s, such as he access impedance modi ied by he
di e en equipmen connec ed o he powe g id, can also educe he
le el o he ecei ed signal. This is also epo ed in [6], whe e i is
indica ed ha he highes a enua ions a e due o he swi ching ON/OFF
equipmen (sudden changes in loads) and he addi ion o new lines o
cos ume s. As an example, in Fig. 6, he a enua ion in he 1.7–10 MHz
equency ange om a Seconda y Subs a ion (SS) o a SM oom in he
LV g id in Spain is shown.
3.2.3. No ma i e amewo k
As he beha io o he g id access impedance in he equency band
assigned o PLC echnologies is s ill unknown [157], he egula o y
amewo k in his ield is s ill limi ed. IEC 61000–4–7 de ines a e e ence
g id impedance in he equency ange 2–9 kHz [91], bu some au ho s
ha e ex ended i s de ini ion up o 150 kHz [96,158,159] o e en up o
500 kHz [98]. Howe e , he measu emen campaign ca ied ou in he
LV g id in Aus ia, Swi ze land, Czech Republic, and Ge many demon-
s a es ha he e e ence g id impedance in IEC 61000–4–7
Fig. 5. Example o he modulus and phase o an impedance a ying wi hin he undamen al pe iod o he mains (20 ms) co esponding o he measu emen o he
cha ging o a mobile phone ca ied ou o e a LISN.
Fig. 6. Example o he a enua ion (dB) om a SS o a SM oom in he LV g id
in Spain.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
8
o e es ima es e en he highes alues o he impedances measu ed in
he ac ual dis ibu ion g id [158]. CISPR 16–1–2 desc ibes h ee
di e en e e ence impedances o de ining speci ic LISNs depending on
he equency ange (9–150 kHz, 150 kHz-30 MHz, o
150 kHz-108 MHz) [93]. Acco ding o [159], he CISPR 16–1–2 e e -
ence impedance de ined in he equency ange 9–150 kHz also exceeds
he alues o he impedances measu ed in indoo en i onmen s in
Ge many, Uni ed Kingdom, and Spain. In Fig. 7, he modulus and phase
o he e e ence impedances de ined in CISPR 16–1–2 and IEC
61000–4–7 a e p esen ed. As he IEC 61000–4–7 only de ines he
modulus o he impedance, Fig. 7(b) only includes he phase speci ied in
CISPR 16–1–2.
The empi ical cha ac e iza ion o he LV g id equi es measu emen
sys ems adap ed o he cha ac e is ics o he ne wo k. The lack o
a ailable comme cial de ices has led some au ho s o design hei own
sys ems wi h di e en cha ac e is ics and based on di e en ha dwa e/
so wa e. These sys ems should comply wi h he maximum ole able
measu emen unce ain y de ined in IEC 61557–3 [160] and IEC
61000–4–19:2014 [161]. Howe e , he absence o a s anda dized e-
quency- and ime-dependen e e ence impedance has p e en ed he
de ini ion o a no malized me hod o he accu acy assessmen o a
speci ic sys em, as well as he compa ison be ween di e en imple-
men a ion echniques. The i s app oach o de ining a e e ence
impedance is epo ed in he Eu opean P ojec Z-NET [162]. In his
p ojec , only he de ini ion o a s a ic impedance is conside ed, which is
no sui able o he e alua ion o sub-cycle a ia ions.
4. Cha ac e is ics and limi a ions o he exis ing measu emen
sys ems
4.1. Non-in en ional emissions
The co ec cha ac e iza ion o he emissions equi es p ope ly
designed and calib a ed measu emen sys ems. In p inciple, he emis-
sions can be eco ded by means o comme cial equipmen , such as
spec um analyze s o ec o ne wo k analyze s (VNAs), connec ed o
he g id. Some au ho s ha e p oposed measu emen sys ems o he
assessmen o he NIEs in he LV g id based on hese de ices [72,76,
163]. Howe e , he undamen al componen o he g id (220 V, 50 Hz),
in addi ion o he o e ol ages gene a ed due o he connec ion o
disconnec ion o o he elec onic equipmen o he measu ing equip-
men i sel , can damage he measu emen sys em. Fo his eason, many
au ho s ha e op ed o using oscilloscopes, also e e ed o as da a
acquisi ion (DAQ) de ices in he li e a u e, connec ed o he g id
h ough coupling ci cui s (CCs) and ol age/cu en p obes [44,45,54,
55,59,62–65,70,71,73,79,83,129,132–134,137]. An al e na i e based
on equipmen designed o he ansmission o PLC signals is p esen ed
in [164]. In Fig. 8, a summa y o he measu emen sys ems o he
assessmen o he NIEs in he PLC equency bands is p esen ed.
Thus, he p o ec ion o he measu ing de ices is a key aspec o be
aken in o conside a ion, wi hou o e looking ha he ci cui s used o
p o ec he sys em can conside ably educe he accu acy o he mea-
su emen [165,166]. I should be no ed ha , ega dless o whe he a
cu en o ol age p obe is used, a p io cha ac e iza ion is essen ial, so
ha he measu emen is no a ec ed. Fo his eason, i is necessa y o
e i y ha he equency esponse o he p obe, ega dless o he
impedance o he g id, is la in he whole equency band o analysis
[167].
Since he emissions show a ime-dependen beha io [76], he
cha ac e iza ion o he p opaga ion and in e ac ion o he emissions
gene a ed by he elec onic equipmen can only be pe o med by syn-
ch onized ials a di e en elec ical poin s o he LV dis ibu ion g id
[168]. A solu ion implemen ed o achie ing synch oniza ion wi h a
p ecision o ns by means o a GPS module esponsible o he gene a ion
o a Pulse pe Second (PPS) signal is desc ibed in [97].
4.2. Impedance and channel esponse
In he same way as o he NIEs, one o he main obs acles o
assessing he g id access impedance and he CFR is he limi ed numbe
o sys ems de eloped o measu emen s in he LV dis ibu ion g id, mos
o hem o equencies below 500 kHz. The p o ec ion equi emen s o
Fig. 7. Modulus and phase o he e e ence impedances de ined in CISPR 16–1–2 and IEC 61000–4–7.
Fig. 8. Summa y o he exis ing measu emen sys ems o he assessmen o he
NIEs in he PLC equency band.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
9
a oid damage o he measu emen de ices, he equisi e o po abili y
and independen powe supply a e some o he challenges o be
add essed in de eloping a measu emen sys em ha can be used o on-
ield ials. Besides, i is essen ial o ob ain in o ma ion on he modulus
and phase o bo h pa ame e s, wi h he aim o comple ely unde s anding
he beha io o he p opaga ion channel.
Conce ning he g id access impedance sys ems, some comme cial
impedance analyze s and VNAs p o ide accu a e impedance measu e-
men s o elec onic isola ed de ices in con olled labo a o y condi ions
[163,169–172]. Howe e , hese measu emen de ices a e no designed
o be used in he ield and a e no usually alid o measu emen s o
g ounded de ices. Mo eo e , hey p esen misma ch issues, due o he
di e ence be ween he in e nal impedance (50 Ω) and he g id alues.
Addi ionally, o a oid damage o hese de ices due o high powe
emissions [167], hey would equi e high-p o ec ion coupling ci cui s
ha modi y he injec ion and ecep ion condi ions and, consequen ly,
educe he measu emen accu acy.
Fo his eason, some au ho s ha e de eloped ad-hoc impedance
measu emen sys ems adap ed o he cha ac e is ics o he LV g id.
Acco ding o he li e a u e, hese sys ems can be classi ied depending on
he o igin o he ha monic cu en used o he sys em exci a ion [173].
Unde his classi ica ion, he e a e non-in asi e me hods, based on
gene a ing a cu en by means o exis ing ne wo k componen s and
which a e only able o measu e a hose equencies whe e conside able
emissions a e eco ded [174,175], and in asi e me hods, which use an
ex e nal sys em o gene a ing and injec ing his cu en in o he
ne wo k. In o de o p o ide in o ma ion o e he whole equency
band, a es signal should be injec ed in o he g id. The in asi e
me hods, in u n, can be classi ied as single o mul iple equency sweep
me hods. The la e allows a conside able educ ion o he measu ing
ime, ob aining simila accu acy o he single equency sweep me hod.
Some examples o in asi e me hods can be ound in [55,157,176–180].
Rega ding he measu emen sys ems o he assessmen o he sub-
cycle impedance a ia ions in he LV g id, di e en me hods ha e
been published so a . Fo example, in [173], a se up based on he in-
jec ion o a single sweep by means o a linea ampli ie is p esen ed o
he equency band om 130 Hz o 150 kHz. Re e ence [181], in
con as , calcula es he impedance wi h espec o ime and equency by
ansmi ing mul iple sweeps in he band 30–500 kHz. To his end, a
de eloped ca d o he synch oniza ion wi h he mains equency is
equi ed. Ano he measu emen sys em is epo ed in [145], which also
allows o measu e cu en and ol age a e injec ing a signal in o he
powe line. This sys em includes signal p ocessing o e o co ec ion
and a se o calib a ions se ups. I should be no ed ha simila sys ems
ha e been p esen ed by he same au ho s in [166,182,183]. Re e ences
[166,182] only co e he equency band assigned o NB-PLC
(30–500 kHz), whe eas [183] p o ides impedance alues up o
1 MHz. Finally, in [184], a measu emen sys em o he cha ac e iza ion
o he mean and sub-cycle g id access impedance in he equency ange
up o 10 MHz is p esen ed. This sys em is based on he injec ion o a es
signal composed o sho single- equency bu s s in he equency band
o in e es ob aining a maximum de ia ion wi hin ±8%.
In Table 3, a compa ison o he published measu emen sys ems o
he assessmen o he g id access impedance in he PLC equency bands
is included.
In he li e a u e, CFR measu emen sys ems based on di e en
ha dwa e and conside ing di e en me hods can be ound. In some
cases, VNAs a e used, including coupling ci cui s o he p o ec ion o
he sys em om he mains signal [163,187–191]. I is impo an o
men ion ha he use o VNAs implies a g ea limi a ion in dis ance, since
i mus be connec ed a bo h sides be ween which he CFR is o be
measu ed. In gene al, hese sys ems include calib a ion p ocesses so ha
he e ec o he couple s does no a ec he measu emen s. In o he
cases, such as [169,192–195], a signal gene a o is used a he
Table 3
Compa ison o he published measu emen sys ems o he assessmen o he g id access impedance in he PLC equency bands.
Re e ence Mean/Sub-
cycle
F equency ange Connec ion o he g id Measu emen me hod
S iegle , 2015[173] Mean & sub-
cycle
130 Hz-150 kHz Coupling de ice +powe ampli ie In asi e me hod (single sweep)
Ch uszczyk, 2015[178] Mean 10 kHz-1 MHz T ans o me +RC il e +powe ampli ie In asi e me hod (sinusoidal signal
sweep)
Hallak, 2017[145] Mean & sub-
cycle
30–500 kHz T ans o me +capaci o +shun esis o +ampli ie In asi e me hod (single- equency
bu s s)
Hallak, 2018[166] Mean & sub-
cycle
30–500 kHz T ans o me +capaci o +shun esis o +ampli ie In asi e me hod
(single- equency bu s s)
El eki, 2018[181] Mean & sub-
cycle
30–500 kHz Capaci i e couple +LCL il e In asi e me hod (single- equency
bu s s)
Ne wo k analyze
Fe n´
andez (IGOR-Me e ),
2019[177]
Mean 1–500 kHz Di ec connec ion (ex e nal p obes) In asi e me hod (single- equency
s eps)
Fe n´
andez (UPV/EHU), 2019
[177]
Mean 20–500 kHz Via capaci i e couple (ex e nal p obes) In asi e me hod (single one-
con inuous sweep)
Fe n´
andez (TUD), 2019[177] Mean 0–200 kHz Di ec connec ion (in e nal p obes) In asi e me hod (single- equency
s eps)
Nieβ, 2020[185] Mean & sub-
cycle
30–500 kHz Coupling ci cui + esis i e shun s +ampli ie In asi e me hod (sweep sine wa e)
Saa ho , 2020[186] Mean DC-200 kHz TRIAC-based phase-con olled swi ch In asi e me hod (In ush ansien )
Nieβ, 2021[182] Sub-cycle 40–500 kHz Coupling ci cui In asi e me hod (cosine cu en signal)
Szymczyk, 2021[183] Mean & sub-
cycle
10 kHz-1 MHz High-pass il e (HPF) +shun esis o In asi e me hod (single- equency
bu s s)
Jensen, 2021[55] Mean 50 Hz-150 kHz Powe ampli ie + ans o me +cu en p obe
+ ol age/cu en p obe
In asi e me hod (mul i- one signal)
A ale, 2022[180] Mean 35.9375–90.625 kHz
98.4375–121.875 kHz
154.6875–487.5 kHz
Shun +50 Hz il e +PLC signal condi ioning ci cui In asi e me hod (G3-PLC signal)
E han, 2022[157] Mean 12–150 kHz Ampli ie +injec ion ans o me +cu en p obe
+ ol age p obe
In asi e me hod
(single- equency sweep)
Ade ibole, 2023[179] Mean 100–200 kHz Shun + wo capaci o s +3 induc ances + ans o me In asi e me hod (sweep sinusoid)
Fe n´
andez, 2023[184] Mean & sub-
cycle
20 kHz-10 MHz Capaci i e couple (ex e nal p obes) In asi e me hod (single- equency
bu s s)
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
10
ansmission side o he injec ion o a signal in o he g id, while a da a
digi ize boa d is esponsible o cap u ing he signal a he ecei e side.
Each equipmen is connec ed o he g id h ough a coupling ci cui , in
cha ge o p o ec ing he measu ing de ices. In Table 4, a compa ison o
he exis ing sys ems o he assessmen o he CFR in he PLC equency
bands is ga he ed.
The impedance and CFR measu emen sys ems show simila di i-
cul ies o be aced. Fi s , he p obes used o measu ing he cu en s
should be co ec ly cha ac e ized, as well as he measu ing ci cui s ha
a e pa o he sys em. Mo eo e , especially a highe equencies, he
pa asi ic e ec s o he cables can ha e a signi ican impac on he
measu emen [198] and, hus, a p io and accu a e cha ac e iza ion o
he connec ion cables is also needed. Finally, he measu emen sys em
should be capable no only o measu ing he mean g id access imped-
ance o CFR, bu also he sub-cycle a ia ions gi en wi hin he 20 ms.
5. Measu emen campaigns in he LV dis ibu ion g id
In his sec ion, a summa y o he measu emen campaigns pe o med
in he LV dis ibu ion g id o he cha ac e iza ion o he NIEs, a enu-
a ion/CFR, and impedance a e p esen ed. Fig. 9 shows a map wi h he
coun ies in which he cha ac e iza ion o he LV elec ical g id has been
add essed.
5.1. Non-in en ional emissions
Se e al au ho s ha e pe o med on- ield ials in di e en LV dis-
ibu ion g ids all a ound he wo ld conside ing ce ain sou ces o dis-
u bances and g id opologies. As an example, in [163], he NIEs a
indus ial, esiden ial, and u al a eas in he Tu kish LV g id a e cha -
ac e ized up o 100 kHz, ob aining emission ampli udes a ound 90
dBµV, 100 dBµV, and 75 dBµV, espec i ely. A s a is ical cha ac e iza-
ion o he emissions in he LV g id based on p obabilis ic unc ions is
p esen ed in [51], co e ing he 2–150 kHz equency band. Ano he
s a is ical analysis can be ound in [164], whe e a long- e m cha ac-
e iza ion o he emissions in he LV g id in Qa a is p esen ed consid-
e ing mo e han 1.8 billion samples a h ee di e en loca ions o e 10
days. The analysis is based on he s a iona i y, au oco ela ion, and in-
dependence o he NIEs in he equency ange om 10 kHz o 490 kHz.
A measu emen campaign ca ied ou in Spain [44], in u n, concludes
ha he p opaga ion channel in u al scena ios is ema kably noisy in
low equencies, as he highes NIEs occu a equencies lowe han
150 kHz, mainly lowe han 40 kHz. Howe e , high-ampli ude emis-
sions can be ound in he whole equency ange up o 500 kHz in u ban
en i onmen s. In [72], a simila cha ac e iza ion o he NIEs in he LV
g id in F ance in he NB-PLC equency band is ga he ed. This a icle
poin s ou ha he noise is p edominan ly cyclos a iona y and concludes
ha he backg ound noise is s able o e pe iods o hou s.
5.2. Channel esponse and a enua ion
The huge a ie y o causes ha in ol e a enua ions in PLC signals
and he dependency o he beha io o he elec ical g id on he
geog aphical a ea [190,192,199], mainly because o he he e ogenei y
o g id opologies and connec ed loads, makes he p edic ion o he PLC
channel esponse a complex s udy ha should be add essed. In [163],
o example, he a enua ion be ween he SS and he SMs in he LV g id
in Tu key up o 1 MHz is analyzed in esiden ial, indus ial, and u al
a eas, e ealing a enua ions a ound 20 dB (dis ances om 50 m o
150 m), 30 dB (dis ances om 80 m o 270 m), and 40 dB (dis ances
om 150 m o 500 m) a each scena io, espec i ely. Re e ence [200]
conside s a se o ep esen a i e g id opologies, concluding, as in [201],
ha an inc ease in dis ance implies highe ansmission losses. Howe e ,
i is also s a ed ha he e is no a linea end be ween he a enua ion
and he dis ance be ween communica ions equipmen , due o he g ea
impac o he numbe o b anches in ee like opologies. In [202], apa
Table 4
Compa ison o he exis ing measu emen sys ems o he assessmen o he CFR in he PLC equency bands.
Re e ence Indoo /
Ou doo
F equency
ange
Connec ion o he g id/P o ec ion Measu emen basis
Co ´
es, 2005[194] Indoo 1–20 MHz Coupling ci cui Signal gene a ion boa d (Tx)-Da a
acquisi ion boa d (Rx)
Tlich, 2008[189],
[190]
Indoo 30 kHz-100 MHz Couple box Vec o Ne wo k Analyze
Colen, 2013[169] Indoo /
Ou doo
1.7–50 MHz PLC couple +Ampli ie (Tx) Signal gene a ion boa d (Tx)-Da a
acquisi ion boa d (Rx)
Gassa a, 2014
[188]
Indoo 10–500 kHz Coupling ci cui s Vec o Ne wo k Analyze
Tonello, 2014
[187]
Indoo 1.8–30 MHz Couple s +ex ension cables (cha ac e is ic impedance o 50 Ω) Vec o Ne wo k Analyze
Co ´
es, 2015[191] Ou doo 3–95 kHz Coupling ci cui +High powe ampli ie (Tx)-
Coupling ci cui +Band-pass il e (BPF) +Analog o digi al con e e (Rx)
Signal gene a ion boa d (Tx)-Da a
acquisi ion boa d (Rx)
Oli ei a, 2017
[192]
Indoo 1.7–30 MHz
(band A)
1.7–50 MHz
(band B)
1.7–100 MHz
(band C)
Couple Signal gene a o (Tx)-Da a digi ize (Rx)
Pico one, 2020
[193]
Ou doo 1.7–100 MHz Couple Signal gene a o +Digi al analog
con e e (Tx)-Da a digi ize +analog
digi al con e e
Masood, 2020/
2021[196],
[197]
Ou doo 50–150 kHz - PLC anscei e s
Ade ibole, 2023
[179]
Indoo /
Ou doo
100–200 kHz Coupling ci cui (2 ans o me s, 3 capaci o s, 3 solid-s a e elays, and 2
shun esis o s) +analog on -end (Low-pass il e +Powe ampli ie in Tx,
BPF +P og ammable Gain Ampli ie s in Rx)
PLC modems
Gonz´
alez-Ramos,
2023[195]
Indoo /
Ou doo
1.7–15MHz Coupling ci cui Signal gene a o (Tx)-Oscilloscope (Rx)
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
11
om measu ing he a enua ion in he equency band associa ed o
NB-PLC, a cha ac e iza ion o he ansmission losses measu ed om
500 kHz o 10 MHz in he LV g id in China is p esen ed. A compa ison
be ween he mean ansmission losses measu ed in u ban and u al
esiden ial a eas is ca ied ou , concluding ha he a enua ion wi h
espec o equency in he u ban scena ios is la e han wha is
measu ed in u al a eas.
As p e iously men ioned, he cha ac e iza ion o he PLC channel is
ca ied ou acco ding o wo app oaches: he op-down app oach, whe e
a model is ob ained om a la ge numbe o ield ials, and he bo om-
up app oach, based on modeling he channel by applying TL heo y
[10].
Rega ding he op-down app oach, he indoo PLC channel has been
ex ensi ely analyzed in e ms o a e age channel a enua ion o channel
gain, delay sp ead, cohe ence bandwid h, and channel capaci y in
[187–190,192,203–210]. In [211], he cha ac e iza ion o he indoo
PLC channel based on mul ipa h phenomenon can be ound. Conce ning
he cha ac e iza ion o he ou doo scena io, o ins ance, in [191], he
g id access impedance, he channel esponse (in e ms o delay sp ead,
channel bandwid h, and a enua ion), and he NIEs in di e en u ban,
semiu ban, and u al scena ios in he LV dis ibu ion g id a e analyzed,
as well as he achie able da a a es. The esul s p esen ed in his pape
only co e he CENELEC A NB-PLC equency band (3–95 kHz). A
simila analysis in he equency ange assigned o BB-PLC
(1.7–100 MHz) is ga he ed in [193], whe e he B azilian ou doo sce-
na io is cha ac e ized conside ing he a e age channel a enua ion, oo
mean squa ed delay sp ead, cohe ence bandwid h, cohe ence ime, and
he achie able da a a e. In e e ence [212], a cha ac e iza ion o he
ou doo channel conside ing he mul ipa h signal p opaga ion heo y is
p esen ed. This model is e i ied by a se o channel esponses ob ained
by means o ield ials.
Since in he bo om-up app oach he ne wo k elemen s a e ma he-
ma ically modeled, a ho ough knowledge o he elec ical g id ( opol-
ogy, cable cha ac e is ics, load impedances, e c.) is needed. This
app oach conside s he wo-conduc o TL (2TL) heo y [213–219], used
o powe ne wo ks connec ed wi h wo-conduc o ansmission lines,
and he mul i-conduc o TL (MTL) heo y [220–227], which is a
gene aliza ion o he 2TL app oach. The TL heo y has also been applied
by some au ho s o he cha ac e iza ion o he ou doo PLC channel
[217,222,223,228,229]. In e e ence [230], bo h he LV ae ial and un-
de g ound cable dis ibu ion lines a e model by means o he MTL
heo y.
A combina ion o he op-down and bo om-up app oaches can be
ound in [196,197], whe e a s a is ical analysis o he NB-PLC channel in
he LV g id in Pakis an is p esen ed up o 150 kHz and 500 kHz,
espec i ely.
The e o e, aking in o accoun ha he CFR is one o he mos c i ical
ac o s ha could cause communica ions o ail and ha he e is a lack o
knowledge abou i s beha io on he elec ical g id, a mo e exhaus i e
cha ac e iza ion o he channel in he equencies assigned o PLC
echnologies is s ill needed.
5.3. Impedance
Only a ew measu emen campaigns o cha ac e ize he g id
impedance ha e been pe o med all a ound he wo ld. In [163], he g id
access impedance in he LV g id in Tu key is cha ac e ized up o
100 kHz, conside ing esiden ial, u al, and indus ial scena ios. The
pape concludes ha , ega dless o he a ea, impedance alues below
10 Ω a e measu ed in he whole equency band. Ano he measu emen
campaign co e ing he same equency ange in Aus ia, Swi ze land,
Czech Republic, and Ge many is p esen ed in [158], showing ha he
impedance inc eases wi h equency and ha emains below he IEC
61000–4–7 e e ence impedance in he whole equency band. In [157],
phase o neu al impedance measu emen s a e ca ied ou a an elec-
ical ehicle cha ging plaza a he uni e si y and ou esiden ial
household ins alla ions in The Ne he lands in he 9–150 kHz equency
ange. This a icle concludes, i s , ha , due o he capaci o included in
hei ci cui y, bo h EVs and PVs imply a low-impedance pa h o
communica ions signals, and, second, ha household appliances ha e a
conside able e ec on he g id impedance. Mo eo e , he impo ance o
he opology on he impedance is also highligh ed. Al hough mos o he
published s udies ocus, exclusi ely, on he 9–150 kHz equency band,
some publica ions p o ide in o ma ion up o 500 kHz. Fo example, in
[8], he access impedance om 35 kHz o 500 kHz a h ee T ans o me
S a ions (TCs) and a a se o access poin s o he LV dis ibu ion g id in
he Basque Coun y (Spain) has been cha ac e ized. Fo his pu pose,
wo u ban and a u al scena io ha e been conside ed. The s udy p o-
ides an insigh in o he g ea di e ences be ween he impedances
measu ed a each elec ical poin and concludes ha he u ban dis i-
bu ion can be modeled as a pa icula scena io o sho cable sec ions
and nume ous homes. In [150], a simila cha ac e iza ion in he e-
quency band 30–500 kHz is pe o med, measu ing he g id access
impedance in an u ban/subu ban a ea in China wi h low- ise apa men
Fig. 9. Wo ld map wi h he coun ies in which measu emen campaigns ha e been ca ied ou o he cha ac e iza ion o he NIEs, a enua ion/CFR, and impedance
in he ou doo LV g id.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 36 (2023) 101217
12
buildings.
The beha io o he g id impedance in he equency ange abo e
1 MHz is s ill unknown. In [231], a measu emen campaign o i s s a-
is ical analysis in he B azilian LV ne wo k is p esen ed, concluding ha
he impedance is a non-s a iona y andom p ocess. Howe e , hese e-
sul s canno be ex apola ed o any dis ibu ion ne wo k, since he
pe o mance o he elec ical g id a he equencies associa ed o PLC
changes d as ically om one coun y o ano he , mainly due o he di-
e si y o loads and g id opologies [190,192,199].
6. Open issues
The s a e o he a p esen ed abo e shows some impo an limi a-
ions. Fi s , he e a e no comme cial measu emen sys ems adap ed o
he challenging condi ions o he LV g id. Impedance and ne wo k an-
alyze s, commonly used unde labo a o y condi ions, a e no in ended
o ne wo k measu emen s, mainly because hey can be damaged by he
undamen al equency o he g id (50 Hz, 220 V) o by o e ol ages
due o he connec ion/disconnec ion o nea by equipmen . Fo his
eason, coupling de ices a e used o p o ec ion pu poses, al hough hey
educe conside ably he accu acy o he measu emen . In his con ex ,
he de elopmen o measu emen sys ems adap ed o he elec ical g id,
co e ing bo h he NB-PLC and BB-PLC equency bands, in addi ion o
being capable o measu ing sub-cycle a ia ions and p o iding in o -
ma ion on he modulus and phase, is o pa icula impo ance. Fo his
pu pose, all he componen s ha a e pa o he sys em (cu en / ol age
p obes, p o ec ion ci cui s, e c.), as well as he connec ion cables, which,
due o pa asi ic e ec s, could ha e a conside able impac on he mea-
su emen accu acy a highe equencies, should be co ec ly cha ac-
e ized. In consequence, he cha ac e iza ion o he beha io o
connec ion cables, especially in he BB-PLC equency band, is also a
ma e o be analyzed [198], which should be based on heo e ical and
physical models conside ing di e en cable leng hs, cable diame e s,
and cable ypes.
The measu emen sys ems o he cha ac e iza ion o he LV g id
should be e alua ed ollowing a s anda dized me hod based on e e -
ence measu emen s. Howe e , up o now, no e e ence impedance has
been de ined o he p ope calib a ion o a speci ic measu emen sys-
em, as well as he compa ison be ween sys ems based on di e en
echniques o ha dwa e/so wa e. The i s a emp o his p ocess can
be ound in [162], concluding ha he measu ing condi ions (connec-
ion o he measu ing equipmen o he ea h and neu al conduc o s o
he supply o he e e ence impedance) could conside ably a ec he
ob ained esul s. Besides, only s a ic impedances, i.e., impedances no
showing sub-cycle a ia ions, ha e been aken in o conside a ion.
Rega ding he conduc ed emissions, he e is no no ma i e measu emen
me hod o he pos -p ocessing applied o he eco ded signals o e al-
ua e NIEs in he equency domain. CISPR 16–1–1 migh be used o
ob aining he QP alues o he eco ded signals. Howe e , his me hod
shows conside able d awbacks, such as high complexi y, compu a ional
bu den o memo y equi emen s [99]. Thus, he design o new and
highe -pe o mance me hods is essen ial o he co ec cha ac e iza ion
o he NIEs in eal g id condi ions. I is impo an o men ion ha his
e alua ion should be based on e e ence signals wi h di e en ime and
equency ea u es, which ha e no been de ined ye . Fo his eason,
se e al au ho s ha e op ed o assessing he pe o mance o measu e-
men me hods based on g id measu emen s, which a e highly dependen
on he cha ac e is ics and accu acy o he acquisi ion measu emen
sys ems, and ad-hoc designed syn he ic signals, which a e no commonly
o public use. Mo eo e , hese me hods only conside he spec al
analysis o he emissions, wi hou add essing hei a ia ion o e ime
and, he e o e, he de ini ion o s anda dized me hods o he ime
cha ac e iza ion o he NIEs in he LV g id is also a esea ch line o be
explo ed.
Ano he aspec ha should be aken in o accoun is he de ini ion o
emissions limi s o he wide ange o elec onic de ices connec ed o he
g id. Al hough he CISPR has al eady speci ied he maximum ampli ude
o conduc ed emissions ha can be gene a ed by ligh ing (CISPR15 [82])
and cooking appliances (CISPR11 [85]), he e a e sca ce speci ic limi s
o o he de ices ha will ha e a high pene a ion in he ollowing yea s,
such as EVs, PVs, o hyd opowe bombs. These limi s should be de ined
o bo h he NB-PLC and BB-PLC equency bands in RMS, a e age, and
QP alues. Simila ly, he CL, he limi s ha should no be o e passed by
conduc ed emissions in he LV g id, ha e only been es ablished o he
2–150 kHz equency band and, he e o e, he e is a need om s an-
da diza ion bodies o ex end hem o he uppe NB-PLC (150–500 kHz)
and BB-PLC equency anges.
The de ini ion o labo a o y scena ios ha a e mo e ep esen a i e o
he ac ual cha ac e is ics o he LV dis ibu ion g id is ano he aspec
cu en ly demanded by he scien i ic communi y. These scena ios would
allow he cha ac e iza ion o he emissions gene a ed by speci ic
equipmen , in addi ion o analyzing hei capaci y o p opaga ion and
in e ac ion wi h o he elec onic de ices, a oiding uncon ollable ac-
o s such as g id impedance, unknown emission sou ces, e c.
Finally, i should be men ioned ha , due o he high pene a ion o
EVs and DERs in ecen yea s, he cha ac e is ics o he LV g id as a
ansmission medium a e expec ed o ha e changed conside ably, and
many mo e changes will ollow in he nea u u e. Since he majo i y o
he s udies add essing he cha ac e iza ion o he ou doo p opaga ion
channel we e published in he ea ly 2000 s, u he esea ch in his a ea
is needed, in o de o de ine and implemen high-pe o mance PLC
echnologies allowing he de elopmen o u u e SG applica ions.
Al hough he esul s ob ained o indoo channels canno be di ec ly
ex apola ed o he dis ibu ion g id scena io, since a each en i onmen
di e en ac i i ies a e occu ing [1], he expe ience and me hodologies
de eloped in his scena io can se e as a i s s ep owa ds he de ini ion
o new me hods o he cha ac e iza ion o he ou doo ansmission
channel.
7. Conclusions
The elec ical g id is a hos ile medium o da a ansmission, as he
powe line cable is no designed o communica ions and i s cha ac e -
is ics change o e ime and equency. Mo eo e , he o e all beha io o
he LV dis ibu ion g id is d as ically changing due o he in oduc ion o
he Sma G id pa adigm. These changes also impose s onge commu-
nica ion equi emen s o u u e se ices, which highligh s he need o
a de ailed cha ac e iza ion o he pe o mance o he p opaga ion me-
dium in bo h NB-PLC and BB-PLC equency bands. This cha ac e iza-
ion should become he basis o he design o obus algo i hms o da a
coding, modula ion, and mul iplexing.
In his con ex , his a icle gi es an insigh in o he aspec s o be
conside ed o he cha ac e iza ion o he LV dis ibu ion g id as a
ansmission medium o PLC signals based on ield ials. The main
cha ac e is ics o NB-PLC and BB-PLC echnologies ha e been analyzed,
as well as he his o ical e olu ion o BB-PLC, om hei o igin in 1990 s
as an al e na i e o DSL o In e ne Access o hei cu en pu pose o
acili a ing he de elopmen o new SG se ices. Then, he main pa-
ame e s o be measu ed, namely, NIEs, a enua ion/channel esponse,
and impedance, ha e been ou lined, desc ibing hei undamen als,
along wi h he s ill insu icien no ma i e amewo k. As a consequence
o he numbe o wo ks ela ed o he p opaga ion and in e ac ion o he
emissions h ough he g id and he special in e es o he scien i ic
communi y in he emissions gene a ed by EVs, hese opics ha e been
co e ed mo e deeply in sepa a e sec ions. Then, he need o de elop
p ope ly designed and calib a ed measu emen sys ems adap ed o he
g id has been highligh ed, gi ing an o e iew o hei cha ac e is ics
and limi a ions o be aced. Mo eo e , an o e iew o he measu emen
sys ems ound in he li e a u e, based on di e en echniques, has also
been p esen ed. Finally, a comp ehensi e e iew o he measu emen
campaigns ca ied ou in di e en LV g ids all a ound he wo ld o he
cha ac e iza ion o he ou doo elec ical ne wo k as a p opaga ion
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
A ailable online 16 Ma ch 2024
0378-7796/© 2024 The Au ho (s). Published by Else ie B.V. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-
nc-nd/4.0/).
Emissions gene a ed by elec ic ehicles in he 9-500 kHz band:
Cha ac e iza ion, p opaga ion, and in e ac ion
Jon Gonz´
alez-Ramos
*
, Alexande Galla e a , Igo Fe n´
andez , I zia Angulo , Da id de la Vega ,
Amaia A inda
Uni e si y o he Basque Coun y (UPV/EHU), Bilbao, Spain
ARTICLE INFO
Keywo ds:
Conduc ed non-in en ional emissions
Elec ic ehicles
In e ac ion
Low ol age g id
P opaga ion
ABSTRACT
Elec ic Vehicle Cha ging P ocesses (EVCPs), due o he in ol ed powe elec onics wi h high-swi ching e-
quencies, gene a e high-ampli ude emissions ha could ha e a nega i e impac on Powe Quali y (PQ) and
Na owband Powe Line Communica ions (NB-PLC) sys ems. In his con ex , his pape deals, i s , wi h he
equency and ime cha ac e iza ion o he dis u bances gene a ed by ou EVCPs in he 9-500 kHz equency
ange. The s udy is based on measu emen s ca ied ou in a con olled Low Vol age (LV) g id. Fo his pu pose, a
no el p ocedu e o he equency and ime cha ac e iza ion is p oposed. The equency cha ac e iza ion is based
on he calcula ion o a se o pa ame e s, which allows e alua ing no only he ampli ude o he dis u bances in
he equency band unde analysis, bu also hei spec al dis ibu ion. The ime a iabili y is cha ac e ized by
means o a Fas Fou ie T ans o m (FFT) analysis ha leads o a simpli ied model o e alua e he ime-dependen
beha io o he dis u bances, which shows a sub-cycle pe iodic pa e n o he emissions in he equency band o
in e es . Second, he p opaga ion o he p e iously cha ac e ized emissions is analyzed, concluding ha hey
p opaga e se e al me e s h ough he LV g id. Al hough in mos cases he dis u bances a e a enua ed wi h
dis ance, he e migh be esonances ha lead o highe ampli udes a an elec ical poin dis an om he sou ce
o he emissions. Finally, he in luence o he simul aneous cha ging o se e al EVs is s udied. The esul s show
ha , in gene al, he ampli udes co espond o he supe posi ion o he indi idual dis u bances, in addi ion o
in e modula ion p oduc s due o he swi ching equencies o he in e e s.
As a conclusion, he high ampli ude ime- a ying emissions, oge he wi h hei capaci y o p opaga ion and
in e ac ion, a e a ma e o be analyzed due o hei in luence on PQ and hei po en ial deg ada ion o he
pe o mance o PLC in he equency band om 9 kHz o 500 kHz.
1. In oduc ion
Elec ic Vehicles (EVs) play a undamen al ole in he concep o
Sma G id, and any Sma Ci y ini ia i e should include EV cha ging in
hei plans [1]. Howe e , EVs ace signi ican ba e y- ela ed challenges,
which lead o con inuous inno a ion in o de o imp o e aspec s such as
d i ing ange, cha ging ime, ba e y cos , e c. [2].
One unin ended side e ec o EV Cha ging S a ions (EVCSs) is
elec omagne ic compa ibili y (EMC) issues due o conduc ed Non-
In en ional Emissions (NIEs) [3], which may po en ially impac Powe
Line Communica ions (PLC) [4–6]. In e e ing emissions due o ec i i-
ca ion caused by powe elec onics wi h high-swi ching equencies up
o se e al hund ed kHz could p opaga e and ha e an e ec no only
close o he EVCS bu also on he su ounding Low Vol age (LV) g id [7,
8]. In o de o co e all he equency ange whe e Na owband PLC
(NB-PLC) echnologies can be de eloped, i is impo an o ex end he
Abb e ia ions: CDF, Cumula i e Dis ibu ion Func ion; EDF, ´
Elec ici e de F ance; EMC, Elec omagne ic Compa ibili y; EV, Elec ic Vehicle; EVCS, Elec ic
Vehicle Cha ging S a ion; EVCP, Elec ic Vehicle Cha ging P ocess; FFT, Fas Fou ie T ans o m; IFFT, In e se Fas Fou ie T ans o m; LISN, Line Impedance S a-
biliza ion Ne wo k; LV, Low Vol age; MV, Medium Vol age; NB-PLC, Na owband Powe Line Communica ions; NIEs, Non-In en ional Emissions; PFBL, Pe cen age o
equency bins exceeding he PLC ou -o -band emission limi s; PLC, Powe Line Communica ions; POC, Poin o Connec ion; PPS, Pulse pe Second; PQ, Powe
Quali y; QP, Quasi-Peak; SoC, S a e o cha ge; STFT, Sho - e m Fou ie T ans o m; TSHV, To al Sup aha monic Vol age.
* Co esponding au ho .
E-mail add esses: [email p o ec ed] (J. Gonz´
alez-Ramos), [email p o ec ed] (A. Galla e a), [email p o ec ed] (I. Fe n´
andez), i zia .
[email p o ec ed] (I. Angulo), [email p o ec ed] (D. de la Vega), [email p o ec ed] (A. A inda).
Con en s lis s a ailable a ScienceDi ec
Elec ic Powe Sys ems Resea ch
jou nal homepage: www.else ie .com/loca e/eps
h ps://doi.o g/10.1016/j.eps .2024.110289
Recei ed 8 June 2023; Recei ed in e ised o m 11 Oc obe 2023; Accep ed 27 Feb ua y 2024
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
2
s udy o hese emissions om he classical sup aha monic ange (2–150
kHz) [9] o he equency band up o 500 kHz. In his con ex , his pape
is ocused on he cha ac e iza ion o he NIEs gene a ed by Elec ic
Vehicle Cha ging P ocesses (EVCPs) in a con olled LV g id in he e-
quency ange 9-500 kHz, in addi ion o analyzing hei capaci y o
p opaga ion and in e ac ion. The s udy is based on measu emen s ca -
ied ou in a con olled LV g id scena io.
2. S a e o he a
2.1. Non-In en ional emissions gene a ed by EVCPs
The e ec o he pene a ion o EVs on he Powe Quali y (PQ) pa-
ame e s (up o 2 kHz) has been deeply s udied in he exis ing li e a u e
[10–12]. Al hough acco ding o [13], he ec i ie s included in he ci -
cui y o he ba e y cha ging p ocess a e a sou ce o high-ampli ude
emissions, only a ew esea ch wo ks ha e been ca ied ou o hei
p ope cha ac e iza ion in he sup aha monic ange (2 kHz o 150 kHz).
In [14], he main causes o he spec al componen s o he emissions
gene a ed by an elec ic bus in he 2-150 kHz equency ange a e
iden i ied. In [7], he sup aha monic cu en s o a bidi ec ional EVCS
a e measu ed and analyzed in he equency domain, showing bo h
wide-band and na ow-band emissions. Simila ly, e e ence [15] ana-
lyzes he sup aha monic emissions gene a ed by nine popula EV models
in he Ne he lands o equencies up o 100 kHz. As s a ed in [16] and
[17], he spec al pa e n o he dis u bances is highly dependen on he
EV model and he cha ging egime, and hus, ex ensi e measu emen
campaigns a e necessa y o model his ype o NIEs.
The au ho s o he cu en wo k also p esen ed a ime- and
equency-cha ac e iza ion o he conduc ed dis u bances gene a ed by
wo comme cial EV models om 9 kHz up o 500 kHz in [18]. The
measu emen s analyzed in [18] we e pe o med on an a i icial Line
Impedance S abiliza ion Ne wo k (LISN). Howe e , conduc ed emis-
sions in eal scena ios will s ongly depend on local g id condi ions and
he in luence and in e ac ion wi h o he elec ical equipmen connec ed
close-by [7].
The e o e, he cu en s a e-o - he-a p esen s some esea ch gaps
ha s ill need o be co e ed in his a ea. Fi s , he e a e sca ce s udies
a ailable o he equency ange abo e 150 kHz, so ha hei po en ial
nega i e e ec on NB-PLC echnology co e ing he equency band up o
500 kHz is unknown. Second, as al eady shown in he li e a u e, he
conduc ed emissions depend on he speci ic cha ging p ocess (EVCS, EV
model and s a e o cha ge). This highligh s he need o ex ensi e
measu emen campaigns conside ing di e en cha ging si ua ions, in
o de o p o ide a big pic u e o he cu en and u u e si ua ions o EV
de elopmen .
2.2. P opaga ion and in e ac ion o NIEs
The conduc ed dis u bances in his equency ange ha e been
p o en o p opaga e o e se e al kilome e s in he LV and Medium
Vol age (MV) g ids [19–21]. Mo eo e , as in a eal si ua ion an EV is
su ounded by o he sou ces o dis u bances, he esul an emissions a a
ce ain elec ical poin a e expec ed o be he combina ion o he
p opaga ed emissions gene a ed by each sou ce. Fo his eason, in o de
o ully de e mine he e ec s o he EVCPs on he ampli ude o he
emissions in he LV g id, i is essen ial no only o p ope ly cha ac e ize
hese emissions in he ime and equency domains a he Poin o
Connec ion (POC) whe e he EVCS is ins alled, bu also o analyze hei
p opaga ion and in e ac ion wi h o he connec ed equipmen .
Up o now, se e al labo a o y and on-si e measu emen s ha e been
ca ied ou o he cha ac e iza ion o he in e ac ion o he dis u bances
gene a ed by di e en elec onic de ices. Acco ding o he exis ing
s udies, he dis u bances measu ed a he e minal o a ce ain de ice
consis o a p ima y and a seconda y emission. The p ima y emission,
dependen on he g id impedance and, he e o e, on he loca ion and
ime o he day, is de ined as he pa o he cu en gene a ed by he
sou ces inside he de ice. In con as , he seconda y emission co e-
sponds o he pa o he cu en o igina ed by sou ces ou side he de ice
[19,22–27]. In [24], whe e a model o es ima ing he emissions o a
labo a o y ins alla ion composed o mul iple de ices is p esen ed, i is
s a ed ha his seconda y emission canno be dis ega ded and should be
aken in o accoun o he co ec cha ac e iza ion o he esul an
emission gene a ed by an indi idual appliance.
In [23], by means o a simula ion model, he in luence o he leng h
o he line on he dis u bance le els is s udied. Fo ha pu pose, he
in e ac ion be ween some speci ic de ices connec ed a he same and
di e en elec ical poin s has been conside ed. The s udy in [23] is
based on he analysis o he emissions a speci ic equencies depending
on he numbe o connec ed de ices and he leng h o he elec ical
cable.
Some wo ks ha e also analyzed he in luence o he in e ac ion be-
ween elec onic equipmen in a labo a o y en i onmen . This in e ac-
ion depends on he de ices and he cha ac e is ics o he sou ce
impedance [19], and he ime o he day [28]. Acco ding o he au ho s
o hose con ibu ions, he emissions gene a ed om a speci ic de ice
inc ease i mo e appliances a e connec ed a he same elec ical ins al-
la ion, whe eas he o al emission o he g id is educed [24].
Rega ding he p opaga ion and in e ac ion o he emissions gene -
a ed by EVCPs, in [29], he long- e m a ia ions o he NIEs gene a ed
om h ee elec ic cha ging in as uc u es a e analyzed up o 100 kHz,
bo h in he equency and ime domains. The conclusion o he wo k is
ha he highe he numbe o EVs, he highe he dis u bances measu ed
in he powe g id. A simila s udy is pe o med in [30], whe e he ime
and equency analysis o he beha io o he sup aha monics om
EVCSs is add essed o equencies up o 100 kHz in labo a o y
condi ions.
Howe e , he exis ing li e a u e only conside s he p opaga ion o
emissions om indi idual de ices, wi hou analyzing he cumula i e
e ec o NIEs om se e al sou ces on hei p opaga ion h ough he
g id. Mo eo e , as s a ed in [31], he dis u bances gene a ed by elec-
onic appliances show a conside able ime-dependen beha io and,
consequen ly, he p opaga ion analysis can only be ca ied ou by means
o synch onized measu emen s a di e en elec ical poin s [32].
Consequen ly, he s a e-o - he-a shows ha he agg ega ion o
simul aneous emissions om a ious EV cha ging p ocesses should be
in es iga ed. Finally, conduc ed NIEs om EVs should be analyzed in
mo e ealis ic g id condi ions, a oiding uncon olled g id ac o s, in
o de o e alua e p opaga ion and in e ac ion wi h o he elec ical
equipmen .
3. Pape con ibu ions and s uc u e
Conside ing he esea ch gaps p esen ed abo e, he main objec i e o
his pape elies on he equency and ime cha ac e iza ion o he
emissions gene a ed by a se o EVs du ing hei cha ging p ocess in he
9-500 kHz equency ange. The s udy, based on measu emen s ca ied
ou in a con olled LV g id, p oposes a no el p ocedu e ha allows
e alua ing a ce ain dis u bance by means o he calcula ion o a se o
pa ame e s and a Fas Fou ie T ans o m (FFT) analysis. Mo eo e , he
p opaga ion o he NIEs along he g id is analyzed by means o syn-
ch onized measu emen s a di e en connec ion poin s. Finally, in o de
o s udy he in e ac ion o he emissions om a ious EVCPs, he ag-
g ega ion o simul aneous emissions is in es iga ed.
Wi h he aim o highligh ing he no el y o his pape , Table I shows
a summa y o he p e iously p esen ed s a e o he a , in addi ion o he
main con ibu ions o he cu en pape .
The es o he pape is o ganized as ollows. In sec ion IV, he
measu emen scena io is p esen ed, including he a ailable echnical
speci ica ions o he EVCSs unde s udy, as well as he measu emen
sys em used o he eco ding o he NIEs and he signal p ocessing. In
sec ion V, he NIEs gene a ed by ou di e en EVCPs a e cha ac e ized
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
3
in he equency and ime domains up o 500 kHz. In sec ion V.A, he
Quasi-Peak (QP) alues o he ampli ude o he dis u bances, acco ding
o he CISPR16-1-1 [33], a e p esen ed. The s udy o he ime- a ian
beha io o he dis u bances o e he measu emen ime is discussed
in sec ion III.B. In sec ions VI and VII, he p opaga ion and in e ac ion o
he emissions p e iously cha ac e ized a e add essed. Finally, in sec ion
VIII, he main conclusions o he wo k a e ga he ed.
4. Me hodology
4.1. Desc ip ion o he measu emen scena io
The measu emen s ha suppo his s udy a e ca ied ou in he
“Concep G id” labo a o y o ´
Elec ici e de F ance (EDF), a unique
es ing acili y ha goes beyond ideal condi ions o labo a o y ials, bu
a he same ime, a oids uncon olled backg ound dis o ion ha may
subs an ially a ec he esul s [7]. This es ing scena io simula es a LV
dis ibu ion g id composed o a Seconda y Subs a ion (SS) and i e
houses, wi h a h ee-phase ins alla ion, o which di e en elec onic
de ices can be connec ed.
In his s udy, h ee di e en EVCSs a e analyzed. EVCS1 is ins alled
a H2, EVCS2 a H3, and EVCS3 a H5. The e a e 3 EV models a ailable,
and each EV model can only be cha ged a i s co esponding EVCS. A
ep esen a ion o he measu emen scena io is shown in Fig. 1, whe e
he dis ance be ween he di e en houses and he loca ion o he EVCSs
a e indica ed.
In o de o ha e a p edominan ly esis i e load a he POC, domes ic
hea e s we e connec ed o each house. The backg ound noise due o
hese loads was cha ac e ized a H4, conside ing his as he de aul si -
ua ion. All he measu emen esul s p esen ed in he ollowing sec ions
we e conduc ed in he same elec ical phase (monophasic
measu emen s).
The ou si ua ions o be analyzed co espond o he cha ging p ocess
o EV1 a 81% s a e o cha ge (SoC), EV2 a 100% and 75% SoC, and EV3
a 68% SoC. The dis u bances we e measu ed a he POC o each house
o which he EVCS unde s udy is ins alled. As he emissions gene a ed
by EVCP1 and EVCP3 we e only measu ed a a gi en SoC, he ea e he
SoC o hese wo EVCPs will no be speci ied each ime hese wo EVCPs
a e ci ed in he ex . In o de o analyze he p opaga ion o he emissions
gene a ed by each EVCP indi idually, synch onized measu emen s we e
ca ied ou a H2, H3, and H5. The analysis o he in e ac ion o he NIEs
is based on measu emen s pe o med a H2, H3, and H5 when he h ee
EVs a e cha ging simul aneously.
Limi ed echnical in o ma ion o he EVCSs unde s udy is a ailable,
wi h no in o ma ion abou EVCS2. In Table II, he echnical speci ica-
ions o EVCS1 and EVCS3 a e ga he ed. In all he ials, EVCS1 co e-
sponds o he cha ging pos o EV1, EVCS2 o he cha ging ci cui y o
EV2, which is ins alled inside EV2, and EVCS3 o he cha ging pos o
EV3.
4.2. Measu emen sys em o NIEs
The measu emen sys em o NIEs, shown in Fig. 2, is based on he
ol age p obe p esen ed in [36], which is connec ed o he elec ical
poin whe e he dis u bances a e measu ed. This ol age p obe shows a
la esponse o a wide ange o impedance alues ha may be ound in
he g id. A digi al oscilloscope, con olled by a lap op, is esponsible o
eco ding he signal wi h high accu acy (15 bi s o esolu ion in
magni ude) and a high sampling equency (8.92 MHz). The acquisi ion
o he NIEs is pe o med by means o pu pose-speci ic so wa e de el-
oped by he au ho s.
In o de o ca y ou synch onized measu emen s a h ee di e en
Table I
Summa y o he p e iously p esen ed s a e o he a and he main con ibu ions o he cu en pape .
Re e ence F equency
ange
Measu emen scena io Scope o he s udy
G asel, [7] 9-150 kHz Recons uc ed LV g id Cha ac e iza ion o he conduc ed emissions gene a ed by a V2G EVCS
Meye , [13] 50 Hz-150 kHz Labo a o y Cha ac e iza ion o he conduc ed emissions gene a ed by 19 EVCPs
Lode i, [14] 2-150 kHz Poin o Common Connec ion (LV
g id)
Cha ac e iza ion o he conduc ed emissions gene a ed by an elec ic bus induc i e cha ging
Slangen, [15] 0-100 kHz Tes lab o ElaadNL Cha ac e iza ion o he conduc ed emissions gene a ed by 9 EVCPs
Sch¨
o ke, [16] 2-150 kHz Labo a o y/LV g id Cha ac e iza ion o he conduc ed emissions gene a ed by 6 EVCPs
Da mawa dana,
[17]
2-150 kHz LV g id/Wa e o m gene a o (supply) Cha ac e iza ion o he conduc ed emissions gene a ed by wo EVCPs
Gonz´
alez-Ramos,
[18]
9-500 kHz LISN Cha ac e iza ion o he conduc ed emissions gene a ed by wo EVCPs
Espín-Delgado,
[20]
2-150 kHz LV g id (TU D esden labo a o y) P opaga ion and in e ac ion s udy o he conduc ed emissions gene a ed by PV in e e s and
LEDs based on a co ela ion and impedance analysis
Cassano, [23] 5.1 kHz and
15.1 kHz
Simula ions (LV g id) P opaga ion and in e ac ion analysis o he conduc ed emissions gene a ed by a as DC EVCS
based on a p ima y/seconda y emission app oach
Su a ia, [26] 2-150 kHz Labo a o y a he Uni e si y o Luleå Cha ac e iza ion o he conduc ed emissions gene a ed by di e en Powe Fac o Co ec ed
ci cui s. The p opaga ion o hese emissions is analyzed based on a p ima y/seconda y emission
app oach
S eubel, [29] 2-150 kHz Th ee di e en pa king ga ages wi h
cha ging in as uc u es
Cha ac e iza ion o he conduc ed emissions gene a ed by h ee EV cha ging in as uc u es in
he long- e m
Slangen, [30] 0-100 kHz Sma G id In e ope abili y Lab Cha ac e iza ion o he conduc ed emissions gene a ed by ou EVs. The p opaga ion and
in e ac ion o hese emissions a e analyzed based on a p ima y/seconda y emission app oach
Cu en pape 9-500 kHz Con olled LV dis ibu ion g id De ini ion o a no el p ocedu e o he cha ac e iza ion o he conduc ed emissions gene a ed by
ou EVCPs ( ime and equency domains). The p opaga ion and in e ac ion o hese emissions is
analyzed based on synch onized measu emen s a di e en elec ical poin s in he LV
dis ibu ion g id
Fig. 1. Measu emen scena io in he “Concep G id” labo a o y o EDF
in ´
Ecuelles.
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
4
loca ions, h ee measu emen sys ems as he one p esen ed in Fig. 2 a e
needed, as well as an iden ical ime se ing in each compu e . Fo his
pu pose, a GPS module is equi ed o gene a ing a Pulse pe Second
(PPS) signal wi h a p ecision o 20 ns. In his way, a de ia ion be ween
compu e s o less han a millisecond is achie ed. In each measu emen ,
he GPS signal is eco ded a he same ime as he dis u bances.
4.3. Signal p ocessing
Once he emissions a e eco ded, he esul s a e ep esen ed in a
colo scale wi h espec o equency (ho izon al axis) and ime ( e ical
axis) in he o m o spec og ams by means o a Ma lab sc ip , p o iding
a ime and equency esolu ion o 2 ms and 50 Hz, espec i ely. Fo his
pu pose, a sliding Lanczos windowing wi h a du a ion o 20 ms and an
o e lapping o 90% be ween consecu i e windows is applied. The
Lanczos window is de ined in (1),
whe e N is he numbe o samples pe 20 ms window.
Then, in o de o ob ain he measu ed spec um e e y 2 ms, a Sho -
Te m Fou ie T ans o m (STFT) is applied o each ime window ac-
co ding o (2),
Z[ c,k] = ∑
N−1
n=0
x[n−k]w[n]e−j2
π
n (2)
whe e
c
a e he equency componen s, and k he ime s eps be ween
consecu i e STFT ou pu s.
As he eco dings o he emissions las 600 s (excep o EV2 a 75%
SoC, whe e he emissions we e eco ded o 250 s), due o compu a-
ional e o , only 50 s o he eco dings a e ep esen ed. As an example,
Fig. 3 shows he spec og am du ing he i s 50 s o he emissions
gene a ed by he cha ging p ocess o EV1.
Subsequen ly, in o de o cha ac e ize he dis u bances in he e-
quency domain, he QP alues o he ampli ude o he emissions a e
ob ained acco ding o CISPR 16–1-1 s anda d [33], ob aining a single
igu e o QP alues o he ampli ude o he NIEs as a unc ion o e-
quency o each ime in e al o 50 s.
By con as , he ime cha ac e iza ion o he emissions is pe o med
by e alua ing he a iabili y o each equency bin o e he measu e-
men ime.
5. Cha ac e iza ion o indi idual emissions
5.1. F equency analysis
Fig. 4 shows he QP alues o he ampli ude o he emissions
gene a ed by he EVCP1, EVCP2 a 75% and 100% SoC, and EVCP3,
espec i ely, oge he wi h he emissions co esponding o he de aul
si ua ion.
As no emission limi s ha e been speci ied o he conduc ed dis u -
bances gene a ed by EVs du ing he cha ging p ocess, he ou -o -band
limi s de ined o communica ions equipmen in EN 50065-1 [37]
migh be conside ed as a conse a i e c i e ion [38] o compa ison
pu poses.
In gene al, he ampli ude o he emissions is e y high i compa ed o
he backg ound noise co esponding o he de aul si ua ion in which no
EV is connec ed o he g id. This does no go in line wi h he obse a ions
in [16], whe e no signi ican emission is isible abo e 50 kHz. This e-
in o ces he idea ha he emissions a e e y dependen on he speci ic
cha ac e is ics o he cha ging p ocess unde analysis.
Fo EVCP1 and EVCP2, he measu ed dis u bances a e in he o m o
ha monics o he swi ching equency o an in e e o he EV cha ge
[7],[13],[16]. Fo ins ance, Fig. 4(a) shows his spec al pa e n, whe e
ha monics o 10 kHz (ampli udes be ween 62 dBµV and 98 dBµV) a e
measu ed in he whole equency band. In he case o EVCP2, ha monics
o 16 kHz wi h ampli udes a ying be ween 82 dBµV and 115 dBµV a
75% SoC and be ween 82 dBµV and 106 dBµV a 100% SoC a e egis-
e ed. By con as , as shown in Fig. 4(c), a high-ampli ude emission
dec easing wi h equency in he o m o colo ed noise is obse ed o
EVCP3 be ween 70 kHz and 500 kHz (ampli udes a ying om 79 dBµV
Table II
Technical speci ica ions o EVCS1 and EVCS3 [34,35].
EVCS Model Mode Ou pu
Vol age
Ou pu
Cu en
Maximum single-
phase powe
EVCS1 NSQC442G 3 50–500 V
DC
0 – 120 A
DC
–
EVCS3 Wi y-
XEV100
3 230-400 V
(Adjus .)
– 32 KVA
Fig. 2. Measu emen sys em o Non-In en ional Emissions.
Fig. 3. Spec og am o he dis u bances gene a ed by he cha ging p ocess o
EV1 du ing 50 s.
w[n] = ⎧
⎪
⎪
⎪
⎨
⎪
⎪
⎪
⎩
sinc(2(2n
N−1))sinc(2n
N−1−1),n∕= N−1
2
1,n=N−1
2
,n=0,1,…,N−1(1)
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
5
o 48 dBµV). This migh be due o an e icien g id-side il e ci cui ,
PWM-con olled de ices using ac i e powe ac o co ec ion and/o a
con ol egime ha con inuously changes he swi ching equency [7,13,
30]. In o de o e i y his, a ime domain analysis is equi ed (see
sec ion V.B).
Mo eo e , Fig. 4(b) demons a es ha he ampli udes o he spec al
componen s do no necessa ily dec ease wi h equency, as obse ed in
[14]. Fig. 4(b) also shows ha he spec al pa e n a ies wi h he s a e
o cha ge [13,16,39]. Al hough simila ampli udes a e ob ained a he
na owband emissions i bo h s a es o cha ge a e compa ed, he e is a
mino shi in he undamen al equency o he spec al componen s
and, hence, he high-ampli ude na owband emissions a e no cen e ed
a exac ly he same equencies. This e ec is also ound in he esul s
p esen ed in [13,17]. As an example, Fig. 5 shows he emission gene -
a ed by he cha ging p ocess o EV2 a 100% and 75% SoC a ound 350
kHz. The peaks o he emission a e loca ed a 351.9 kHz and 352.4 kHz
espec i ely.
The e o e, he p e ious igu es clea ly show ha he QP alues o he
ampli ude o he emissions highly depend on he pa icula i ies o he
EVCPs (EVCS, EV model and s a e o cha ge), as di e en spec al
componen s and ampli udes o he emissions ha e been egis e ed. This
e ec migh be caused by he il e ci cui s o he EVs [13]. Conside ing
ha he de ails o he elec onic componen s a e no usually p o ided by
he manu ac u e s, his highligh s he need o basing he analysis o
emissions om EVCPs on expe imen al e idence [17].
In o de o nume ically cha ac e ize he QP alues o he ampli ude
o he emissions o he whole equency band o in e es , he To al
Sup aha monic Vol age (TSHV) [40–42] is calcula ed o he ou si u-
a ions unde s udy acco ding o (3).
TSHV =
∑
i=9821
i=1
Vi2
√
√
√
√(3)
This pa ame e , calcula ed in linea scale and hen con e ed o
loga i hmic, gi es a gene al o e iew o he ampli ude o he dis u -
bances gene a ed by each EVCP in he whole equency band. The
summa ion conside s he QP alues o he ampli ude o he emissions o
he 9821 equency bins co e ing he band unde analysis (9-500 kHz)
wi h a equency s ep size o 50 Hz.
Mo eo e , in o de o de e mine i he emissions abo e he limi a e
concen a ed a speci ic equencies o sp ead o e wide equency
bands, he pe cen age o equency bins exceeding he PLC ou -o -band
emission limi s (PFBL) is calcula ed.
PFBL =numbe o equency bins abo e he limi
o al numbe o equency bins ⋅100 (4)
Fig. 4. QP alues o he ampli ude o he emissions (dBµV) gene a ed by EVCP1 (a), EVCP2 a 75% and 100% SoC (b), and EVCP3 (c).
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
6
In Table III, he TSHV and PFBL a e ga he ed o each si ua ion unde
s udy.
Table III clea ly shows ha he emissions gene a ed by he cha ging
p ocess o EV1 in he 9-500 kHz equency band a e conside ably lowe
in compa ison wi h EVCP2 and EVCP3 (di e ences be ween 9 dB and 19
dB), in addi ion o being concen a ed in ewe equency bins (18%). I
he TSHV alues o EVCP2 a 75% and 100% SoC a e compa ed, i can
be concluded ha he emissions in oduced by his EVCP a e highe
when EV2 is no ully cha ged. This di e ence in he emission ampli-
udes is also e lec ed in he noise loo ob ained a 75% and 100% SoC,
which explains he change in he PFBL om 97% a 75% SoC o 62% a
100% SoC. Finally, i should be men ioned ha , despi e dis ibu ing he
emissions in only 48% o he equency bins, he cha ging p ocess o EV3
gene a es he highes dis u bance ampli udes acco ding o he TSHV.
The PFBL shows i he emission ampli udes abo e he PLC ou -o -
band emission limi s a e concen a ed a pa icula equencies o
sp ead o e wide equency bands, bu i does no e lec he spec al
dis ibu ion o he emissions o e he 9-500 kHz equency band. In
o de o show he spec al dis ibu ion o he emissions abo e he PLC
ou -o -band emissions limi s, he Cumula i e Dis ibu ion Func ion
(CDF) o he equency bins ha co espond o emission ampli udes
abo e he emission limi is calcula ed o each si ua ion unde es . In
Fig. 6, he ob ained CDFs o EVCP1, EVCP2 a 75% and 100% SoC, and
EVCP3, a e depic ed.
As shown in Fig. 6, he slope o he CDF o EVCP1 is much s eepe a
equencies abo e 150 kHz, which implies ha a highe numbe o
equency bins exceed he PLC ou -o -band emission limi s in he e-
quency ange 150-500 kHz. A simila beha io is obse ed abo e 70 kHz
i he CDF o EVCP2 a 100% SoC is analyzed. Rega ding he CDF ob-
ained o EVCP2 a 75% SoC, a linea end is shown in almos he
whole equency band o analysis, which demons a es a con inuous
uni o m dis ibu ion om 23 kHz o 500 kHz (i.e., all he emissions a e
abo e he limi in his equency ange). Again, he di e ence in he
esul s ob ained o bo h SoCs is a ec ed by he di e ence in he noise
loo measu ed in he wo si ua ions. Finally, he CDF o EVCP3 allows o
conclude ha , as shown in Fig. 6, all he equency bins exceed he limi s
be ween 9 kHz and 250 kHz (uni o m dis ibu ion in his equency
band).
These esul s a e o p ac ical in e es o planning PLC sys ems in his
equency band because, as shown in Fig. 4, he QP alues o he
ampli ude o he emissions gene a ed in he ou si ua ions unde s udy
exceed he limi s de ined o he PLC ou -o -band emissions a leas o
some equency bands. I he eigh channels de ined o PRIME 1.4 a e
conside ed [43], and e alua ing he CDFs ob ained in Fig. 6, i is possible
o op o he communica ion channels ha p esen lowe dis u bance
ampli udes. Fo example, in p esence o emissions simila o he si ua-
ion co esponding o he cha ging p ocess o EV1, channel 1 (42-89
kHz) and channel 2 (97–144 kHz) would p esen be e channel condi-
ions in e ms o dis u bances, as e y ew spec al componen s o he
emissions a e abo e he PLC ou -o -band emission limi s. In he case o
EV3, channels 5 (261-308 kHz), 6 (315–362 kHz), 7 (370–417 kHz) and
8 (424-471 kHz) would imply less dis u bances caused by he EVCP.
5.2. Time analysis
The measu emen s gi e ise o spec og ams ha a e composed o
many samples co esponding o each equency bin and ime ins an ,
which co e a wide ange o ampli ude alues. The e o e, due o
compu a ional issues, he analysis in he ime domain is pe o med o
sho e ime pe iods o 50 s leng h.
The analysis o he ime a iabili y o he dis u bances is o ganized as
ollows. Fi s , in sec ion V.B.1), he di e ences in he QP alues o he
ampli udes o he emissions a e compa ed o he esul an ime pe iods
o 50 s. Second, in V.B.2), he a ia ions o e ime gi en wi hin he i s
50 s pe iod o each eco ding a e modelled by means o a FFT analysis.
5.2.1. Analysis o he ime a iabili y wi hin he eco ding ime
The analysis o he ime a iabili y wi hin he eco ding ime is
based, i s , on he compa ison o he QP alues o he ampli ude o he
emissions calcula ed o e each pe iod o 50 s wi hin ha eco ding ime
(600 s, i.e. 12 pe iods, excep o EVCP2 a 75% SoC, whe e 250 s a e
a ailable, i.e. 5 pe iods). In Fig. 7, he QP alues o he ampli ude o he
emissions gene a ed by he cha ging p ocess o each EV o e each pe iod
o 50 s a e supe imposed, oge he wi h he PLC ou -o -band emissions
limi s.
Al hough he spec al o m o he emissions is main ained in he
whole equency band o he consecu i e pe iods o 50 s o he ou
Fig. 5. F equency shi o he emission gene a ed by EVCP2 a ound 350 kHz o
di e en s a es o cha ge.
Table III
TSHV and PFBL o each EVCP.
EVCP TSHV PFBL
EVCP1 112 dBµV 18%
EVCP2 75% 128 dBµV 97%
EVCP2 100% 121 dBµV 62%
EVCP3 131 dBµV 48%
Fig. 6. CDFs o he equency bins ha co espond o QP emission ampli udes
abo e he PLC ou -o -band emission limi s.
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
7
si ua ions unde s udy, sligh a ia ions in he ampli ude o he dis u -
bances can be obse ed a ce ain equencies. Fo his eason, in o de
o quan i y hese di e ences, he TSHV and PFBL a e calcula ed o he
12 pe iods o 50 s (5 pe iods in he case o EVCP2 a 75% SoC). Then, he
ollowing igu es a e ob ained in o de o cha ac e ize he maximum
di e ence o he TSHV (ΔTSHV) and PFBL (ΔPFBL) o each analyzed
si ua ion:
ΔTSHV =max
i∈{1,2,…n}TSHVi−min
i∈{1,2,…n}TSHVi(5)
ΔPFBL =max
i∈{1,2,…n}PFBLi−min
i∈{1,2,…n}PFBLi(6)
whe e n =12 excep o EVCP2 a 75% SoC, whe e n =5. In Table IV,
ΔTSHV and ΔPFBL alues o each EVCP a e ga he ed.
In he case o EVCP1, a maximum di e ence o he TSHV o 3 dB is
obse ed be ween di e en pe iods o he emissions gene a ed by i s
cha ging p ocess. In he emaining analyzed cases, hese di e ences do
no exceed 1 dB. Rega ding he PFBL, in he cases o EV1 and EV2 a 75%
SoC, maximum di e ences o 1% and less han 1% a e ob ained,
espec i ely. The di e ences in he PFBL o EV2 a 100% SoC and EV3,
in u n, inc ease up o 11% and 6%, espec i ely. Howe e , hese la ge
di e ences do no esul in g ea a ia ions in he TSHV in ha cases,
which implies ha he emissions exceeding occasionally he limi s do
no p esen high ampli ude. The e o e, i can be concluded ha , in
gene al, he dis u bances gene a ed by he EVCPs unde s udy p esen a
quasi-s a iona y beha io in pe iods anging om 50 s o se e al
minu es.
Apa om ha , al hough no explici ly shown in Table III, i is
impo an o highligh ha he TSHV o he las pe iod o 50 s o EV2 a
75% SoC (129 dBµV) is signi ican ly highe i compa ed o he TSHV
eco ded o he i s 50-second pe iod a 100% SoC (122 dBµV). This
means ha he emissions can subs an ially di e o di e en s a es o
cha ge. The e o e, despi e he ac ha i is no possible o de e mine
when hese ime a ia ions occu , i can be assumed ha a ia ions
Fig. 7. QP alues o he ampli ude o he emissions (dBµV) gene a ed by EVCP1 (a), EVCP2 a 75% (b) and 100% (c) SoC, and EVCP3 (d) in each pe iod o 50 s wi hin
he eco ding ime.
Table IV
ΔTSHV and ΔPFBL.
ΔTSHV ΔPFBL
EVCP1 3 dB 1%
EVCP2 a 75% SoC 1 dB 0%
EVCP2 a 100% SoC 0 dB 11%
EVCP3 1 dB 6%
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
8
gi en in pe iods g ea e han se e al minu es due o changes in he s a e
o cha ge a e o be expec ed. As EVs use cons an cu en /cons an
ol age modes o cha ging hei ba e ies [44], his e ec can be due o
he cha ging p o ile o hese wo modes. Ab up di e ences in he
emission ampli udes 30 minu es a e s a o he cha ging cycle ha e
been also epo ed in [16].
5.2.2. Analysis o he ime a iabili y wi hin 50 s
In he p e ious sec ion, i was concluded ha no ime a ia ions in
pe iods longe han 50 s a e occu ing, since simila TSHV and PFBL
we e ob ained in each 50-second pe iod wi hin he eco ding ime. Fo
his eason, ime a ia ions can only occu in pe iods o less han 50 s. In
o de o quan i y his po en ial ime a iabili y, a FFT is applied o he
ime samples co esponding o he i s 50-second pe iod o each e-
quency bin o he spec og am, ob aining he FFT equency componen s
o he ime a iabili y o each equency bin.
Wi h he aim o compa ing he ampli udes o he FFT equency
componen s co esponding o he di e en equency bins, he esul an
FFTs a e no malized wi h espec o he co esponding ampli ude o he
FFT componen a 0 Hz. As an example, in Fig. 8, he modulus o he
no malized FFT ob ained o he i s 50-second pe iod o he emissions
gene a ed by he cha ging p ocess o EV2 a 100% SoC a he equency
bin 383.901 kHz is p esen ed. Fig. 8 only shows he posi i e pa o he
FFT spec um, since he emissions gene a ed by he cha ging p ocess o
an EV a e eal signals, whose FFT is e en.
In he 39,284 analyzed equency bins (9821 equency bins/signal ⋅
4 signals), he modulus o he no malized FFTs we e simila o he one
shown in Fig. 8, whe e na owband FFT componen s a 50 Hz, 100 Hz,
150 Hz, 200 Hz and 250 Hz a e obse ed. This FFT pa e n co esponds
o a pe iodic unc ion wi h pe iod T =1/50 =20 ms [45].
The e o e, he no malized FFT, ega dless o he equency bin and
EVCP unde s udy, can be cha ac e ized by a simpli ied model composed
o he main componen s a 50 Hz, 100 Hz, 150 Hz, 200 Hz, and 250 Hz.
Fo each equency bin, he main ampli ude o hose componen s is
de ec ed using he no malized FFT in dB, as ep esen ed in Fig. 8. Then,
a simpli ied model is p oposed in linea scale acco ding o (7):
FFT ( FFT) = δ( FFT) + ∑
5
n=1
cn⋅δ( FFT −50n)(7)
whe e δ( FFT)is he uni sample unc ion.
In (7), c
n
a e 2ℼ imes he complex Fou ie coe icien s o he signal,
gi en in linea scale and no malized wi h espec o he 0 Hz FFT
componen .
Wi h he aim o de e mining how he o iginal signal esembles he
signal ob ained by app oxima ing he FFT wi h he p oposed simpli ied
model (syn hesized signal), he absolu e alue o he In e se FFT (IFFT)
o he impulse ain in (7) is calcula ed and exp essed in loga i hmic
scale. Fo his pu pose, he non-no malized coe icien s in (7) a e
conside ed. In Fig. 9, he o iginal and syn hesized emissions gene a ed
by he cha ging p ocesses o EVCP2 a 100% SoC a 255.951 kHz (a) and
EVCP3 a 10 kHz (b) a e shown du ing a ime pe iod o 0.11 s.
As i can be obse ed, he syn hesized signal app oxima es accu a ely
he ampli ude o he o iginal signal, as well as he peak- o-peak and he
shape o he a ia ion. This means ha he complex coe icien s o he
simpli ied model cnp o ide a quan i a i e cha ac e iza ion o he signal
a ia ion in he ime domain. The e o e, in o de o quan i y he o al
ime a ia ions o he emissions and ela e hei ime-dependen
beha io wi h hei spec al cha ac e is ics, he o al a iabili y o he
dis u bances is calcula ed, as shown in (8), as he sum in linea scale o
he modulus o he ampli ude o he FFT componen s a 50 Hz, 100 Hz,
150 Hz, and 200 Hz, which is hen con e ed o loga i hmic.
To al a iabili y (dB) = 20⋅log10(∑
4
n=1
|cn|)(8)
Fo he wo examples shown in Fig. 9, he o al a iabili y is -11.8 dB
o (a) and -7.9 dB o (b), so ha he highe he sum o he FFT com-
ponen s, he g ea e he a ia ion o he signal o e ime.
Fig. 10 shows he o al a iabili y o he dis u bances ob ained in
e ms o he FFT componen s, along wi h he QP alues o he ampli ude
o he emissions.
Simila ly as ob ained o he equency analysis, he ime a ia ions
o he ampli udes o he NIEs also depend on he speci ic cha ging
p ocess (EVCS, EV model, and s a e o cha ge). Finally, om Fig. 10 i
can be assumed ha , while he backg ound noise emains p ac ically
s a ic o e ime in he ou cha ging si ua ions unde s udy, highe
a ia ions occu a he equency bins co esponding o dis u bances,
bo h o na owband emissions and colo ed noise.
6. P opaga ion o he measu ed emissions
6.1. F equency analysis o he p opaga ion o he emissions
In o de o analyze he p opaga ion o he emissions in he equency
domain, he QP alues o he ampli ude o he emissions measu ed a
H2, H3, and H5 a e supe imposed when each EV model is cha ged
indi idually. Fig. 11 shows, o he h ee measu emen loca ions, he
emissions gene a ed by he cha ging p ocess o EV1 (a), EV2 a 75% (b)
and 100% (c) SoC, and EV3 (d).
As i can be seen, he emissions do no emain in he p oximi y o he
sou ce and hey p opaga e se e al me e s h ough he elec ical g id. In
mos o he cases, he spec al pa e ns o he dis u bances a e main-
ained, bu hey a e a enua ed wi h he dis ance. In o de o clea ly
obse e his beha io , Fig. 12 shows, as an example, a zoom o he QP
alues o he ampli ude o he emissions (dBµV) gene a ed by EVCP1
measu ed a H2, H3, and H5 in he equency ange 230-250 kHz.
I he emission a 240 kHz is aken in o accoun , a dec ease in he
ampli udes gene a ed by he cha ging p ocess o his EV model can be
obse ed when measu ing a dis ances o 46 m (POC o H3) and 120 m
(POC o H5). The gene a ed emission su e s an a enua ion o 5 dB
( om 79.5 dBµV o 74.5 dBµV) i he NIEs a e measu ed a H3, while he
a enua ion inc eases up o 10.7 dB ( om 79.5 dBµV o 68.8 dBµV) a
H5.
By con as , a ce ain equencies o he spec um when EV1 is
cha ging, highe emissions a e measu ed a loca ions di e en o he
POC o he sou ce. As an example, Fig. 13 p esen s he QP alues o he
ampli ude o he emissions gene a ed by EVCP1 a H2, H3, and H5 in he
equency ange 70-90 kHz. Despi e EV1 being cha ged a H2, he
componen a ound 80 kHz is g ea e when measu ed a H5 han when
measu ed a he same POC whe e he EV is cha ging.
Fig. 8. Modulus o he no malized FFT (dB) o he ime samples co esponding
o he emissions gene a ed by EVCP2 a 100% SoC a 383.901 kHz.
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
9
Fig. 9. Syn hesized and o iginal emissions gene a ed by he cha ging p ocesses o EVCP2 a 100% SoC a 255.951 kHz (a) and EVCP3 a 10 kHz (b).
Fig. 10. No malized ampli ude (dB) o he a iabili y o he emissions gene a ed by EVCP1 (a), EVCP2 a 75% (b) and 100% (c) SoC, and EVCP3 (d) in he 9-500 kHz
equency band du ing he i s 50 s o measu emen ime oge he wi h he QP alues o he ampli ude o he emissions.
J. Gonz´
alez-Ramos e al.
Elec ic Powe Sys ems Resea ch 231 (2024) 110289
10
As i is s a ed in [46],[47], he in oduc ion o mode n
ene gy-e icien appliances implies esonances in he g id access
impedance, which may in ol e signi ican inc eases in he emissions a
he swi ching equency and i s ha monics. This migh be he main cause
why, in spi e o he a enua ion su e ed by he gene a ed emissions due
o he dis ance, he ampli udes o he dis u bances a e highe a an
elec ical poin dis an om he sou ce o emissions. Hence, as i is
poin ed ou in [23],[25],[46–48], esonances in he g id impedance a e
a key aspec in he ampli ude o NIEs gene a ed by elec onic de ices. An
analysis o he in luence o he g id impedance on he p opaga ion o he
dis u bances is p esen ed in [49]. Re e ence [7] poin s ou ha a
high-impedance pa allel esonance causes high-ampli ude emissions a
he esonance equency, whe eas a dec ease in he dis u bance ampli-
udes is obse ed i he e is a se ies esonance wi h low impedance
alues.
As ca ied ou in sec ion V.A, in o de o nume ically cha ac e ize he
p opaga ion o he emissions, TSHV and PFBL alues a e calcula ed o
he dis u bances eco ded a each loca ion o each cha ging si ua ion
unde es (see Table V).
The esul s p esen ed in Table V lead o conclude ha highe TSHVs
a e ob ained a he POC o each house o which he EVCS unde s udy is
ins alled. This goes in line wi h he conclusions p esen ed om Fig. 12,
whe e, in gene al, a dec ease in he emissions wi h dis ance was
obse ed. The g ea es di e ence occu s when compa ing he TSHV
calcula ed o EVCP3 a H5, 131 dBµV, and H3, 124 dBµV (di e ence o
7 dB). Howe e , as concluded om Fig. 13, he dis ance is no he only
ac o a ec ing he ampli ude o he emissions, since a esonance in he
g id access impedance could imply an inc ease in he dis u bances a
ha equency. Fo his eason, simila o e en highe TSHVs a e
calcula ed a elec ical poin s mo e dis an om he sou ce o he
emissions han a nea by POCs. This beha io can be obse ed when
compa ing, o example, he TSHV o EVCP1 a H3 and H5. Fo his EV,
an iden ical TSHV is ob ained a bo h loca ions, despi e he ac ha H5
is 74 me e s a he away om H2 han H3. This e ec a ises in he ou
cha ging si ua ions unde s udy.
The PFBL calcula ed o EVCP2 a 75% SoC and EVCP3 is simila a
he h ee POCs whe e he emissions a e egis e ed. In he case o EVCP2
a 75% SoC, alues close o 100% (90%-100%) a e ob ained a he h ee
measu emen loca ions. The sligh di e ences a e due o he noise loo ,
which does no exceed he PLC ou -o -band emissions limi s a some
equencies when measu ing he dis u bances a H2 and H5. Simila ly,
his is he case o EVCP3, whe e 50% o he equency bins (app oxi-
ma ely up o 250 kHz) co espond o emission ampli udes abo e he
emission limi s a H2, H3, and H5. By con as , o EVCP2 a 100% SoC,
Fig. 11. Synch onized QP alues o he ampli ude o he emissions (dBµV) measu ed a H2, H3, and H5 gene a ed by EVCP1 when EV1 is cha ging a H2 (a), EVCP2
a 75% and 100% SoC when EV2 is cha ging a H3 (b), and EVCP3 when EV3 is cha ging a H5 (c).
J. Gonz´
alez-Ramos e al.
142
A.3. Jou nal Pape JP3
This subsec ion p esen s he ollowing jou nal pape :
J. González-Ramos, A. Galla e a, I. Fe nández, I. Angulo, A. A inda, D. de la
Vega, “Compa ison o Conduc ed Emissions Due o Elec ic Vehicle Cha ging
P ocesses unde Isola ed and On-Line Condi ions in he 9-500 kHz F equency
Range”, Sus ainable Ene gy, G ids and Ne wo ks, Volume 38, 2024, 101333, ISSN
2352-4677, h ps://doi.o g/10.1016/j.segan.2024.101333.
The mos ep esen a i e quali y indica o s conce ning his pape a e lis ed below:
Publishe : Else ie
Jou nal: Sus ainable Ene gy, G ids and Ne wo ks
Yea o publica ion: 2024
Type o publica ion: Jou nal pape indexed in JCR
A ea: Elec ical & Elec onic Enginee ing | Ene gy & Fuels
Ranking (JCR - 2023): 70/353 (Q1) | 75/171 (Q2)
Impac ac o (JCR - 2023): 4.8
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
A ailable online 28 Feb ua y 2024
2352-4677/© 2024 The Au ho (s). Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-
nc-nd/4.0/).
Compa ison o conduc ed emissions due o elec ic ehicle cha ging
p ocesses unde isola ed and on-line condi ions in he 9–500 kHz
equency ange
Jon Gonz´
alez-Ramos
a
,
*
, Alexande Galla e a
a
, Igo Fe n´
andez
a
, I zia Angulo
b
,
Da id de la Vega
a
, Amaia A inda
a
a
Depa men o Communica ions Enginee ing, Uni e si y o he Basque Coun y (UPV/EHU), Bilbao, Spain
b
Depa men o Applied Ma hema ics, Uni e si y o he Basque Coun y (UPV/EHU), Bilbao, Spain
ARTICLE INFO
Keywo ds:
Conduc ed non-in en ional emissions
Elec ic ehicle cha ging p ocess
Low ol age g id
Line impedance s abiliza ion ne wo k
ABSTRACT
This pape aims a compa ing he conduc ed emissions gene a ed by Elec ic Vehicle Cha ging P ocesses (EVCPs)
unde isola ed and on-line condi ions in he equency and ime domains, co e ing he 9–500 kHz equency
ange. The isola ed condi ions co espond o he use o a Line Impedance S abiliza ion Ne wo k (LISN), whe eas
measu emen s in he Low Vol age (LV) g id whe e he EV is connec ed a e e e ed o as on-line condi ions.
Rega ding he equency analysis, he esul s lead o conclude ha di e en ampli ude and spec al ea u es o
he Non-In en ional Emissions (NIEs) a e measu ed depending on he measu emen condi ions, egis e ing
highe -ampli ude NIEs when he measu emen s a e conduc ed in he LV g id. Mo eo e , he pape also shows
ha he spec al cha ac e is ics o he NIEs co espond o onal and na owband emissions a speci ic equencies
in he 9–150 kHz equency ange, while backg ound noise wi h low-ampli ude emissions (in he case o isola ed
condi ions) and colo ed noise (in he case o on-line condi ions) a e epo ed in he 150–500 kHz band. In he
ime domain, a pe iodic ime-dependen beha io wi hin he undamen al cycle o he mains (20 ms) is epo ed
in all he analyzed cases, excep o he onal emissions wi h oscilla ing cen al equency gene a ed by a speci ic
EV model. The ime a iabili y be ween bo h measu emen con igu a ions only di e s by mo e han 3 dB o
h ee ou o he wel e analyzed EVCPs, whe e a highe a iabili y is obse ed i he measu emen s a e ca ied
ou using a LISN. As a conclusion, conside ing he di e ences in he equency and ime cha ac e is ics o he
NIEs unde isola ed and on-line condi ions, e alua ing he emissions gene a ed by EVCPs migh lead o an un-
de es ima ion o he NIEs in eal LV g ids.
1. In oduc ion
In ecen yea s, he e has been a conside able inc ease in he numbe
o in e e -based de ices connec ed o he powe g id, such as Elec ic
Vehicles (EVs) cha ging in as uc u e o Pho o ol aic (PV) panels. This
kind o powe elec onic de ices gene a e Non-In en ional Emissions
(NIEs) a he swi ching equency o he in e e and i s mul iples [1,2].
Al hough he imp o emen o he e iciency o he powe con e sion has
led o a educ ion in he ampli ude o he dis o ion in he ha monic
equency ange (<2 kHz), he emission le els in he 2–500 kHz e-
quency band ha e signi ican ly inc eased [3]. These high-ampli ude and
ime- a ian dis o ions co e he whole Na owband Powe Line
Communica ions (NB-PLC) equency ange and can be a c i ical aspec
o communica ions using he Low Vol age (LV) g id in as uc u e [1,4,
5]. Mo eo e , hey can lead o Powe Quali y (PQ) issues, such as
equipmen mal unc ion, educ ion o he li e ime o he elec onic
Abb e ia ions: BPF, Band-Pass Fil e ; CDF, Cumula i e Dis ibu ion Func ion; EUT, Equipmen Unde Tes ; EV, Elec ic Vehicle; EVCS, Elec ic Vehicle Cha ging
S a ion; EVCP, Elec ic Vehicle Cha ging P ocess; FFT, Fas Fou ie T ans o m; LISN, Line Impedance S abiliza ion Ne wo k; LV, Low Vol age; NB-PLC, Na owband
Powe Line Communica ions; NIE, Non-In en ional Emission; PFBHOn-line, Pe cen age o equency bins in which he emissions a e highe when he measu emen is
ca ied ou unde on-line condi ions; PFBL, Pe cen age o equency bins exceeding he PLC ou -o -band emission limi s; POC, Poin o Connec ion; PQ, Powe
Quali y; PV, Pho o ol aic; RF, Radio equency; TSHV, To al Sup aha monic Vol age; QP, Quasi-Peak.
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (J. Gonz´
alez-Ramos).
Con en s lis s a ailable a ScienceDi ec
Sus ainable Ene gy, G ids and Ne wo ks
jou nal homepage: www.else ie .com/loca e/segan
h ps://doi.o g/10.1016/j.segan.2024.101333
Recei ed 30 No embe 2023; Recei ed in e ised o m 2 Feb ua y 2024; Accep ed 24 Feb ua y 2024
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
2
de ices, audible noise o ligh licke , among o he s [6,7].
Up o now, se e al con ibu ions ha e been published add essing he
cha ac e iza ion o he emissions gene a ed by EV cha ging p ocesses
(EVCPs) in di e en measu emen condi ions co e ing he equency
ange om 2 kHz o 150 kHz (also e e ed o as sup aha monic ange).
Some o hese wo ks ha e been ca ied ou in he LV dis ibu ion g id
[8–13], whe e a mo e ealis ic o e iew o he beha io o he dis u -
bances is o e ed. In his scena io, he in luence o o he connec ed de-
ices and he equency- and ime-dependen g id impedance canno be
dis ega ded [14]. O he s udies, such as [15–17], ha e add essed his
cha ac e iza ion in econs uc ed LV g ids wi h he aim o simula ing
eal g id condi ions in a con olled scena io. Finally, in o he publica-
ions [1,18], he EV is connec ed o a Line Impedance S abiliza ion
Ne wo k (LISN). This way, he eco ded emissions a e only due o he
EVCP unde s udy and a e e alua ed o a well-cha ac e ized imped-
ance. In his sense, hey can be conside ed as isola ed condi ions [15].
The e is a lack o s udies in he li e a u e in which a compa ison o
he emissions gene a ed by he same EVCPs unde bo h isola ed and LV
g id condi ions is add essed. In his con ex , his pape deals wi h he
cha ac e iza ion o he emissions gene a ed by wel e EVCPs,
compa ing he esul s ob ained o he measu emen s in he LV g id (on-
line condi ions) and when using a comme cially a ailable LISN (isola ed
condi ions). The s udy e alua es he dis u bances in he equency and
ime domains and co e s he 9–500 kHz equency band.
This pape is o ganized as ollows. In sec ion II, he me hodology o
he s udy is p esen ed. This includes he desc ip ion o he measu emen
scena ios o he eco ding o he emissions wi h and wi hou LISN, he
signal p ocessing applied o he eco ded emissions, and a summa y o
he empi ical da a ob ained. Sec ion III discusses he equency and ime
cha ac e iza ion o he emissions. In sec ion III. A, a compa ison o he
spec al ea u es o he dis u bances gene a ed by hese EVCPs a e
p esen ed, while sec ion III. B gi es an o e iew o hei ime-dependen
beha io . Finally, in sec ion IV, he main conclusions o his s udy a e
ga he ed.
2. Me hodology
2.1. Desc ip ion o he measu emen scena io
2.1.1. Measu emen scena io o he eco ding o he emissions unde
isola ed condi ions
The on-si e measu emen s we e ca ied ou on a comme cial Elec ic
Vehicle Cha ging S a ion (EVCS) ope a ing in mode 3 acco ding o IEC
61851–1 [19]. All he measu emen s p esen ed in his a icle we e
pe o med in monophasic mode. Wi h he aim o implemen ing a
con olled measu emen scena io, he se up was isola ed om he LV
g id by means o a LISN [20], a small ans o me (Polylux PD400 [21]
wi h a a ed powe o 4000 VA), and a band-pass il e (BPF) adjus ed o
he PLC equency band. Bo h he LISN and he ans o me ope a ed in
monophasic mode.
The EVCS was di ec ly connec ed o he Equipmen Unde Tes (EUT)
po o he LISN. The esul ing emissions we e measu ed om he
Radio equency (RF) po wi h a digi al oscilloscope, which was
con olled by a lap op cha ging a a powe s a ion. The oscilloscope
(Picoscope Se ies 5000 [22]) egis e ed he NIEs wi h a e ical eso-
lu ion o 16 bi s, using a sampling a e o 4.1667 MS/s o he equency
ange 9–500 kHz. As he measu emen s a e ob ained om he RF po , a
co ec ion ac o should be applied in he pos -p ocessing s age o
compensa e he inse ion loss o he LISN be ween he EUT and RF po s.
Typical alues o he inse ion loss o he LISN as a unc ion o equency
can be ound in [20]. In Fig. 1, a ep esen a ion o he measu emen
se up used o he eco ding o he NIEs using a LISN is shown.
2.1.2. Measu emen scena io o he eco ding o he emissions unde on-
line condi ions
As shown in Fig. 2, he measu emen s we e pe o med a a pa king
plaza, whe e he EVCS was di ec ly connec ed o he LV dis ibu ion
g id. As in sec ion II. A. 1), he measu emen sys em o he assessmen
o he NIEs is composed o an oscilloscope esponsible o egis e ing he
emissions wi h a e ical esolu ion o 16 bi s and a sampling a e o
4.1667 MS/s. In o de o a oid he measu emen sys em om being
damaged by he undamen al componen o he g id (220 V, 50 Hz), he
ol age p obe p esen ed in [23] is connec ed as a coupling ci cui a he
Poin o Connec ion (POC). This ol age p obe was designed o
measu ing he NIEs unde unknown g id impedance condi ions in he
equency ange unde analysis.
2.2. Signal pos -p ocessing
Once he dis u bances a e measu ed, a Lanczos window wi h a
du a ion o 20 ms and an o e lapping o 90% is applied o he eco ded
signals [24]. Subsequen ly, in o de o cha ac e ize he dis u bances in
he equency domain, he Quasi-Peak (QP) alues o he ampli ude o
he emissions a e ob ained acco ding o he CISPR 16–1–1 s anda d
[25]. Mo e de ailed in o ma ion abou he signal pos -p ocessing s age
can be ound in [16].
2.3. Summa y o he eco ded emissions
The analysis p esen ed in his pape is based on he dis u bances
gene a ed by i e comme cially a ailable EV models. Wi h he aim o
Fig. 1. Measu emen scena io o he eco ding o he NIEs gene a ed by EVs du ing hei cha ging p ocess using a LISN (isola ed condi ions).
Fig. 2. Measu emen scena io o he eco ding o he NIEs gene a ed by EVs
du ing hei cha ging p ocess in he LV g id (on-line condi ions).
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
3
analyzing he po en ial in luence o he cha ging cu en on he eco ded
emissions, he di e en cha ging cu en s suppo ed by he EVCS a e
sequen ially con igu ed o each EV model unde s udy. In he case o
EV1, only one cha ging cu en was possible (16 A), whe eas wo
cha ging cu en s we e possible o EV4 (12 A, 16 A). Fo he es o he
EV models unde es , h ee cha ging cu en s we e e alua ed (8 A,
12 A, and 16 A). The e o e, he s udy add esses he cha ac e iza ion o
he emissions gene a ed by wel e di e en EVCPs (combina ions o EV
model and cha ging cu en ) wi h and wi hou using a LISN. This
cha ac e iza ion conside s eco dings o 600 s o du a ion, excep o
EV3 wi h a cha ging cu en o 8 A unde isola ed condi ions, o which
only 150 s we e eco ded.
3. Resul s
3.1. F equency cha ac e iza ion
The spec al cha ac e iza ion o he emissions in he 9–500 kHz
equency band, add essed in sec ion III. A. 1), is based on he p ocedu e
p oposed by he cu en au ho s in [16]. Fi s , he QP alues o he
ampli ude o he emissions co esponding o he i s 50 s o measu e-
men a e calcula ed and ep esen ed acco ding o CISPR 16–1–1 s an-
da d [19]. Since no speci ic emission limi s a e de ined o he cha ging
o EVs in his equency ange, he PLC ou -o -band emission limi s
de ined in EN 50065–1 [20] a e included o compa ison pu poses.
Then, he To al Sup aha monic Vol age (TSHV), he pe cen age o e-
quency bins exceeding he PLC ou -o -band emission limi s (PFBL), and
he Cumula i e Dis ibu ion Func ion (CDF) o he equency bins co -
esponding o dis u bances abo e hese limi s a e calcula ed. De ailed
in o ma ion abou he de ini ion and calcula ion o hese pa ame e s can
be ound in [16].
Mo eo e , in o de o ho oughly analyze he beha io o he emis-
sions in he anges 9–150 kHz and 150–500 kHz, sepa a e analyses o
each equency band a e p esen ed in sec ions III. A. 2) i) and III. A. 2)
ii).
3.1.1. Compa ison o he emissions gene a ed by EVCPs in he 9–500 kHz
equency band
As an example, in Fig. 3, he QP alues o he ampli ude o he
emissions gene a ed by EV2 and EV3 using a cha ging cu en o 8 A o
he i s pe iod o 50 s o eco ding a e p esen ed unde isola ed and on-
line condi ions.
Fig. 3 clea ly shows ha bo h he spec al shape and he ampli ude o
he emissions depend on he measu emen condi ions. The dis u bances
measu ed using a LISN a e lowe in p ac ically he whole equency
band, excep o emissions occu ing a ce ain equencies ( o example,
a ound 45 kHz o EVCP3 wi h a cha ging cu en o 8 A). I should be
no ed ha , unlike when he emissions a e eco ded di ec ly in he LV
g id, he dis u bances measu ed using a LISN do no gene ally exceed
he PLC ou -o -band limi s.
Acco ding o Fig. 3, he e a e di e ences in he emissions eco ded
o di e en models o EV, e en hough he same EVCS was used.
Howe e , simila beha io s can be obse ed. In he 9–150 kHz e-
quency ange, EVCPs gene a e onal emissions ( o example, a ound
45 kHz and 90 kHz o EVCP3 and a cha ging cu en o 8 A) o
na owband emissions ( o ins ance, be ween 45 kHz and 50 kHz and
be ween 90 kHz and 98 kHz o EVCP2 and a cha ging cu en o 8 A).
These emissions show simila ampli ude in bo h measu emen condi-
ions. Since he na owband emissions migh be caused by a pe iodically
changing swi ching equency [17], a ime analysis is equi ed (see
sec ion III. B). By con as , bo h he ampli ude and he spec al shape o
he backg ound emissions a y conside ably in his equency band i
isola ed and on-line condi ions a e compa ed.
In he 150–500 kHz equency ange, la and low-ampli ude dis-
o ions (40 dBµV, app oxima ely) a e measu ed when a LISN is used o
bo h EVCPs. Unde on-line condi ions, in u n, colo ed noise dec easing
wi h equency ( om 80 dBµV o 55 dBµV o EV2 and om 70 dBµV o
50 dBµV o EV3) is eco ded.
In o de o show ha he spec al pa e n o he emissions is simila
o he 12 EVCPs unde s udy, in Fig. 4, he QP alues o he ampli ude o
he emissions gene a ed by he wel e analyzed cases a e p esen ed
unde isola ed (a) and on-line (b) condi ions.
Fig. 4 shows ha he beha io obse ed o EV2 and EV3 wi h a
cha ging cu en o 8 A in Fig. 3 applies o he wel e EVCPs unde
s udy. The e o e, hese esul s lead o conclude ha he unde s anding
o he ampli ude and spec al ea u es o he emissions gene a ed by
EVCPs om 9 kHz o 150 kHz, he equency ange ha is no mally
add essed in he li e a u e, canno be ex apola ed o he 150–500 kHz
equency band. This highe equency ange is cu en ly being used in
No h Ame ica o PLC ansmissions and i is expec ed o be also
adop ed in Eu ope in he coming yea s. Fo his eason, u he s udies
conside ing hese highe equencies a e s ill needed.
In o de o gi e an insigh in o he ampli ude o he emissions in he
9–500 kHz equency band in all he analyzed cases and pe o m a
quan i a i e compa ison be ween bo h measu emen con igu a ions
(wi h and wi hou LISN), he TSHV and PFBL a e calcula ed o all he
(a) (b)
Fig. 3. QP alues o he ampli ude o he emissions (dBµV) gene a ed by EV2 (a) and EV3 (b) wi h a cha ging cu en o 8 A unde isola ed and on-line condi ions.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
4
EVCPs unde s udy as de ined in [16] (see Table 1).
As shown in Table 1, a highe TSHV can be measu ed wi h o wi hou
LISN depending on he EV model and he cha ging cu en unde anal-
ysis. Fo ins ance, EV2 wi h a cha ging cu en o 16 A shows a lowe
TSHV when measu ing wi h a LISN (113 dBµV and 135 dBµV when he
measu emen s a e pe o med wi h and wi hou LISN, espec i ely). Fo
EV3 wi h a cha ging cu en o 8 A, in u n, a highe THSV is ob ained
when a LISN is used o he eco ding o he emissions (132 dBµV when a
LISN is used, and 125 dBµV when he measu emen is pe o med
di ec ly in he LV g id). Due o he de ini ion o he TSHV (in linea
scale), high-ampli ude emissions occu ing a ce ain equencies can
con ibu e signi ican ly o he alue o his pa ame e , in such a manne
ha hey can cause he TSHV o be highe e en i he NIEs a e consid-
e ably lowe in he emaining equency bins. This phenomenon can be
clea ly obse ed i he PFBL is analyzed. Fo all he EVCPs unde s udy,
PFBLs lowe han 4% a e ob ained when a LISN is used. By con as ,
when he measu emen s a e pe o med in he LV g id, conside ably
highe PFBLs a e measu ed, being up o 100% o some EVCPs. This
means ha , when a LISN is used, he e a e high-ampli ude emissions a
speci ic equency bins ha gi e ise o TSHV alues highe han he
ones ob ained when he measu emen is pe o med in he LV g id,
al hough in his measu emen con igu a ion mos o he emissions
exceed he ampli ude o he NIEs eco ded wi h LISN. In o de o show
his e ec , he pe cen age o equency bins in which he emissions a e
highe when he measu emen is ca ied ou unde on-line condi ions
wi h espec o he isola ed condi ions is calcula ed (PFBHOn-line).
Mo eo e , in o de o quan i y he di e ence in ampli ude be ween bo h
measu emen condi ions, he di e ence o he QP alues o he ampli-
ude o he emissions is calcula ed o each equency bin
i
acco ding o
(1).
ΔQP( i) = QPOn−line( i) − QPIsola ed( i)(1)
In Table 2, he minimum, median, and maximum alues o he di -
e ence o each equency bin ΔQP( i)a e ga he ed o each EVCP unde
s udy, oge he wi h he PFBHOn-line.
Table 2 shows ha , in all he cases unde s udy, a leas in a 99% o
he equency bins, he emissions gene a ed by an EVCP when
measu ing in he LV g id exceed he dis o ions measu ed using a LISN.
Mo eo e , Table 2 also poin s ou ha he median di e ence is highe
han 11 dB in all he analyzed cases, which implies ha in 50% o he
equency bins, he NIEs measu ed in he LV g id a e a leas 11 dB
highe han he ones eco ded using a LISN. I he maximum di e ences
be ween bo h con igu a ions a e conside ed, alues g ea e han 30 dB
a e epo ed, wi h a maximum di e ence o 49 dB o EV2 wi h a
cha ging cu en o 16 A. Rega ding he minimum alues, i should be
men ioned ha nega i e di e ences a e measu ed o en ou o he
wel e analyzed EVCPs. These nega i e alues indica e ha he emis-
sions in hose equency bins a e highe i a LISN is used. A maximum
nega i e di e ence o −19 dB is egis e ed o EV5 wi h a cha ging
cu en o 16 A.
The p e ious analysis only conside s he ampli ude o he emissions,
wi hou aking in o accoun hei spec al dis ibu ion in he 9–500 kHz
equency band. Fo his eason, ollowing he p ocedu e p esen ed in
[16], Fig. 5 shows he CDFs o he equency bins in which he QP alues
o he ampli ude o he emissions exceed he PLC ou -o -band emission
o he wel e EVCPs when measu ing unde isola ed (a) and on-line (b)
condi ions.
Fig. 5(a) shows ha he equency bins in which he QP alues o he
ampli ude o he emissions exceed he PLC ou -o -band emission limi s
when measu ing wi h a LISN a e concen a ed in he equency ange up
o, app oxima ely, 150 kHz. This means ha , abo e his equency, he
emissions do no exceed he PLC ou -o -band limi s o his measu emen
condi ion. When he measu emen is pe o med di ec ly in he LV g id,
in u n, as p esen ed in Fig. 5(b), he spec al dis ibu ion o he highe -
ampli ude emissions is sp ead o e a highe equency ange. In ou o
he wel e analyzed cases, emissions exceeding he limi s a e epo ed in
he whole equency ange. The e o e, he 150–500 kHz equency
ange seems o be mo e app op ia e o PLC ansmissions i he
ampli ude o he emissions measu ed wi h a LISN a e aken in o
conside a ion. Howe e , due o he di e en beha io obse ed unde
on-line condi ions, i can be concluded ha he planning o PLC ne -
wo ks should no only be based on isola ed obse a ions using a LISN.
3.1.2. Compa ison o he emissions in he 9–150 kHz and 150–500 kHz
equency bands
In Fig. 4, i was shown ha he spec al ea u es o he NIEs depend
conside ably on he measu emen condi ions and he equency band. In
he 9–150 kHz equency band, onal o na owband emissions a spe-
ci ic equencies we e epo ed, while backg ound noise wi h low-
ampli ude emissions (wi h LISN) o colo ed noise (wi hou LISN) we e
egis e ed in he 150–500 kHz equency ange. This sec ion aims o gi e
a mo e in-dep h e alua ion o he di e ences occu ing be ween isola ed
and on-line condi ions depending on he equency ange unde analysis
(9–150 kHz and 150–500 kHz).
Fig. 4. QP alues o he ampli ude o he emissions (dBµV) gene a ed by he wel e EVCPs unde s udy conside ing he eco dings unde isola ed (a) and on-line
(b) condi ions.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
5
3.1.2.1. 9–150 kHz equency band. This sec ion aims a compa ing he
ampli ude o he onal and na owband emissions egis e ed be ween
bo h measu emen con igu a ions. Since he ield ials conduc ed wi h a
LISN isola e he NIEs gene a ed by he EVCP i sel , wi hou being
a ec ed by ex e nal sou ces, his compa ison may shed ligh on how
hese emissions beha e when he EV is cha ging unde eal LV g id
condi ions.
Fo his pu pose, his s udy ocuses on he NIEs wi h an ampli ude
exceeding 50 dBµV and a p ominence o 5 dB. The p ominence measu es
how much a ce ain na owband emission s ands ou due o i s in insic
heigh and i s loca ion ela i e wi h espec o o he na owband
dis u bances.
In o de o conduc his analysis, i s , he emissions ul illing he
p e iously men ioned condi ions in he measu emen s pe o med using
a LISN a e iden i ied. These emissions a e ep esen ed by means o a ed
ci cle in Fig. 6(a). Then, he same p ocedu e is applied o he NIEs
eco ded di ec ly in he LV g id. The emissions de ec ed in he same
equency bins unde bo h measu emen condi ions a e ep esen ed by
means o a ed ci cle in Fig. 6(b). I , by con as , no onal o na owband
emission is de ec ed when eco ding he emissions wi hou LISN in a
equency bin whe e an emission was p esen unde isola ed condi ions,
i is indica ed using a black c oss. An example o his can be obse ed in
Fig. 6(b) o EV5 wi h a cha ging cu en o 8 A.
Fig. 6(a) shows ha wen y onal o na owband emissions ul ill he
es ablished condi ions when eco ding he NIEs gene a ed by EV5 wi h a
cha ging cu en o 8 A wi h a LISN in he equency band om 9 kHz o
150 kHz. By con as , in he measu emen ca ied ou in he LV g id (see
Fig. 6(b)) only h ee ou o hese wen y emissions (115 dBµV a 44 kHz,
92 dBµV a 89 kHz, and 71 dBµV a 133 kHz) a e iden i ied as onal o
na owband NIEs. Thus, i seems ha mos o he emissions gene a ed by
EVCPs occu ing in isola ed condi ions canno be iden i ied when he
measu emen is conduc ed di ec ly in he LV g id. In o de o p o e i
his beha io is obse ed o all he EVCPs, he pe cen age o he
emissions ha occu in bo h measu emen si ua ions (common emis-
sions), as well as he equency (kHz), wid h a −10 dB (kHz), and
ampli ude (dBµV) o hese NIEs when measu ing wi h and wi hou LISN
a e ga he ed in Table 3 o he wel e EVCPs unde s udy.
As concluded om Fig. 6, no all he emissions de ec ed wi h LISN
Table 1
TSHV (dBµV) and PFBL (%) o each EVCP unde isola ed (Is.) and on-line (O-L) condi ions.
EV1 EV2 EV3 EV4 EV5
16 A 8 A 12 A 16 A 8 A 12 A 16 A 12 A 16 A 8 A 12 A 16 A
Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L
TSHV (dBµV) 109 123 136 132 137 133 113 135 132 125 109 121 113 125 110 118 111 118 133 124 110 117 113 119
PFBL (%) 1 33 3 100 4 100 3 100 2 35 1 33 1 95 0 12 0 13 2 19 1 12 1 28
Table 2
Pe cen age o equency bins in which he emissions gene a ed by he EVCPs a e highe unde on-line han isola ed condi ions, oge he wi h he minimum, median,
and maximum di e ences be ween he QP alues o he ampli ude o he emissions eco ded unde on-line and isola ed condi ions.
EV model PFBHOn-line (%) ΔQP (
i
) (dB)
Min. Median Max.
8 A 12 A 16 A 8 A 12 A 16 A 8 A 12 A 16 A 8 A 12 A 16 A
EV1 - - 100.00 - - 4 - - 26 - - 42
EV2 99.01 99.02 99.21 -6 -6 -2 26 26 36 38 39 49
EV3 99.70 99.90 99.99 -15 -4 0 17 24 29 32 37 42
EV4 - 99.92 99.91 - -7 -7 - 18 13 - 34 30
EV5 99.50 99.80 99.47 -14 -4 -19 11 16 21 30 32 36
Fig. 5. CDFs o he equency bins in which he QP alues o he ampli ude o he emissions exceed he PLC ou -o -band emission limi s when measu ing unde
isola ed (a) and on-line (b) condi ions.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
6
a e also de ec ed unde eal condi ions in he LV g id. In all he analyzed
cases, he maximum pe cen age o common emissions is 50.0%, which is
gi en o EV3 wi h a cha ging cu en o 12 A, and EV3 and EV5 wi h a
cha ging cu en o 12 A (see Table 3).
Rega ding he di e ence in ampli ude be ween he NIEs in bo h
measu emen condi ions, in six ou o he eigh een analyzed emissions, a
highe ampli ude is epo ed when a LISN is used. In hese cases, a
highes di e ence o 11 dB is gi en a 44 kHz o EV5 wi h a cha ging
cu en o 8 A. In he emaining cases, in u n, he ampli ude o he NIE
is highe in he LV g id, wi h a g ea es di e ence o 20 dB occu ing a
45 kHz o EV2 wi h a cha ging cu en o 16 A. Thus, he esul s do no
show a clea end in e ms o he ampli ude o he onal o na owband
emissions when measu ed in he LV g id compa ed o measu emen s
pe o med using a LISN.
In ou EVCPs unde s udy, onal o na owband emissions exceeding
120 dBµV a e epo ed when measu ing unde isola ed condi ions,
which a e se e al dB lowe in ampli ude when de ec ed unde on-line
condi ions. These cases showed a highe TSHV when he measu emen
is ca ied ou using a LISN (see Table 1 and Table 3, EV2 wi h a cha ging
cu en o 8 A and 12 A, EV3 wi h a cha ging cu en o 8 A, and EV5
wi h a cha ging cu en o 8 A). The e o e, hese high-ampli ude NIEs
co espond o he emissions wi h a majo con ibu ion o he calcula ion
o his pa ame e , ega dless o he beha io obse ed o he emaining
equency bins.
The di e ences obse ed o hose emissions ha a e de ec ed as
onal o na owband NIE wi h LISN, bu do no ul ill he c i e ia o be
de ec ed when eco ded in he LV g id, ha e been also analyzed. Fo his
pu pose, as in sec ion III. A. 1), he minimum, mean, and maximum
alues o he di e ences be ween he emissions measu ed wi hou and
wi h LISN o each equency bin a e calcula ed o each EVCP unde
s udy. Table 4 ga he s he minimum, mean, and maximum di e ences
be ween bo h si ua ions, oge he wi h he numbe o emissions de ec ed
as onal o na owband emission when he eco ding is pe o med wi h
LISN, bu no iden i ied as so i he measu emen is pe o med in he LV
g id.
In gene al, he ampli ude o he non-common emissions be ween
bo h measu emen con igu a ions is conside ably highe when he
measu emen is pe o med wi hou a LISN. In hese cases, he
(a) (b)
Fig. 6. QP alues o he ampli ude o he emissions (dBµV) gene a ed by EV5 wi h a cha ging cu en o 8 A unde isola ed (a) and on-line (b) condi ions.
Table 3
F equency (kHz), ampli ude (dBµV) o he common emissions eco ded unde isola ed and on-line condi ions in he 9–150 kHz equency ange, as well as he pe -
cen age o common emissions de ec ed wi h espec o he o al emissions iden i ied unde isola ed condi ions.
EV
model
Cha ging
cu en
On-line emissions /
Isola ed emissions
Pe cen age o common emissions
in bo h con igu a ions (%)
F equency
(kHz)
Wid h a
−10 dB (kHz)
Ampli ude unde
isola ed condi ions
(dBµV)
Ampli ude unde on-line
condi ions (dBµV)
EV1 16 A 1/11 9.1 130 0.4 69 76
EV2 8 A 2/14 14.3 45 4.3 121 117
90 8.5 91 93
12 A 2/16 12.5 45 4.3 123 118
90 8.3 93 94
16 A 2/6 33.3 45 4.3 80 120
90 8.1 97 96
EV3 8 A 2/13 15.4 44 0.5 125 116
89 0.8 91 92
12 A 1/2 50.0 89 0.8 89 92
16 A 1/5 20.0 89 0.9 86 88
EV4 12 A 1/20 5.0 33 0.7 92 93
16 A 1/10 10.0 33 0.7 93 93
EV5 8 A 3/20 15.0 44 0.5 126 115
89 0.4 91 92
134 0.6 59 71
12 A 1/3 33.3 89 0.4 91 93
16 A 1/5 20.0 44 0.4 95 89
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
7
backg ound emission o he on-line scena io, in which emissions due o
EVCPs a e combined wi h po en ial NIEs om o he sou ces, b ings EV
emissions o be unde ec able. This beha io can be obse ed i he me-
dian di e ence is aken in o conside a ion, since, in all he analyzed
cases, posi i e di e ences anging om 1 dB o 25 dB a e measu ed. I
should be men ioned ha , in six ou o he wel e cases, a maximum
di e ence be ween bo h con igu a ions highe han 20 dB is epo ed.
Finally, i he minimum di e ences a e conside ed, i can be obse ed
ha he e a e also some cases whe e emissions wi h a highe ampli ude
when measu ing using a LISN a e egis e ed.
3.1.2.2. 150–500 kHz equency band. As p e iously men ioned, in he
150–500 kHz equency ange, la backg ound wi h low-ampli ude
na owband emissions a e eco ded when a LISN is used, whe eas
colo ed noise is egis e ed i he measu emen is conduc ed in he LV
g id. In Fig. 4, i was shown ha he emissions in his equency band
when measu ing in he LV g id conside ably exceed he dis u bances
eco ded i a LISN is used. In o de o quan i y he exis ing di e ences
be ween bo h measu emen condi ions in his equency band, he di -
e ence be ween he QP alues o he ampli ude o he emissions wi hou
and wi h LISN a e calcula ed (ΔQP(
i
) o 150 <=
i
(kHz) <=500). In
Fig. 7, a boxplo o hese di e ences o each EVCP unde s udy is
shown. In his igu e, he ed line indica es he median di e ence alue,
while he bo om and op edges ep esen he 25 h and 75 h pe cen iles
o he di e ences. The whiske s indica e he mos ex eme da a poin s
no conside ed ou lie s. Finally, he ou lie s a e depic ed by he ed “+”
symbol.
Fig. 7 shows median alues highe han 7.7 dB ega dless o he
EVCP unde es . In se en ou o he wel e analyzed EVCPs, he median
di e ence exceeds 20 dB, wi h a maximum median alue o 36.6 dB o
EV2 wi h a cha ging cu en o 16 A. The maximum di e ence, 48.3 dB,
is also epo ed o his EVCP a 150.3 kHz.
The e o e, as i was shown in Fig. 4, he ampli ude o he emissions
measu ed in he LV g id in he 150–500 kHz equency band is consid-
e ably highe han he ones eco ded using a LISN, ega dless o he EV
model and cha ging cu en unde es . This implies ha he measu e-
men s pe o med unde isola ed condi ions migh no be ep esen a i e
o he emissions gene a ed by EVCPs in a eal si ua ion in his equency
band.
3.2. Time cha ac e iza ion
In o de o analyze he ime-dependen beha io o he dis u bances
du ing he 600 s o eco ding ime, in sec ion III. B. 1), he QP alues o
he ampli ude o he emissions in each pe iod o 50 s a e calcula ed and
ep esen ed. Then, wi h he aim o s udying he ime a iabili y wi hin
each pe iod o 50 s, in sec ion III. B. 2), a Fas Fou ie T ans o m (FTT)
analysis is pe o med.
3.2.1. Analysis o he ime a iabili y wi hin he eco ding ime (600 s)
In Fig. 8, as an example, he QP alues o he ampli ude o he
emissions gene a ed by EV2 wi h a cha ging cu en o 8 A a e shown o
he wel e 50-second pe iods (600 s in o al) when he measu emen is
pe o med using a LISN (a) and in he LV g id (b).
Fig. 8(a) shows ha he di e ences in he QP alues o he ampli ude
o he emissions gene a ed by EV2 be ween consecu i e 50-second pe-
iods a e negligible when a LISN is used. By con as , in Fig. 8(b), wo
di e en s a es a e epo ed when he eco ding is pe o med in he LV
g id. Al hough he spec al ea u es o he dis u bances a e simila in bo h
s a es excep o he emission occu ing a ound 45 kHz, he ampli ude o
he NIEs a ies conside ably in he whole equency band. I should also
be no ed ha a d as ic change in he PFBL can be obse ed be ween hese
wo clea ly di e en ia ed s a es. While in he lowe -ampli ude s a e he
dis o ions a e below he PLC ou -o -band emission limi s in he majo i y
o he equency bins, hese limi s a e exceeded in he whole equency
band in he highe -ampli ude s a e. In he speci ic case shown in Fig. 8(b),
he emissions co espond o he highe - and lowe -ampli ude s a es in i e
and se en 50-second pe iods, espec i ely.
Wi h he aim o quan i ying he a ia ion o he emissions wi hin he
600 s and e alua ing i he ime a ia ions shown in Fig. 8 a e epo ed
o o he EVCPs, he maximum di e ence be ween he TSHV and PFBL
o he wel e pe iods o 50 s wi hin 600 s is calcula ed o each EVCP
unde s udy acco ding o (2) and (3) [16]:
ΔTSHV =max
i∈{1,2,…n}TSHVi−min
i∈{1,2,…n}TSHVi(2)
ΔPFBL =max
i∈{1,2,…n}PFBLi−min
i∈{1,2,…n}PFBLi(3)
whe e n is he numbe o 50-second pe iods a ailable o each mea-
su emen (n=12, excep o EV3 wi h a cha ging cu en o 8 A unde
isola ed condi ions, o which n=3).
Table 4
Numbe o emissions, minimum, mean, and maximum di e ence o he emis-
sions iden i ied in he 9–150 kHz equency band as na owband emissions
unde isola ed condi ions and ha a e no de ec ed unde on-line condi ions.
EV
model
Cha ging
cu en
No
de ec ed
on-line
emissions /
Isola ed
emissions
Min. ΔQP
(
i
) o he
emissions
no de ec ed
unde on-
line
condi ions⋅
(dB)
Median
ΔQP(
i
) o
he
emissions
no de ec ed
unde on-
line
condi ions⋅
(dB)
Max. ΔQP
(
i
) o he
emissions
no de ec ed
unde on-
line
condi ions⋅
(dB)
EV1 16 A 10/11 4 14 29
EV2 8 A 12/14 13 22 29
12 A 14/16 3 17 30
16 A 4/6 18 25 31
EV3 8 A 11/13 -3 11 18
12 A 1/2 2 2 2
16 A 4/5 4 7 8
EV4 12 A 19/20 -7 9 26
16 A 9/10 -7 7 21
EV5 8 A 17/20 -7 6 18
12 A 2/3 -4 1 6
16 A 4/5 -19 2 11
Fig. 7. Boxplo o he di e ences o he QP alues o he ampli ude o he
emissions unde on-line and isola ed condi ions (dB) in he 150–500 kHz e-
quency band.
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
8
The esul s ga he ed in Table 5 show ha he emissions measu ed in
a LISN ba ely a y wi h ime in pe iods g ea e han 50 s. Fo his
measu emen condi ion, a maximum di e ence in he TSHV and PFBL o
1 dB and 1%, a e epo ed, espec i ely, which can be conside ed
negligible. Howe e , as obse ed in Fig. 8(b) o EV2 wi h a cha ging
cu en o 8 A, impo an a ia ions wi h ime can be obse ed in ou
ou o he wel e analyzed cases i he measu emen is pe o med
di ec ly in he LV g id. In hese cases, di e ences in he TSHV and PFBL
be ween wo 50-second pe iods up o 26 dB and 97% a e measu ed,
espec i ely.
Table 5 also shows ha hese signi ican ime a ia ions mainly occu
o EV2 unde on-line condi ions, ega dless o he con igu ed cha ging
cu en . Fo his eason, in o de o ho oughly analyze he ime-
dependen beha io o he emissions gene a ed by his EV model, in
Table 6, he TSHV (dBµV) and PFBL (%) calcula ed o each 50-second
pe iod o EV2 wi h cha ging cu en s o 8 A, 12 A, and 16 A a e
ga he ed.
Fo he cha ging cu en s o 8 A and 12 A, a d as ic dec ease in he
ampli ude o he dis u bances is egis e ed in he eigh h and six h 50-
second pe iod, espec i ely. In hese cases, he TSHV d ops om 132
o 133 dBµV and 133–134 dBµV o 119 dBµV, in ol ing a conside able
educ ion o he PFBL om 100% o 12–13%. I should be no ed ha , o
hese cha ging cu en s, once he TSHV and PFBL dec ease in a ce ain
50-second pe iod, hey do no ake highe alues a subsequen ime
ins an s du ing he 10 minu es.
Fo he cha ging cu en o 16 A, in u n, a di e en beha io is
obse ed. The ampli ude o he emissions a ies om one s a e o
ano he in di e en pe iods o 50 seconds, and can inc ease and
dec ease depending on he 50-second pe iod conside ed. Fo example, in
he i s and ou h pe iods, a TSHV and PFBL o 135 dBµV and 100% a e
measu ed, while, in he second and hi d pe iods, hese pa ame e s ake
alues o 109 dBµV and 3%, espec i ely. Thus, he e seems o be
di e en s a es in he EVCP o EV2 ha gi e ise o di e ences in he
ampli ude o he emissions, which means ha he ampli ude o he NIEs
gene a ed by his EVCP ises and alls in di e en ime in e als wi hin
10 minu es. In o de o gi e an insigh in o he a ia ions o he NIEs
om one s a e o ano he , as an example, in Fig. 9, he spec og am o
he emissions gene a ed by EV2 wi h a cha ging cu en o 16 A unde
on-line condi ions a e depic ed du ing he i s 50 seconds o eco ding.
In Fig. 9, sha p changes in he ampli ude o he emissions gene a ed
by EV2 wi h a cha ging cu en o 16 A when measu ing unde on-line
condi ions a e shown. This can be clea ly obse ed i he high-
ampli ude emission occu ing a ound 40–50 kHz om 22 seconds on-
wa ds is conside ed, which does no appea in he i s 22 seconds o he
Fig. 8. QP alues o he ampli ude o he emissions gene a ed by EV2 wi h a cha ging cu en o 8 A in each pe iod o 50 s du ing he eco ding ime unde isola ed
(a) and on-line (b) condi ions.
Table 5
ΔTSHV and ΔPFBL o each EVCP unde isola ed (Is.) and on-line (O-L) condi ions.
EV1 EV2 EV3 EV4 EV5
16 A 8 A 12 A 16 A 8 A 12 A 16 A 12 A 16 A 8 A 12 A 16 A
Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L Is. O-L
ΔTSHV (dB) 1 0 0 14 0 15 0 26 0 0 0 0 0 0 0 0 1 0 1 8 0 0 0 1
ΔPFBL (%) 0 1 1 87 1 88 1 97 0 0 0 1 0 3 0 0 1 2 0 7 0 0 0 2
Table 6
TSHV (dBµV) and PFBL (%) o each 50-second pe iod o EV2 wi h cha ging
cu en s o 8 A, 12 A, and 16 A unde on-line condi ions.
Pe iod o
50 s
EV2
8 A 12 A 16 A
TSHV
(dBµV)
PFBL
(%)
TSHV
(dBµV)
PFBL
(%)
TSHV
(dBµV)
PFBL
(%)
Pe iod 1 132 100 133 100 135 100
Pe iod 2 132 100 134 100 109 3
Pe iod 3 132 100 134 100 109 3
Pe iod 4 132 100 134 100 135 100
Pe iod 5 132 100 134 100 119 12
Pe iod 6 133 100 119 13 119 12
Pe iod 7 133 100 119 12 119 13
Pe iod 8 119 13 119 12 119 12
Pe iod 9 119 13 119 12 119 12
Pe iod 10 119 13 119 13 119 12
Pe iod 11 119 13 119 13 119 12
Pe iod 12 119 13 119 12 119 12
J. Gonz´
alez-Ramos e al.
Sus ainable Ene gy, G ids and Ne wo ks 38 (2024) 101333
9
eco ding. This implies ha signi ican ime a ia ions can occu
ab up ly wi hin 50 seconds o his EV model.
I should be no ed ha Fig. 9 also shows ha he na owband
emissions a ound 45–50 kHz and 90–98 kHz co espond o a onal
emission whose cen al equency oscilla es wi h ime. This oscilla ion is
he main cause o he appea ance o na owband NIE in Fig. 3(a). These
emissions ha e been obse ed o he h ee cha ging cu en s con igu ed
du ing he cha ging p ocess o EV2.
The e o e, he esul s p esen ed in his sec ion lead o conclude ha
o some o he EVCPs unde analysis subs an ial ime a ia ions a e
egis e ed wi hin 10 minu es. These ime a ia ions may occu a
di e en ime ins an s wi hin he eco ding ime. This ype o ime
a iabili y was no obse ed in a p e ious s udy whe e EVCPs co e-
sponding o di e en EV and EVCS models we e e alua ed [16], in
which only sub-cycle a ia ions we e epo ed. This implies ha he
NIEs due o EVCPs a e e y dependen on he speci ic EV model.
3.2.2. Analysis o he ime a iabili y wi hin 50 s
The s udy p esen ed in his sec ion is ocused on he common
emissions p esen ed in Table 3 (emissions de ec ed bo h o isola ed and
on-line condi ions).
The analysis o he ime a iabili y wi hin 50 s is based on he FFT
analysis o he signal co esponding o each equency bin, which gi es
ise o a model composed o he componen s a ound 0 Hz, 50 Hz,
100 Hz, 150 Hz, and 200 Hz. This model accu a ely esembles he
o iginal signal and, hus, i is sui able o cha ac e izing he ime-
dependen beha io o he NIEs gene a ed by EVCPs whe e a pe iodic
pa e n wi h he undamen al pe iod o 50 Hz is obse ed [16].
As p esen ed in [16], in o de o be able o compa e he ampli ude o
he FFT componen s o di e en equency bins and EVCPs, he ampli-
udes o he FFT componen s o each equency bin a e no malized wi h
espec o he ampli ude o he componen a 0 Hz o his equency bin.
As an example, in Fig. 10, he modulus o he no malized FFT o he ime
samples co esponding o he emission gene a ed by EV5 wi h a
cha ging cu en o 12 A a 89 kHz unde isola ed and on-line condi ions
a e depic ed.
Fig. 10 shows ha he emissions gene a ed by his EVCP ollow he
model p esen ed in [16]. This FFT pa e n co esponds o a pe iodic
signal wi h a epe i ion a e o 50 Hz and, he e o e, as p esen ed in
[16], he emissions gene a ed by EVCPs a y wi hin he 20 ms o he
undamen al pe iod o he mains. This beha io has been obse ed o all
he common emissions analyzed, excep o hose gene a ed by EV2,
which a e cha ac e ized by onal emission whose cen al equency os-
cilla es wi h ime. Fo his eason, his EV model will be independen ly
analyzed a e wa ds.
In o de o gi e an o e iew o he o al a ia ion o e ime o each
emission o all he EVCPs (excep o EV2), he o al a iabili y is
calcula ed as he sum in linea scale o he ampli ude o he FFT com-
ponen s a 50 Hz, 100 Hz, 150 Hz, and 200 Hz, which is hen con e ed
o loga i hmic scale [16]. In Table 7, he o al a iabili y o he common
emissions de ec ed unde isola ed and on-line condi ions a e ga he ed
Fig. 9. Spec og am o he NIEs gene a ed by EV2 wi h a cha ging cu en o 16 A unde on-line condi ions conside ing he i s 50 seconds o eco ding.
Fig. 10. Modulus o he no malized FFT (dB) o he ime samples co e-
sponding o he emission gene a ed by EV5 wi h a cha ging cu en o 12 A a
89 kHz when measu ing unde isola ed ( ed) and on-line (black) condi ions.
J. Gonz´
alez-Ramos e al.
2
o he hi d LISN, whe e a PLC de ice is connec ed o ac as a ecei e ), conside ing an open-
ci cui con igu a ion, i.e., when no ex e nal load is connec ed o he il e .
Resul s and discussion
The in oduc ion o he EMC il e s in he se up shows a ia ions in he measu ed impedance
ha imply a ia ions in signal a enua ion. The ob ained esul s lead o he conclusion ha he
a enua ion in oduced by he il e s g ea ly depends on he poin o connec ion. A enua ion
measu ed when he il e is connec ed a poin s A o B is e y low, being p ac ically 0 dB when
connec ing hem in he EUT po o he second LISN. By con as , he a enua ion in oduced
by he il e s when connec ed a he same elec ical po as he ecei e is e y high on some
equency channels. The e o e, he impac is conside ably mo e no iceable when he il e is
connec ed nea he communica ions equipmen , especially i i wo ks as a ecei e . I should be
aken in o accoun ha he communica ion equipmen deployed in he ield pe o ms bo h
ansmi ing and ecei ing unc ions. The e o e, in p ac ice, an EMC il e nea a
communica ion equipmen migh imply an addi ional p oblem o NB-PLC, especially when
ecei ing ames om o he equipmen .
As he in oduc ion o he il e is only e y in luen ial when i is connec ed close o he ecei e ,
he FER-SNR cu es we e only ep esen ed when he il e was connec ed a poin C. The
h esholds ob ained using a DBPSK modula ion a e e y simila in he absence o a il e and
when each o he ou EMC il e s a e connec ed. This implies ha in a be e si ua ion, ha is,
when using mo e obus modula ions, he FER-SNR cu es will nei he be a ec ed.
As a consequence, i can be concluded ha he channel equency esponses due o he il e s
a e no su icien ly selec i e o deg ade NB-PLC in e ms o he SNR h esholds. This implies
ha he es ima ion and equaliza ion p ocesses de ined o PRIME 1.4 pe o m p ope ly unde
hese ci cums ances, no being a ec ed by he channel cha ac e is ics caused by he
in oduc ion o he EMC il e s.
In o de o analyze he conside able a enua ion in oduced by he EMC il e s, he schema ics
o he ou EMC il e s we e e alua ed. The ou o hem show capaci o s in pa allel o he L-
N po s (line side). Howe e , i he il e p esen ed a high-impedance induc i e in e ace, i
would decouple i s capaci i e load om he ne wo k, which could educe he ansmission
losses. Fo ha eason, we epea ed he es s connec ing wo o he ou il e s e e sed, so ha
he i s elemen s a e induc i e.
These ials lead o conclude ha he impedance measu ed o he equency band o in e es is
no in luenced when he il e is connec ed e e sed, ob aining equal esul s wi hou connec ing
he il e and when connec ing i h ough he load side o he EUT po o he LISNs.
Following he same p ocedu e explained be o e, he a enua ion o he il e s in his new
con igu a ion was calcula ed. Rega dless o he loca ion o he il e , he a enua ion in oduced
by hem when connec ed o he se up h ough he L’-N’ po s can be conside ed negligible.
Thus, i is demons a ed ha he a enua ion in oduced by he EMC il e s is mainly due o he
capaci i e bu den hey p esen in he ne wo k in e ace.
Conclusion
As a conclusion, i is shown ha he EMC il e s used in he s udy do no in oduce ab up
a ia ions in he channel equency esponse, so ha h esholds ha se he quali y o
communica ions a e no a ec ed. Howe e , hey do p esen conside able signal a enua ions
when hey a e connec ed nea he ecei e equipmen . These a enua ions could cause
communica ions o ail i he noise le el is high o i he ecei ed signal powe is close o he
sensi i i y limi o he ecei ing equipmen . This implies ha he EMC il e s used o powe
con e e s ha e an unin ended side e ec o a enua ing communica ion signals ha had no
been analyzed be o e. These high a enua ions a e due o he capaci i e in e ace o he il e s
wi h he g id.
255
B.1.5. In e na ional Con e ence Pape ICP5
This subsec ion p esen s he ollowing con e ence pape :
J. González-Ramos, B. G asel, I. Angulo, I. Fe nández, A. A inda, "E alua ion
o PRIME 1.4 in a Recons uc ed Low Vol age G id," 14 h Wo kshop o Powe line
Communica ions (WSPLC), Mannheim, Ge many, 2023.
Then, he mos ep esen a i e quali y indica o s conce ning his pape a e lis ed below:
Publishe : -
Con e ence: 14 h Wo kshop o Powe line Communica ions (WSPLC)
Yea o p esen a ion: 2023
Type o publica ion: Con e ence P oceedings
A ea: Elec ical & Elec onic Enginee ing
XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE
E alua ion o PRIME 1.4 in a Recons uc ed Low Vol age G id
Jon González-Ramos*, Be nha d G asel†, I zia Angulo*, Igo Fe nández*, Amaia A inda*
Uni e si y o he Basque Coun y (UPV/EHU), Bilbao, Spain*
Uni e si y o Applied Sciences Technikum Vienna, Vienna, Aus ia†
{jon.gonzalez , i zia .angulo, igo . e nandez, amaia.a inda }@ehu.eus*
[email p o ec ed]†
Keywo ds— Na owband Powe Line Communica ions, PRIME 1.4, Recons uc ed Low Vol age G id.
I. INTRODUCTION
Powe Line Communica ions (PLC) a e he mos ex ensi ely deployed echnologies by he Dis ibu ion Sys em Ope a o s
(DSOs) o ansmi da a o e he elec ical g id [1]. In he ecen de elopmen o Ad anced Me e ing In as uc u es (AMI),
Na owband PLC (NB-PLC) ha e been he p e e ed op ion o ansmi da a be ween Sma Me e s (SMs) and da a
concen a o s. Howe e , he e icien pe o mance o PLC is challenging, as he elec ical g id was no designed o da a
communica ion. Un il now, only equencies below 500 kHz ha e been conside ed, as hey a e expec ed o allow longe
ansmission dis ances. Ne e heless, he high le el o in e e ing noise and dis u bances ha a e p esen in his ange, oge he
wi h he ansmission losses due o he changing impedance o he medium, hampe he communica ions and p e en om
ob aining ull- ime a ailabili y o he ansmission de ices.
In his con ex , his pape aims a e alua ing he pe o mance o NB-PLC acco ding o PRIME 1.4 s anda d in a
econs uc ed Low Vol age (LV) g id conside ing he connec ion o mul iple elec onic de ices ha may ha e a nega i e impac
on communica ions. Mo eo e , a cha ac e iza ion o he emissions a he measu emen loca ions is also p esen ed, in o de o
ela e hei ampli ude and spec al shape o he po en ial deg ada ion o NB-PLC.
II. METHODOLOGY
The measu emen s we e pe o med in he econs uc ed LV g id a Uni e si y o Applied Sciences Technikum Vienna shown
in Fig. 1. This measu emen scena io is composed o a Seconda y Subs a ion (SS) and ou houses (H1, H2, H3, and H4), o
which di e en elec onic de ices can be connec ed [2]. The dis ance be ween he SS and each house is 17 m. All he
measu emen s we e ca ied ou a phase 2 (L2) conside ing a s a g id opology. The connec ed de ices a e as ollows: 2 SMs
(L123) a H1, 3 SMs (L123) and a Pho o ol aic (PV) sys em (L2) a H2, 2 SMs (L123), a PV sys em (L2) and a 1200 W load
(L123) a H3, and 3 SMs (L123), a PV sys em (L123), and a s o age sys em (L123) a H4.
Fig. 1. Recons uc ed LV g id a Uni e si y o Applied Sciences Technikum Vienna.
NB-PLC acco ding o PRIME we e e alua ed by means o wo PL360G55CF-EK boa ds, one ac ing as a ansmi e and
connec ed o he SS and he o he ope a ing as a ecei e and connec ed o he co esponding house. The s udy conside s he
equency channels 1 and 3-8 de ined in PRIME 1.4 s anda d [3], as well as DBPSK_C, DQPSK_C, R_DQPSK, and
R_DBPSK modula ions. In all he ials, 1000 ames wi h a 256-by e leng h we e ansmi ed.
The eco ding o he emissions was conduc ed using he measu emen sys em de eloped by he cu en au ho s p esen ed in
[4].
III. RESULTS AND CONCLUSIONS
In o de o e alua e he pe o mance o PRIME 1.4, o each con igu a ion, he mean alue o he Signal o Noise Ra io
(SNR) o all he ecei ed ames is calcula ed, in addi ion o he F ame E o Ra e (FER), which is de ined as he a io be ween
he e oneous ecei ed ames and he o al numbe o ansmi ed ames. In Table I and Table II, he mean SNR and FER
ob ained o each con igu a ion when he ecei e is loca ed a H1, H2, H3, and H a e ga he ed.
Table I. Mean SNR and FER ob ained o each equency channel and modula ion when he ecei e is loca ed a H1 (a) and H2 (b).
Modula ion
Channel
Mean
SNR
FER
DBPSK_C
CH1
5.8
1.000
CH3
20.3
0.000
CH4
13.0
0.000
CH5
15.0
0.002
CH6
20.5
0.000
CH7
20.8
0.000
CH8
19.8
0.000
DQPSK_C
CH1
7.5
1.000
CH3
18.3
0.000
CH4
11.3
0.006
CH5
14.8
0.000
CH6
21.8
0.000
CH7
22.0
0.000
CH8
20.5
0.000
R_DBPSK
CH1
5.8
0.000
CH3
17.5
0.000
CH4
9.8
0.000
CH5
14.8
0.000
CH6
21.0
0.000
CH7
21.0
0.000
CH8
19.0
0.000
R_DQPSK
CH1
7.3
0.213
CH3
18.8
0.000
CH4
11.5
0.000
CH5
16.0
0.000
CH6
21.5
0.000
CH7
21.3
0.000
CH8
19.3
0.000
Modula ion
Channel
Mean
SNR
FER
DBPSK_C
CH1
11.1
0.002
CH3
10.1
0.231
CH4
5.6
0.861
CH5
13.0
0.000
CH6
15.9
0.000
CH7
19.8
0.000
CH8
19.0
0.000
DQPSK_C
CH1
11.6
0.026
CH3
11.9
0.825
CH4
7.2
0.932
CH5
13.6
0.000
CH6
16.3
0.000
CH7
20.6
0.000
CH8
19.6
0.000
R_DBPSK
CH1
11.3
0.000
CH3
10.0
0.000
CH4
5.7
0.000
CH5
13.5
0.000
CH6
16.1
0.000
CH7
19.9
0.000
CH8
19.1
0.000
R_DQPSK
CH1
11.7
0.000
CH3
11.4
0.000
CH4
6.9
0.264
CH5
13.6
0.000
CH6
16.2
0.000
CH7
20.5
0.000
CH8
19.7
0.000
Table II. Mean SNR and FER ob ained o each equency channel and modula ion when he ecei e is loca ed a H3 (a) and H4 (b).
Modula ion
Channel
Mean
SNR
FER
DBPSK_C
CH1
8.4
0.029
CH3
12.5
0.010
CH4
10.9
0.000
CH5
18.5
0.000
CH6
21.0
0.000
CH7
21.3
0.000
CH8
21.6
0.000
DQPSK_C
CH1
9.1
0.936
CH3
14.4
0.085
CH4
11.1
0.000
CH5
19.1
0.000
CH6
22.5
0.000
CH7
22.6
0.000
CH8
23.1
0.000
R_DBPSK
CH1
6.6
0.000
CH3
9.4
0.000
CH4
7.6
0.000
CH5
10.7
0.000
CH6
18.1
0.000
CH7
19.8
0.000
CH8
19.5
0.000
R_DQPSK
CH1
7.6
0.000
CH3
10.3
0.000
CH4
8.3
0.000
CH5
10.4
0.000
CH6
18.4
0.000
CH7
19.8
0.000
CH8
19.2
0.000
Modula ion
Channel
Mean
SNR
FER
DBPSK_C
CH1
4.0
1.000
CH3
12.0
0.001
CH4
4.2
1.000
CH5
5.5
0.999
CH6
15.2
0.000
CH7
18.5
0.000
CH8
16.2
0.000
DQPSK_C
CH1
5.9
1.000
CH3
12.8
0.010
CH4
6.1
1.000
CH5
7.6
1.000
CH6
15.8
0.000
CH7
19.3
0.000
CH8
17.1
0.000
R_DBPSK
CH1
4.1
0.846
CH3
12.2
0.000
CH4
4.3
0.707
CH5
6.1
0.000
CH6
14.6
0.000
CH7
19.2
0.000
CH8
17.2
0.000
R_DQPSK
CH1
6.0
1.000
CH3
14.4
0.000
CH4
6.8
0.896
CH5
7.8
0.018
CH6
16.4
0.000
CH7
19.7
0.000
CH8
19.9
0.000
The esul s shown in Table I lead o conclude ha channel 1 (42-89 kHz) is signi ican ly mo e hos ile o communica ions.
This beha io can be clea ly obse ed, o example, when analyzing he esul s ob ained o he DQPSK_C modula ion. While
in channel 1 all ames a e ecei ed wi h e o s, in he uppe equency channels (3-8) only a FER highe han 0 is epo ed in
channel 5 (0.02). The deg ada ion o communica ions in channel 1 is mainly caused by he emissions o highe ampli ude
epo ed in his equency band (see Fig. 2). I should be no ed ha , in Eu ope, his is he equency ange assigned by
CENELEC o sma me e ing applica ions and, hus, an imp ope pe o mance o he sys em could be expec ed.
Howe e , in some cases, as shown in Table II a loca ion 4, a same FER (1.000) is ob ained in channel 1 and highe equency
channels ( o ins ance, channel 4 o DBPSK_C modula ion), al hough he ampli ude o he emissions is conside ably highe
in he 42-89 kHz equency band. Fo his eason, in hese ins ances, in o de o de e mine he cause o he deg ada ion o he
ope a ion o NB-PLC acco ding o PRIME 1.4, u he s udy is needed. This analysis should conside he in luence o he g id
impedance, bo h a he ansmi e and ecei e sides, and he channel equency esponse be ween he SS and he co esponding
loca ion.
Finally, i should be men ioned ha , as i could be expec ed, modula ions including epe i ion codes, i.e., R_DQPSK and
R_DBPSK, pe o m conside ably be e han hose using only Fo wa d E o Co ec ion (FEC) codes (DBPSK_C and
DQPSK_C). In he case o R_DBPSK, all he ames a e ecei ed co ec ly ega dless o he equency channel in which he
PLC signal is ansmi ed.
Fig. 2. QP alues o he ampli ude o he emissions measu ed a H1, H2, H3, and H4.
ACKNOWLEDGMENT
This wo k was inancially suppo ed in pa by he Basque Go e nmen unde he g an s IT436-22, PRE_2021_1_0006 and
PRE_2021_1_0051, and by he Spanish Go e nmen unde he g an s PID2021 124706OB-I00 (MCIU/AEI/FEDER, UE,
unded by MCIN/AEI/10.13039/5011000011033 and by “ERDF A way o making Eu ope”).
REFERENCES
[1] L. Lampe, A. Tonello, and T. Swa , Powe line communica ions: P inciples, s anda ds and applica ions om mul imedia o sma g id: Second
edi ion. 2016. doi: 10.1002/9781118676684.
[2] B. G asel, J. Bap is a, and M. T agne , “Sup aha monic and Ha monic Emissions o a Bi-Di ec ional V2G Elec ic Vehicle Cha ging S a ion and
Thei Impac o he G id Impedance,” Ene gies (Basel), ol. 15, no. 8, 2022, doi: 10.3390/en15082920.
[3] “PRIME Alliance, ‘PRIME 1.4 Whi e Pape ’, PRIME Alliance,” B ussels, 2014.
[4] I. Fe nández, D. de la Vega, A. A inda, I. Angulo, N. U ibe-Pé ez, and A. Llano, “Field T ials o he Cha ac e iza ion o Non-In en ional Emissions
a Low-Vol age G id in he F equency Range Assigned o NB-PLC Technologies,” Elec onics (Basel), ol. 8, no. 9, 2019, doi:
10.3390/elec onics8091044.
260
B.1.6. In e na ional Con e ence Pape ICP6
This subsec ion p esen s he ollowing con e ence pape :
J. González-Ramos, I. Angulo, A. A inda, I. Fe nández, D. De La Vega and A.
Galla e a, "BB-PLC o e LV ne wo ks – a s ep o wa d owa ds Sma G id
implemen a ion," ICSC-CITIES: V Ibe o-Ame ican Cong ess o Sma Ci ies, 2022.
Then, he mos ep esen a i e quali y indica o s conce ning his pape a e lis ed below:
Publishe : Commi ee o ICSC-CITIES
Con e ence: ICSC-CITIES: V Ibe o-Ame ican Cong ess o Sma Ci ies
Yea o p esen a ion: 2022
Type o publica ion: Con e ence P oceedings
A ea: Elec ical & Elec onic Enginee ing
BB-PLC o e LV ne wo ks – a s ep o wa d owa ds
Sma G id implemen a ion
Jon González-Ramos 1[0000-0001-9221-6957], I zia Angulo 1 [0000-0002-7855-3516], Amaia
A inda 1 [0000-0001-6058-1837], Igo Fe nández 1[0000-0002-9312-5886], Da id de la Vega 1 [0000-
0003-4811-4173] and Alexande Galla e a 1 [0000-0002-5336-8824]
1 Uni e si y o he Basque Coun y (UPV/EHU), Bilbao Bizkaia 48013, Spain
[email p o ec ed]
Abs ac . This a icle e iews he cu en s a e o he a o B oadband Powe
Line Communica ions (BB-PLC) o e he Low Vol age (LV) g id. Fi s , he e o-
lu ion o BB-PLC is summa ized, including he pas , p esen , and u u e o he
echnology. Second, he main oppo uni ies and applica ions a e ga he ed, includ-
ing sma me e ing, sma ci y se ices, con ol o dis ibu ed gene a ion, and
Elec ic Vehicles (EVs). Then, he challenges ha BB-PLC should ace a e p e-
sen ed, highligh ing he need o cha ac e izing he g id as a ansmission medium
in e ms o g id access impedance, a enua ion, and Non-In en ional Emissions
(NIEs), as well as he cybe secu i y issues ha should be aken in o conside a ion.
Finally, he cu en s a e o he al eady olled ou BB-PLC pilo p ojec s is de-
sc ibed.
Keywo ds: Low Vol age G id, B oadband Powe Line Communica ions, Sma
G ids.
1 In oduc ion
The Sma G id (SG) concep is b inging a g ea e olu ion, mainly in he dis ibu-
ion and cus ome domains, due o he ac ha i allows consume s o be mo e in ol ed
in he ope a ion o he powe sys em [1], [2]. The de elopmen o he SGs means he
end o he classic p oduc ion-dis ibu ion-gene a ion model o he elec ical g id, e ol -
ing in o a dis ibu ed, lexible and e icien model. In his way, i is pa ing he way o
he in eg a ion o Dis ibu ed Ene gy Resou ces (DERs), he managemen o Elec ic
Vehicle (EV) cha ging, and he inc easing ene gy demand esponse [3]-[6]. Mo eo e ,
i is also leading o a educ ion o he losses and illegal usage o he ansmission and
dis ibu ion lines, as well as an easy ope a ion o he g id and au oma ed moni o ing
[5], [7].
In his con ex , communica ions a e c i ical and essen ial o he p ope ope a ion
and pe o mance o he SGs [8]. As desc ibed in sec ion 5, some Dis ibu ed Sys em
Ope a o s (DSOs) ha e selec ed Powe Line Communica ions (PLC) as he al e na i e
o he de elopmen o new SG applica ions, because o he ad an ages i o e s com-
2
pa ed o o he communica ion echnologies. Fi s , as he elec ical g id is al eady de-
ployed, addi ional in as uc u e cos s a e a oided; and, second, co e age p oblems ha
may occu in wi eless communica ions a e p e en ed. Finally, hey also p o ide he
DSOs he oppo uni y o ha e con ol o hei own da a wi hou elying on hi d pa ies
[9].
Se e al DSOs ha e al eady deployed Na owband Powe Line Communica ions
(NB-PLC) echnologies wo ldwide o applica ions such as eleme y, moni o ing o
he de ices connec ed o he powe g id, and signal quali y analysis [10]-[13]. Ne e -
heless, u u e SG applica ions imply mo e demanding equi emen s (highe da a a e
and lowe la ency), which canno be accomplished wi h NB-PLC. Fo his eason, some
DSOs a e conside ing he deploymen o B oadband Powe Line Communica ions
(BB-PLC), also known as B oadband o e Powe Lines (BPL), in he Low Vol age
(LV) dis ibu ion g id.
This pape gi es an o e iew o he cu en s a e o he a o BB-PLC o e he LV
g id. In sec ion 2, he e olu ion o BB-PLC echnologies is p esen ed, whe eas sec ion 3
ga he s he main oppo uni ies and applica ions o BB-PLC o e he LV g id. Sec ion
4 summa izes he challenges ha should be aced by BB-PLC and, inally, in sec ion 5,
he cu en s a e o BB-PLC is included, highligh ing some o he pilo p ojec s ca ied
ou up o da e.
2 Pas , p esen , and u u e o BB-PLC
BB-PLC e e s o PLC sys ems ope a ing in he equency ange o abou 1 MHz o
30 MHz and beyond, wi h a signal bandwid h o ens o MHz and wi h da a a es ang-
ing om se e al Mbps o hund eds o Mbps.
The o igins o BB-PLC da e back o he la e 1990s. In a con ex o de egula ion o
he elecommunica ion and ene gy ma ke s in Eu ope, elec ic powe u ili ies consid-
e ed p o iding addi ional consume se ices h ough hei powe lines, including In e -
ne access by means o BB-PLC sys ems as an al e na i e o Digi al Subsc ibe Lines
(DSL) sys ems. Howe e , he high cos s and g owing Elec omagne ic Compa ibili y
(EMC) issues p e en ed he echnology om a mo e success ul deploymen . The e o e,
in ea ly 2000, he in e es s o he indus y ocused on he use o BB-PLC o in-home
communica ion and mul imedia applica ions [14], [15].
In pa allel, a he beginning o he 2010s, se e al indus y alliances and s anda ds
de eloping o ganiza ions s a ed de ining a new gene a ion o PLC sys ems ope a ing
a equency bands below 500 kHz. These NB-PLC sys ems a e cu en ly widely de-
ployed in he LV g id wo ldwide, especially o sma me e ing pu poses in Eu ope
[14]. Some examples o hese NB-PLC echnologies a e PoweRline In elligen Me e -
ing E olu ion (PRIME) [16], G3-PLC [17], Me e s and Mo e (M&M), CX1, and Open
Sma G id P o ocol (OSGP). In all cases, hese echnologies u n in o echnical s and-
a ds accep ed by s anda diza ion bodies such as he ITU o he IEEE.
Howe e , as mo e applica ions o he SG concep a e becoming a eali y (in eg a ion
o DERs, EVs, secu i y issues, e c.), a mo e eliable communica ion ne wo k wi h
highe da a a es and lowe la ency is needed. As a i s s ep, some DSOs a e al eady
3
using BB-PLC sys ems o e he Medium Vol age (MV) g id o backhauling a ic
wi hou he need o ibe op ics o wi eless communica ions [14], [18]. Bu mo e de-
manding equi emen s in e ms o secu i y, speed, la ency, and co e age also apply o
he communica ions o e he LV g id in o de o ob ain an enhanced con ol o his
dis ibu ion segmen . In his con ex , he concep o BB-PLC o e he LV ne wo k e-
su ges, bu no as a echnology o o e In e ne access o subsc ibe s, bu as a commu-
nica ion ne wo k o SG applica ions [18]. In Fig. 1, he e olu ion o BB-PLC echnol-
ogies is ga he ed.
Fig. 1. E olu ion o BB-PLC echnologies: pas , p esen and u u e.
Finally, in Table 1, a compa ison o he cha ac e is ics o NB-PLC and BB-PLC
echnologies is ga he ed, including he main applica ions and ad an ages and disad-
an ages.
Table 1. Compa ison o he cha ac e is ics o NB-PLC and BB-PLC echnologies.
Technology
NB-PLC
BB-PLC
F equency band
3-500 kHz
1.8-250 MHz
Modula ion
O hogonal F equency Di ision
Mul iplexing
O hogonal F equency Di ision
Mul iplexing
Bi a e
500 kbps
1.5 Gbps (indoo )
Applica ions
Sma me e ing
Elec ical signal Quali y
Remo e moni o ing
Sma me e ing
Sma ci y se ices
Dis ibu ed gene a ion
In eg a ion o EVs
Ene gy demand esponse
Ad an ages
Low powe consump ion
Long dis ances
High da a a e
Low la ency
Highe complexi y o he
ha dwa e
Disad an ages
Low da a a es
High la ency
High Non-In en ional
Emissions
No egula ion
High a enua ion
High Non-In en ional Emis-
sions
10
[38] Enabling deca boniza ion h ough sel - egula ing ene gy sys ems, Co po a e P esen a ion,
CORINEX. A ailable online: h ps://uploads-ssl.web-
low.com/62337650701b6 122d6b4d65/62753d15c0e ab5dd909a5ed_Co inex%20Co po-
a e%20Deck.pd (accessed on 04 Oc obe 2022).
[39] G. Abdulla, “The deploymen o ad anced me e ing in as uc u e”, 2015 Fi s Wo kshop
on Sma G id and Renewable Ene gy (SGRE), 2015, pp. 1-3, doi:
10.1109/SGRE.2015.7208738.
[40] Moma—Sma Ci y Mannheim Resea ch P ojec . A ailable online: h ps://www.ppc-
ag.de/wp-con en /uploads/media/PPC-Case-S udy-MOMA-16-2026-2E.pd (accessed on
04 Oc obe 2022).
[41] B enziko e , Alain & Mülle , Flo ian & Ke se zis, Alexand os & Kienzle, Flo ian & Man-
gani, Ma co & Eisen eich, Ma c & Fa ha , Yamshid & Bache , Raine . (2016). G idBox
pilo p ojec : A pla o m o moni o ing and ac i e con ol o dis ibu ion g ids. Compu e
Science - Resea ch and De elopmen . 10.1007/s00450-016-0308-5.
271
B.1.7. In e na ional Con e ence Pape ICP7
This subsec ion p esen s he ollowing con e ence pape :
J. González-Ramos, I. Angulo, A. A inda, I. Fe nández, A. Galla e a, D. de la
Vega, A. Sendin, I. Be ganza, R. Ayala and J. S. Gómez, "Cha ac e iza ion o he
LV dis ibu ion g id o he deploymen o a pilo BB-PLC ne wo k," 2023
IEEE In e na ional Symposium on Powe Line Communica ions and i s Applica ions
(ISPLC), Manches e , Uni ed Kingdom, 2023, pp. 19-24, doi:
10.1109/ISPLC57122.2023.10104185.
Then, he mos ep esen a i e quali y indica o s conce ning his pape a e lis ed below:
Publishe : IEEE
Con e ence: IEEE In e na ional Symposium on Powe Line Communica ions
and i s Applica ions (ISPLC)
Yea o p esen a ion: 2023
Type o publica ion: Indexed con e ence in IEEExplo e
A ea: Elec ical & Elec onic Enginee ing
XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE
Cha ac e iza ion o he LV dis ibu ion g id o he
deploymen o a pilo BB-PLC ne wo k
Jon González-Ramos*, I zia Angulo†, Amaia A inda*,
Igo Fe nández*, Alexande Galla e a*, Da id de la Vega*
*Dep . o Communica ions Enginee ing
†Dep . o Applied Ma hema ics
Uni e si y o he Basque Coun y (UPV/EHU)
Bilbao, Spain
{jon.gonzalez , i zia .angulo, amaia.a inda, igo . e nandez,
alexande .galla e a, da id.dela ega}@ehu.eus
Albe o Sendin, Iñigo Be ganza, Raquel Ayala,
Juan Sebas ián Gómez
Di ision o Con ol Sys ems and Telecommunica ions
Ibe d ola
Bilbao, Spain
{asendin, ibe ganza, ayala, jgua}@ibe d ola.es
Abs ac — As a p e ious s ep be o e he ollou o a pilo
B oadband Powe Line Communica ions (BB-PLC) ne wo k,
his pape desc ibes he measu emen campaign ca ied ou o
he cha ac e iza ion o he Low Vol age (LV) elec ical g id as a
ansmission medium in e ms o g id access impedance,
a enua ion, and Non-In en ional Emissions (NIEs). Fi s , he
main easons why BB-PLC is he echnology selec ed o he
de elopmen o new Sma G id (SG) applica ions o e he LV
dis ibu ion g id a e discussed. Second, he pilo ne wo k o be
olled ou is desc ibed. Then, he measu emen sys ems designed
o he assessmen o he g id impedance, a enua ion, and NIEs
a e p esen ed, as well as a b ie desc ip ion o he selec ed
loca ions ha co espond o nine scena ios o he LV
dis ibu ion g id in a own in he no h o Spain. These loca ions
ep esen se e al g id opologies and a eas o di e en
popula ion densi ies. Finally, some esul s ob ained in he ield
ials a e shown, which demons a e he impo ance o a p io
cha ac e iza ion o he LV g id as a ansmission medium o an
e icien planning o a BB-PLC deploymen . These esul s a e
only a ep esen a i e selec ion o he da a ob ained in he
measu emen campaign and, hus, u he analysis is equi ed
o d aw accu a e conclusions.
Keywo ds— B oadband Powe Line Communica ions,
channel cha ac e iza ion, Low Vol age g id, pilo ne wo k.
I. INTRODUCTION
Up o now, Na owband Powe Line Communica ions
(NB-PLC) ha e acili a ed he de elopmen o se e al Sma
G id (SG) applica ions, such as sma me e ing, analysis o
he elec ical signal quali y o g id au oma ion, among o he s
[1], [2]. Howe e , NB-PLC a e limi ed in e ms o da a a e
and la ency [3]. Fo his eason, he de elopmen o new
applica ions, such as he cha ging managemen o Elec ic
Vehicles (EVs) o he ene gy demand esponse, is o ally
condi ioned o new ansmission echnologies o e ing
high-demanding ea u es [1], [4]. Se e al Dis ibu ion
Sys em Ope a o s (DSOs) ha e op ed o B oadband Powe
Line Communica ions (BB-PLC), also known as B oadband
o e Powe Lines (BPL), o o e come his challenge, as hey
p o ide high da a a es and low la encies h ough he Low
Vol age (LV) g id. While he maximum achie able da a a e
by NB-PLC p o ocols ises up o 1 Mbps, G.hn, one o he
p e e ed op ions o he de elopmen o hese new SG
se ices, can p o ide da a a es o up o 1 Gbps [1].
The deploymen o NB-PLC echnologies all a ound he
wo ld has clea ly demons a ed ha he elec ical g id is a
ha sh p opaga ion medium o da a ansmission since he
cable cha ac e is ics depend on ime and equency [5]-[7].
Al hough he e is a lack o knowledge o he beha io o he
LV g id in he equency ange assigned o BB-PLC,
communica ions a e also expec ed o be a ec ed a hese
highe equencies by g id impedance a ia ions, a enua ion,
and noise, also e e ed o as Non-In en ional Emissions
(NIEs) in he li e a u e [5], [8]-[11]. This pape desc ibes he
measu emen campaign ca ied ou in a own in he no h o
Spain o he cha ac e iza ion o he LV dis ibu ion g id as a
ansmission medium in e ms o g id access impedance,
a enua ion, and NIEs in he BB-PLC equency ange
(1.7-15 MHz ), as he i s s ep o he ollou o a pilo
BB-PLC ne wo k. In sec ion II, he main easons why
BB-PLC has been selec ed as an op ion o he deploymen o
new SG applica ions o e he LV dis ibu ion g id a e
discussed. Sec ion III desc ibes he pilo ne wo k o be
deployed. In sec ion IV, he main cha ac e is ics o he
selec ed measu emen loca ions a e desc ibed, as well as he
measu emen sys ems used o he assessmen o he g id
access impedance, a enua ion, and NIEs. Finally, in
sec ion V, some example esul s a e p esen ed.
II. BROADBAND POWER LINE COMMUNICATIONS IN THE
LV DISTRIBUTION GRID
BB-PLC echnology has al eady been success ully used in
he Medium Vol age (MV) dis ibu ion g id o p o ide
b oadband backbone connec i i y among Seconda y
Subs a ions (SSs) o ca y au oma ion, emo e con ol and
moni o ing, and emo e me e managemen - ype SG se ices.
BB-PLC echnologies ha e also been a success o in-home
scena ios, c ea ing local home a ea ne wo ks o e he LV
domes ic wi ing.
BB-PLC echnologies we e also deployed o e he LV
dis ibu ion g id in he con ex o he ea ly days o he
elecommunica ions ma ke de egula ion in Eu ope o
p o ide esiden ial In e ne access. This esul ed in limi ed
comme cial success bu c ea ed a legacy o echnical
solu ions ha enabled i s adop ion o SG and in-home
applica ions [12], [13].
Rega ding Ad anced Me e ing In as uc u e (AMI)/sma
me e ing applica ions, NB-PLC echnologies ha e achie ed
ull co e age o each SS in equencies below 500 kHz in
2023 IEEE In e na ional Symposium on Powe Line Communica ions and i s Applica ions (ISPLC)
978-1-6654-9135-8/23/$31.00 ©2023 IEEE 19
2023 IEEE In e na ional Symposium on Powe Line Communica ions and i s Applica ions (ISPLC) | 978-1-6654-9135-8/23/$31.00 ©2023 IEEE | DOI: 10.1109/ISPLC57122.2023.10104185
Au ho ized licensed use limi ed o: Uni e sidad Pais Vasco. Downloaded on No embe 12,2024 a 14:42:27 UTC om IEEE Xplo e. Res ic ions apply.
Eu ope, suppo ed on sys ems ha ha e epe i ion
capabili ies so ha sma me e s (SMs) help o he SMs and
da a concen a o s in SSs o each e e y co ne o he LV
dis ibu ion g id. AMI/sma me e ing ollou s ha e
ans o med u ili y ope a ions. Fo me ly labo -in ensi e
p ocesses ha e been subs i u ed by au oma ed and emo ely
con olled in e en ions. T adi ional elec ic se ice
‘subsc ibe s’ a e now educa ed consume s who no only wan
imely in o ma ion on hei consump ion bu also he abili y
o in e ac wi h he elec ic se ice based on comme cial and
in as uc u e solu ions.
Howe e , bo h he equency band cons ain s and he
implemen a ions ha need o cope wi h he na u al hos ili y
o he LV g id channel cha ac e is ics, ha e p oduced limi ed
h oughpu sys ems ha sa is y he equi emen s o manda ed
AMI/sma me e ing ollou s, bu ha a e no eady o suppo
new ea u es o se ices ela ed o sma me e ing (e.g. highe
equency o eadings, eal- ime SM managemen by
cus ome s, e c.) o ela ed o he expansion o SG applica ions
in o he LV dis ibu ion g id (e.g. LV au oma ion, eal- ime
ol age and cu en moni o ing, e c.).
Ne e heless, BB-PLC o e LV access applica ions we e
no he ocus o he AMI/sma me e ing e o s in p e ious
yea s o a a ie y o easons. On he one side, he sma
me e ing ma ke had no ye unde gone he needed
ans o ma ion o accommoda e he new age o digi al
solu ions; on he o he , BB-PLC echnologies had p o en no
o be eady o he ype o una ended low-cos ollou s ha
AMI p og ams demanded. Thus, while he basic BB-PLC
echnology migh ha e been almos eady, he de ail aspec s
ha could make i success ul o AMI pu poses we e no
de eloped. Nei he he s anda ds no he companies and g id
knowledge could cope wi h he needs.
BB-PLC sys ems can wo k in pa allel wi h NB-PLC
sys ems, as al eady demons a ed [14]. Mo eo e , [15] shows
a comple e se o scena ios o BB-PLC and NB-PLC
e olu ion o gua an ee a oadmap o co e he di e en needs
and capabili ies. These scena ios allow om jus highe
bandwid h sma me e ing o a ull BB-PLC ne wo k on op
o which sma me e ing plus LV moni o ing and con ol can
be implemen ed.
III. D
ESCRIPTION OF THE PILOT
BB-PLC
NETWORK
The speci ic use case ha he pilo BB-PLC ne wo k
in ends o es is such ha a BB-PLC ne wo k will be
deployed om he SS (whe e a LV BB-PLC Head End will
be ins alled) up o a numbe o s ee cabine s and me e
ooms (i.e., in he icini y o he use box whe e he u ili y
LV g id ends) in which LV BB-PLC endpoin s will be
loca ed.
Since he SMs hemsel es a e s ill expec ed o be cos -
e ec i e NB-PLC de ices [16], speci ically using PRIME
echnology below 500 kHz [17] a PRIME Base Node will
also be loca ed along wi h each LV BB-PLC endpoin , such
ha a second, mo e “local”, NB-PLC subne wo k is o med
be ween he PRIME Base Node and he ele an SMs o ha
s ee cabine / me e oom. The PRIME Base Node ac s hen
as a Ga eway be ween he BB-PLC and he NB-PLC
ne wo ks [18].
In e ec , wha we achie e is he c ea ion o many
independen NB-PLC subne wo ks wi h a educed numbe o
elemen s (hence inc eased pe o mance), which a e hen all
backhauled h ough he single LV BB-PLC ne wo k o ha
SS. This is he way ha SSs wi h up o one housand SMs a e
expec ed o be ully ead e e y 15 minu es, along wi h
addi ional capabili ies o LV con ol and moni o ing which
a e being de eloped. In Fig. 1, he schema ic o he basic
ne wo k a chi ec u e o he pilo BB-PLC ne wo k is
p esen ed.
Fig. 1. Schema ic o he basic ne wo k a chi ec u e o he pilo BB-PLC
ne wo k.
The loca ion has been chosen o be ep esen a i e o
ypical deploymen scena ios, including SSs wi h all anges
o numbe o me e s and elec ical opologies, wi hin he
limi ed geog aphical ex ension ha make bo h measu emen s
and deploymen s ep esen a i e and a o dable. This a ea
also o e s a combina ion o easily accessible SSs and use
box loca ions, wi h o he s p esen ing challenges o BB-PLC
de ice ins alla ion; i also includes zones whe e me e s can
be easily concen a ed in buildings, while a he same ime,
he LV eede s connec dispe se me e s. These aspec s a e
impo an , since BB-PLC echnologies mus unc ion
p ope ly e en in scena ios whe e he e a e impo an
limi a ions o he loca ions in which de ices can be ins alled.
U ili ies canno adap hei in as uc u e, and any PLC
echnology mus be enhanced o cope wi h g id cons ain s,
whe he hey allow o signal epe i ion o no .
IV. M
ETHODOLOGY FOR THE CHARACTERIZATION OF THE
LV
GRID
A. Planning o ield ials
Since he beha io o he elec ical g id depends on se e al
ac o s: g id opology, ype o cable and dis ance be ween he
ansmi e and ecei e , among o he s [1], [19], ield ials
should cha ac e ize ep esen a i e en i onmen s conside ing
a wide ange o hese g id con igu a ions.
In o de o p o ide a comp ehensi e scena io o he
beha io o he elec ical g id, di e en dis ances and
opologies be ween he SS and he access poin s ha e been
conside ed, as well as a eas wi h di e en popula ion
densi ies (u ban/subu ban and u al scena ios). The numbe
o cus ome s a each SS may ange om a ew ones in u al
scena ios, o many hund eds in dense u ban zones. The
dis ances om he SS (i.e., he s a o he LV dis ibu ion
g id) o he SMs ange om ens o me e s (dense u ban
a eas) o hund eds (u ban and subu ban a eas) and up o
kilome e s in u al a eas. Thus, di e en g oups o simila
ypes o SSs may be de ined [20].
As a esul , he measu emen campaign p esen ed in his
a icle was ca ied ou in nine scena ios o he LV dis ibu ion
g id in he selec ed own. Each scena io consis s o a SS and
he co esponding downs eam las -mile dis ibu ion g id o
he cus ome s. The measu emen s we e pe o med bo h a he
Low-Vol age Dis ibu ion Swi chboa ds (LVDSs) inside he
SSs and he access poin s wi hin he dis ibu ion g id (labeled
2023 IEEE In e na ional Symposium on Powe Line Communica ions and i s Applica ions (ISPLC)
20
Au ho ized licensed use limi ed o: Uni e sidad Pais Vasco. Downloaded on No embe 12,2024 a 14:42:27 UTC om IEEE Xplo e. Res ic ions apply.
as Rx). I is impo an o men ion ha some imes hose access
poin s a e use boxes (s ee cabine s) while o he s hey a e
SM panels a he en ance o basemen o buildings.
In all cases, he h ee elec ical phases ha e been
conside ed. Fo his eason, a each elec ical poin , h ee
impedance and NIE measu emen s, one o each phase, and
nine a enua ion measu emen s (phase1-phase1,
phase1-phase2, phase1-phase3, phase2-phase2,
phase2-phase3, e c.) a e pe o med. Rega ding he
a enua ion measu emen s, he ansmi e is connec ed a he
SS, while he ecei e is connec ed a he access poin .
Fu he mo e, in i e cases, a enua ions a e also measu ed by
injec ing a he access poin and ecei ing a he SS. In his
way, he symme y o he elec ical g id can be e alua ed.
In o al, he measu emen campaign consis ed o 207
measu emen s o g id access impedance (36 a he SSs and
171 a he access poin s), 549 measu emen s o signal
a enua ion (504 om he SS o he Rx poin s and 45 om he
Rx poin s o he SS), and 207 measu emen s o NIE (36 a he
SSs and 171 a he access poin s).
As an example o he selec ion o measu emen loca ions
in o de o s udy he in luence o dis ance in he esul s, he
in o ma ion abou he access poin s co esponding o SS 8 is
e lec ed in Table I. I can be seen ha he e is la ge a ia ion
in dis ance, om 8 m o 136 m. I all he measu emen
loca ions a e conside ed, he dis ance anges om 8 m o
472 m. In his way, all he ypical dis ances in he LV g id in
he no h o Spain can be analyzed.
Table I. In o ma ion abou he access poin s co esponding o SS 8.
ID LV
Boa d
LV
Feede
Dis ance o
SS 8 (m)
Rx 109 11 1 8
Rx 110 11 1 136
Rx 111 11 6 18
Rx 112 11 8 81
Rx 113 11 8 64
As an example o di e en g id opologies, Fig. 2 shows
wo elec ical LV eede s co esponding o SS 1. The i s
eede p esen s long dis ances, up o a ound 300 m, and
se e al b anches. The second, in u n, is a signi ican ly sho
single b anch. This shows ha he e is big di e ences among
opologies p esen in he LV g id, which can ha e an impac
on g id access impedance and he a enua ion o he
ansmi ed signal.
Fig. 2. Two elec ical eede s co esponding o SS 1.
B. Desc ip ion o he measu emen sys em
1) Measu emen sys em o he assessmen o he
Non-In en ional Emissions
The measu emen sys em o he cha ac e iza ion o he
NIEs p esen in he LV g id, shown in Fig. 3, is based on a
cu en p obe esponsible o measu ing he cu en (I1)
lowing h ough a capaci o (C1).
Then, a digi ally implemen ed quasi-peak (QP) de ec o is
applied o he eco ded cu en emissions acco ding o
CISPR16-1-1 [21], using a Lanczos window o 20 ms wi h an
o e lap o 90 %. In o de o gi e he same g anula i y in he
whole equency band, he ime-windowing ul ills a 6 dB
bandwid h o 200 Hz om 9 kHz o 15 MHz.
Fig. 3. Measu emen sys em o he assessmen o he NIEs.
2) Measu emen sys em o he assessmen o he g id
access impedance
The measu emen sys em o he assessmen o he g id
access impedance, shown in Fig. 4, is desc ibed in de ail in
[22]. I calcula es he impedance as he a io be ween a
ol age and a cu en o e a well-cha ac e ized e e ence
impedance. To do so, a signal gene a o is esponsible o
injec ing a sweep co e ing he whole equency band
be ween 20 kHz and 15 MHz. The signal gene a o is
connec ed o a cu en couple , which injec s he ansmi ed
signal in o he ci cui by means o he magne ic ield induced
in he wi e.
The impedance measu emen sys em de e mines he
impedance as he a io be ween he ol age V and he cu en
I3 shown in Fig. 4. I3 is ob ained by applying Ki chho ’s i s
law, I3 = I1 – I2, whe e he cu en s I1 and I2 a e measu ed by
means o wo cu en p obes. Since he ne wo k impedance
o be measu ed is in pa allel wi h he well-cha ac e ized
e e ence impedance (ZRC), composed o a esis o (R) and a
capaci o (C2) in se ies, he ol age ac oss ZRC and ZGRID is
he same. The e o e, he ol age V can be calcula ed as he
p oduc o ZRC, p e iously measu ed wi h a p ecision LCR
me e in he whole analyzed equency band, and he cu en
I2 (V = ZRC · I2).
I mus be men ioned ha , in gene al, ne wo k ol age
VGRID is negligible wi h espec o he injec ed signal a he
equency band o in e es . I i is no he case o a speci ic
equency bin, his equency bin is dis ega ded and ob ained
by in e pola ion.
Fo he signal p ocessing, a sliding Gaussian windowing
wi h 100 µs du a ion and an o e lapping o 90 % be ween
consecu i e windows is applied. Fi s , he FFT o each ou pu
window o he measu ed cu en s (I1 and I2) is pe o med, so
ha wo ma ixes ep esen ing he spec og ams o hose
eco ded signals in he equency and ime domains a e
2023 IEEE In e na ional Symposium on Powe Line Communica ions and i s Applica ions (ISPLC)
21
Au ho ized licensed use limi ed o: Uni e sidad Pais Vasco. Downloaded on No embe 12,2024 a 14:42:27 UTC om IEEE Xplo e. Res ic ions apply.
ob ained. Then, he equency- empo al bins in which he
injec ed sweep is p esen a e de ec ed, in o de o ge he
measu ed cu en s only as a unc ion o equency (I1 ( ) and
I2 ( )). Finally, he impedance wi h espec o equency is
calcula ed by applying he ollowing exp ession:
Z
∙
.
Fig. 4. Measu emen sys em o he assessmen o he g id access impedance.
3) Measu emen sys em o he assessmen o he
a enua ion
The measu emen sys em o he cha ac e iza ion o he
a enua ion, shown in Fig. 5, is based on he injec ion o
mul iple sho -du a ion sweeps (1 ms) om 20 kHz o
15 MHz on he ansmission side. The ol age le el a he
injec ion poin is calcula ed ollowing a simila p ocedu e o
he one desc ibed in sec ion IV.B.2), ha is, measu ing he
ol age o e a well-cha ac e ized impedance (ZRC) by means
o he cu en measu ed by he cu en p obe (I2). To ob ain
he ansmission losses, he ol age le el a he ecep ion
poin is compa ed wi h he ol age le el a he injec ion poin .
As he desi ed ecei ed signal is e y low in compa ison wi h
he NIEs, con inuous a e ages o he eco ded sweeps, once
hey a e aligned, a e needed, so ha he noise le el is educed.
This a e age p ocess is i al o he p ope assessmen o he
ansmission losses be ween wo elec ical poin s.
I is wo h men ioning ha synch oniza ion be ween he
injec ion and eco ding de ices in e ms o sampling
equency is needed. Fo his eason, an ex e nal e e ence o
GPS synch oniza ion is necessa y o he signal gene a o and
he oscilloscopes on bo h sides.
Fig. 5. Measu emen sys em o he assessmen o he a enua ion.
V. E
XAMPLE RESULTS
In his sec ion, example esul s o measu ed g id access
impedance, a enua ion, and NIEs a e p esen ed.
A. Example o measu ed g id access impedance
Fig. 6 and Fig. 7 show he modulus and phase o he g id
access impedance measu ed a one o he loca ions
conside ing he h ee elec ical phases.
Fig. 6. Modulus o he g id access impedance measu ed a a ce ain Rx.
Fig. 7. Phase o he g id access impedance measu ed a a ce ain Rx.
In Fig. 6, high-ampli ude esonances can be obse ed a
speci ic equencies. A hose equencies, ab up changes in
phase a e shown in Fig. 7. These esonances could imply
conside able impedance misma ches, a ec ing he
p opaga ion o BB-PLC signals and inc easing he emissions
in oduced by ce ain elec onic equipmen a he swi ching
equency and i s ha monics [23], [24]. These esonances
ha e also been epo ed in [25] in he equency ange
0-500 MHz, whe e mean impedance alues up o 400 Ω a e
ound.
Al hough simila spec al pa e ns a e shown in he
impedance o he h ee phases, signi ican di e ences a e
egis e ed in ampli ude. Fo ins ance, a ound 10 MHz, he
impedance ises o 200 Ω a phases 1 and 3, whe eas a
conside ably highe impedance modulus, up o 400 Ω, is
measu ed a phase 2.
Impedance magni ude ( )
2023 IEEE In e na ional Symposium on Powe Line Communica ions and i s Applica ions (ISPLC)
22
Au ho ized licensed use limi ed o: Uni e sidad Pais Vasco. Downloaded on No embe 12,2024 a 14:42:27 UTC om IEEE Xplo e. Res ic ions apply.
[Document text truncated for crawler view.]
