IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 61, NO. 2, MARCH/APRIL 2025 1839
Expe imen al E alua ion and Coo dina ed
Deploymen o Ramp-Ra e Limi a ion Agains Rapid
Vol age Changes in Dis ibu ion Sys ems
S elios C. Dimoulias , S uden Membe , IEEE, Ky iaki-Ne eli D. Malamaki , Membe , IEEE,
And ei Mihai G oss , F ancisco de Paula Ga cía-López , Geo gios C. K yonidis , Senio Membe , IEEE,
and Manuel Ba agán-Villa ejo
Abs ac —Clima e change has expedi ed he in eg a ion o
con e e -in e aced enewable ene gy sou ces (CIRESs) in dis i-
bu ion sys ems. Ne e heless, he ac i e powe ola ili y o CIRESs
and he esul ing high amp a es a e known o downg ade he ol -
age quali y o dis ibu ion sys ems, causing apid ol age changes
(RVCs). In his con ex , he ope a ion o CIRESs unde amp- a e
limi a ion (RRL) schemes, employing an ene gy s o age sys em,
could se e as a p e en i e ac ion agains RVCs, he eby es o ing
ol age quali y. To his end, his pape p o ides insigh s in o he
deploymen o RRL con ol agains RVCs in wo s ages. Fi s ly,
he ap ness o RRL as a p e en i e ac ion is expe imen ally e alu-
a ed in a scaled-down labo a o y es bed o he CIGRE Eu opean
benchma k medium- ol age eede , hos ing CIRES p o o ypes.
The RVCs a e de ined as pe he IEEE 1547:2018 and IEC 61000-
4-30:2015 S anda ds. The expe imen al esul s demons a e ha
he ope a ion o CIRESs unde RRL con ol can enhance ol age
quali y by supp essing RVCs. Subsequen ly, a no el, sys em-le el
s a egy o he coo dina ed mi iga ion o RVCs is de eloped. Fo
a gi en dis ibu ion sys em, he p oposed s a egy alloca es he
RRL unc ionali ies among CIRES uni s, o achie e he elimina ion
o RVCs unde minimal CIRES engagemen and ene gy s o age
equi emen s. The e ec i eness o he s a egy is es ed ia dynamic
ms simula ions on he IEEE Eu opean Low Vol age es eede .
Index Te ms—Ac i e dis ibu ion ne wo ks, ene gy s o age
sys ems, amp- a e limi a ion, apid ol age changes, sys em
planning, ol age quali y.
Recei ed 2 July 2024; e ised 17 Sep embe 2024; accep ed 24 Oc obe 2024.
Da e o publica ion 27 Decembe 2024; da e o cu en e sion 4 Ap il 2025.
Pape 2024-ESC-1007.R1, p esen ed a he 2023 In e na ional Con e ence on
Sma Ene gy Sys ems and Technologies (SEST), Mugla, Tü kiye, Sep. 04–06,
and app o ed o publica ion in he IEEE TRANSACTIONS ON INDUSTRY APPLI-
CATIONS by he Ene gy Sys ems Commi ee o he IEEE Indus y Applica ions
Socie y [DOI: 10.1109/SEST57387.2023.10257447]. This wo k was suppo ed
by he Eu opean Union unde he Ho izon Eu opean p ojec COCOON unde
G an 101120221. (Co esponding au ho : Ky iaki-Ne eli D. Malamaki.)
S eliosC.Dimoulias,Ky iaki-Ne eliD.Malamaki,andGeo giosC.K yonidis
a e wi h he Depa men o Elec ical and Compu e Enginee ing, A is o le Uni-
e si y o Thessaloniki, 54124 Thessaloniki, G eece (e-mail: [email p o ec ed];
[email p o ec ed]; [email p o ec ed]).
And eiMihai G oss,F ancisco dePaula Ga cía-López,and ManuelBa agán-
Villa ejo a e wi h he Depa men o Elec ical Enginee ing, Uni e sidad de
Se illa, 41004 Se ille, Spain (e-mail: [email p o ec ed]; [email p o ec ed]; manuel-
[email p o ec ed]).
Colo e sions o one o mo e igu es in his a icle a e a ailable a
h ps://doi.o g/10.1109/TIA.2024.3523456.
Digi al Objec Iden i ie 10.1109/TIA.2024.3523456
I. INTRODUCTION
AMID he ongoing ene gy ansi ion, new challenges a e
in oduced o elec ici y g ids, pa icula ly due o he
p oli e a ion o con e e -in e aced enewable ene gy sou ces
(CIRESs). Rega ding dis ibu ion sys ems, he ola ile gene a-
iono CIRESscauses echnicalp oblems, suchas e e sepowe
lows and o e ol ages [1]. In his con ex , a phenomenon ha
has been highligh ed by echnical s anda ds [2],[3],[4] and
epo s om egula o y au ho i ies [5], bu emains ela i ely
o e looked by he echnical li e a u e, is ha o apid ol age
changes (RVCs). De ined in he IEEE 1547:2018 [3] and he
IEC 61000-4-30:2015 [4] S anda ds, RVCs can eme ge due o
high amp a es o CIRES ac i e powe and ha e been iden i ied
o deg ade he pe o mance o elec ical equipmen and p oduce
licke , hus comp omising ol age quali y [6],[7]. As a p oac-
i e esponse,CIRESscanope a e,inconjunc ionwi hanene gy
s o age sys em (ESS), unde powe smoo hing schemes, so as o
p e en high amp a es,andhence,RVCs[8],[9].Amongpowe
smoo hing app oaches, amp a e limi a ion (RRL) echniques
a e p e e ed due o hei supe io pe o mance in e ms o he
achie ed smoo hing e ec and he mo e e icien ESS usage, as
a es ed by ecen e iew [10],[11] and compa a i e [12],[13],
[14],[15],[16] s udies. Mo e speci ically, bo h he simula ion-
based [10],[11],[12],[15] and he expe imen al [13],[14],[16]
s udies ha e demons a ed ha RRL echniques ou pe o m al-
e na i e app oaches, such as mo ing-a e age o low-pass il e
me hods, in e ms o he equi ed ESS sizing and ope a ional
cycles, while also achie ing ewe amp a e limi iola ions.
Despi e he ex ensi e li e a u e on RRL con ol, only a ew
wo ks examine he po en ial mi iga ing impac o RRL schemes
on ol age a ia ions, and e en ewe ocus on RVCs. Mo e
speci ically, in [17], RRL con ol is applied o alle ia e ol -
age luc ua ions. Howe e , he s udy neglec s he RVC me ic,
conside ing he 1-minu e licke e ec as a pe o mance index,
and i p ima ily ocuses on echno-economic aspec s o he
employed ESS. The RVC me ic is also neglec ed in [18], whe e
a ol age-smoo hing algo i hm, combining RRL wi h eac i e
powe con ol, is p oposed. Mo eo e , he s udy ocuses on
he e ec o eac i e powe con ol, while he he ein analysis
does no quan i y he achie ed ol age-smoo hing e ec . In [8],
he au ho s in es iga e he e ec o RRL agains apid ol age
© 2024 The Au ho s. This wo k is licensed unde a C ea i e Commons A ibu ion-NonComme cial-NoDe i a i es 4.0 License. Fo mo e in o ma ion, see
h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/
1840 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 61, NO. 2, MARCH/APRIL 2025
a ia ions caused by cloud passing in pho o ol aic (PV) plan s,
and hei s udy is ex ended o wind u bines in [9]. None he-
less, hese wo ks do no consul he de ini ions o he ele an
S anda ds [3],[4].
Addi ionally, and mos impo an ly, he abo e s udies a e
limi ed o demons a ing and quan i ying he e ec o RRL
agains ol age luc ua ions, assuming uni o m applica ion o
heRRL con olbyall CIRESuni s o adis ibu ionsys em. This
app oach, al hough e ec i e, is cha ac e ized by excessi e capi-
alexpendi u e,sincei implies ha e e yCIRESuni is equi ed
o be equipped wi h an ESS and be upg aded wi h RRL con ol
unc ionali ies. The e o e, in he con ex o RVC mi iga ion,
he echnical li e a u e lacks a s a egy o he coo dina ed and
e icien alloca ion o he RRL unc ionali ies among he CIRES
uni s o a dis ibu ion sys em, conside ing ac o s such as he
dis ibu ion sys em opology and echno-economic indices, e.g.,
ESS size and cos . F om he poin o iew o he indus y, his
impedes dis ibu ion sys em ope a o s (DSOs) om add essing
he issue o RVCs in p ac ice, e en hough i s impo ance is
highligh ed in he echnical s anda ds and li e a u e, and some
Eu opean coun ies, e.g., No way, ha e o icially imposed limi s
on he allowable numbe o RVCs pe day [5].
Rega ding he expe imen al implemen a ion o RRL, he ele-
an wo ks ha e been limi ed o he expe imen al alida ion and
compa a i e e alua ion be ween no el and con en ional RRL
echniques a CIRES le el. In his con ex , he cu en li e a u e
has ocused on he powe smoo hing e ec i eness o he es ed
echniques and he echnical equi emen s o he employed
ESSs. Mo e speci ically, in [19], a no el RRL scheme based on
he p edic o -co ec o logic is expe imen ally alida ed, while
in [20], an al e na i e me hod, combining supe capaci o s and
uel cells, is es ed. Simila ly, he au ho s in [21] p opose and
expe imen ally e alua e he pe o mance o a new, il e -based
echnique ha employs supe capaci o s. Mo eo e , in [14],se -
e al RRL con ol schemes a e expe imen ally compa ed in a
mic og id en i onmen . Fu he mo e, al hough he co ela ion
be ween CIRES ac i e powe ola ili y and ol age luc ua ions
is expe imen ally e i ied in [9], he au ho s do no ex end hei
expe imen al analysis o he posi i e impac o RRL on he
mi iga ion o hose luc ua ions. Consequen ly, a comp ehensi e
expe imen al e alua ion o he po en ial mi iga ing e ec o
RRL agains RVCs in dis ibu ion sys ems is missing om he
echnical li e a u e.
Ha ing iden i ied hese exis ing esea ch gaps, he scope o
his pape is wo- old: (a) o expe imen ally demons a e he
ap ness o RRL schemes o he mi iga ion o RVCs in ac i e
dis ibu ion sys ems; (b) o p opose a me hod o he e icien
alloca ion o he RRL con ol among he dis ibu ed CIRES,
aiming o he elimina ion o RVCs unde minimal equi emen s
in CIRES engagemen and ene gy s o age. An ini ial expe imen-
al s age o he wo k was o iginally p esen ed in [1]. Building
upon [1], his wo k en iches he expe imen al e alua ion wi h
addi ional esul s and in oduces he dis ibu ion sys em-le el
coo dina ion s a egy o RVC mi iga ion.
Conce ning he expe imen al s age, he RRL scheme o [16]
is inco po a ed in h ee CIRES labo a o y p o o ypes wi hin
a scaled-down e sion o he CIGRE Eu opean benchma k
medium- ol age (MV) eede , p esen ed in [22]. Each CIRES
is equipped wi h a supe capaci o , which se es as he ESS o
pe o m he RRL con ol. To achie e mo e comp ehensi e e-
sul s, wo dis inc powe p o iles a e used o emula e he CIRES
ola ili y. The expe imen al esul s a e assessed acco ding o he
guidelines o [3] and [4].
The second s age o he pape ega ds he de elopmen o
a no el, uni e sal s a egy o he coo dina ed deploymen o
RRL con ol agains RVCs, designed o dis ibu ion sys em
ope a o s. Speci ically, by ma hema izing he p oblem o RVC
p e en ion, he p oposed me hod de e mines which CIRES uni s
need o pe o m RRL in o de o supp ess RVCs wi h he
minimum numbe o RRL-pe o ming CIRESs and wi h he
minimum o al ins alledESS capaci y. Addi ionally, he equi ed
deg ee o RRL pe CIRES uni is speci ied. The pe o mance
o he me hod is assessed ia ms simula ions on he IEEE
Eu opean Low Vol age (LV) es eede .
In summa y, he pape in oduces he ollowing con ibu ions
wi h espec o he cu en s a e-o - he-a :
Expe imen ally e alua es he impac o RRL agains RVCs
in ac i e dis ibu ion sys ems.
In oduces a new, sys em-planning me hod o e icien ly
p e en RVCsinac i edis ibu ionsys ems,unde minimal
CIRES engagemen and ene gy s o age equi emen s.
The es o he manusc ip is o ganized as ollows: Sec ion II
p o ides an o e iew o he IEC and IEEE de ini ions o RVCs.
Sec ion III b ie ly p esen s he conside ed RRL echnique. In
Sec ion IV, he expe imen al e alua ion o he RRL con ol is
discussed, wi h a ho ough p esen a ion o he labo a o y se up
and he pe o med es s. Sec ion Vp esen s he new me hod o
he coo dina ed deploymen o RRL agains RVCs. The me hod
is alida ed in Sec ion VI. Finally, Sec ion VII concludes he
pape , discussing i s main indings.
II. REVIEW OF RVC DEFINITIONS
In his sec ion, he RVC de ini ions o he IEC 61000-4-
30:2015 [4] and IEEE 1547:2018 S anda ds [3] a e p esen ed.
A. IEC 61000-4-30 S anda d
In IEC 61000-4-30:2015 [4]-Class A, RVCs a e de ined o e
a e e ence in e al, Δ e , wi h a du a ion o 1-s. The de ini ion
is based on he U ms(1/2) quan i y, i.e., he ms ol age alue,
measu edo e onecycleand e esheda eachhal -cycle.Hence,
Δ e accoun s o Mcalcula ed alues o U ms(1/2), whe e Mis
100 o 120, o 50 o 60 Hz sys ems, espec i ely. In each hal -
cycle, hemoni o ed ol ageisins eady-s a e,i all Mp eceding
alues and he new U ms(1/2) alue o his hal -cycle lie wi hin a
h eshold, calcula ed as a pe cen age, h(%), o hei a i hme ic
mean, Umean
ms(1/2). Thus, an RVC s a s i a leas one o he M+1
alues o U ms(1/2) exceeds he h eshold. No e ha , du ing an
RVC e en , a hys e esis is applied o he h(%). Pe cen ages
be ween 1%–6% o U ms(1/2) o h; and a ound 50% o h, o
he hys e esis, a e sugges ed. The e o e, bo h pa ame e s a e
use -de ined.
Fu he mo e, he S anda d quan i ies RVCs h ough he ol-
lowing cha ac e is ic pa ame e s: magni ude (ΔUss), maximum
DIMOULIAS e al.: EXPERIMENTAL EVALUATION AND COORDINATED DEPLOYMENT OF RRL AGAINST RVC IN DISTRIBUTION SYSTEMS 1841
Fig. 1. Indica i e RVC e en acco ding o [4].
de ia ion (ΔUmax), and du a ion (ΔT). ΔUss is he absolu e
di e ence be ween he inal Umean
ms(1/2) alue p io o he e en
and he i s Umean
ms(1/2) alue a e he e en . ΔUmax deno es
he maximum absolu e di e ence be ween any o he U ms(1/2)
alues, du ing he e en , and he inal Umean
ms(1/2) alue, p io o i .
Essen ially, ΔUss and ΔUmax exp ess he in ensi y o an RVC.
ΔTis de ined as ΔT= − s−Δ e , whe e sand a e
he commencing and he s eady-s a e es o a ion ins an s o he
e en , espec i ely. To be e comp ehend he de ec ion p ocess,
he eade is e e ed o Fig. 1, whe e an indica i e RVC e en
is iden i ied. He e, h =1%, while UB,LB s and o he uppe
and lowe bounda y o U ms(1/2), espec i ely.
B. IEEE 1547-2018 S anda d
The IEEE 1547:2018 S anda d [3] speci ies ha an indi idual
CIRES, a MV le el, should no cause s ep o amp changes
in he ms ol age a i s poin o connec ion exceeding 3% o
nominal and 3%/s, a e aged o e a pe iod o 1-s. The espec i e
limi is 5% o he LV le el. None heless, when e e ing o he
combined e ec o all he RVC-inducing sou ces, he s ic e
limi o 3% is used o bo h ol age le els. Conce ning he
de ec ion p ocess, no guidelines a e p o ided.
III. CONSIDERED RRL CONTROL
Gene ally, RRL con ol app oaches aim o con ain he ou pu
amp a e o a CIRES uni , RRou , wi hin a desi ed, p e-de ined
limi , deno ed as RRlim. Mo e speci ically, he inpu ac i e
powe o he uni , pin[ ], is measu ed, and he co esponding
amp a e, deno ed as RRin[ ], is compa ed agains RRlim.I
|RRin[ ]|<RR
lim, he ESS does no in e ene; o he wise, i
injec s/abso bs he equi ed powe so ha |RRou [ ]|=RRlim.
In his s udy, he di ec RRL con ol o [16] is conside ed,
bo h o he expe imen aland he simula ion-basedanalysis.The
employed RRL con ol has been compa ed agains con en ional
powe smoo hing echniques, such as he mo ing a e age and
low pass il e , bo h h ough simula ions [15] and expe imen-
ally [16], demons a ing supe io pe o mance in e ms o ESS
e iciency and amp a e limi iola ions. The con ol logic
behind he conside ed RRL scheme is depic ed in Fig. 2and
p esen ed in algo i hmic o m in Algo i hm 1. The inpu s o
he algo i hm a e pin[ ], and he posi i e (max[ ]=RRlim) and
nega i e (min[ ]=-RRlim) limi s o RRou , o amp-ups and
amp-downs, espec i ely. Δ e e s o he measu emen eso-
lu ion o pin[ ]. The ou pu o he con ol block is he smoo hed,
Fig. 2. Con ol logic o he conside ed RRL echnique [16].
Algo i hm 1: Algo i hm o RRL Con ol.
Requi e: pin[ ],pou [ −Δ ],min[ ],max[ ],Δ
Ensu e: pou [ ],pESS[ ]
1: RRin[ ]←pin[ ]−pou [ −Δ ]
Δ
2: i RRin[ ]<min[ ] hen
3: RRou [ ]←min[ ]
4: else i RRin[ ]>max[ ] hen
5: RRou [ ]←max[ ]
6: else
7: RRou [ ]←RRin[ ]
8: end i
9: pou [ ]←RRou [ ]·Δ +pou [ −Δ ]
10: pESS[ ]←pin[ ]−pou [ ]
ou pu powe pou [ ], by which he e e ence powe , pESS[ ],o
he ESS is de e mined. Among pin[ ],pou [ ], and pESS[ ], i holds
ha :pESS[ ]=pin[ ]−pou [ ]. Mo ede ails ega ding hecon ol
a e a ailable in [16].
IV. EXPERIMENTAL EVALUATION OF RRL AGAINST RVCS
In his sec ion, RRL is expe imen ally e alua ed as a p e-
en i e ac ion agains RVCs. Ini ially, he labo a o y se up wi h
he employed CIRES p o o ypes is p esen ed. Subsequen ly, he
expe imen al scena ios a e ho oughly assessed acco ding o he
IEEE and IEC S anda ds.
A. CIRES P o o ypes
The RRL algo i hm o Sec ion III is inco po a ed in h ee
labo a o y ol age sou ce con e e p o o ypes (VSCPs). Mo e
speci ically, each VSCP se up ea u es a 20-kVA, h ee-phase,
h ee-wi e ol age sou ce con e e (dc/ac con e e ), wi h a ed
dc and ac ol ages o 750 V and 400 V, espec i ely. Rega ding
he employed ESS, supe capaci o s wi h a capaci ance o 6 F,
a a ed ol age o 160 V, and a maximum ins an aneous powe
o 2 kW, a e selec ed. The supe capaci o s a e coupled a he
dc link o he VSCP ia a dedica ed dc/dc con e e (Fig. 3).
No e ha , al hough he heo e ical ene gy s o age capaci y o
he supe capaci o s is 21.33 Wh, he ac ual a ailable ene gy o
RRL con ol is es ic ed o 14.58 Wh, due o limi a ions o he
dedica ed dc/dc con e e s o ope a e wi h a minimum ol age
o 90 V. The p ima y ene gy sou ce o he VSCP is emula ed
by a con ollable cu en sou ce. The con ol o he p o o ype
has been implemen ed in a Texas Ins umen s TMS320F28335
1842 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 61, NO. 2, MARCH/APRIL 2025
Fig. 3. Con igu a ion o he supe capaci o (SC) and dc/dc con e e .
Fig. 4. Single-line diag am o he scaled-down MV dis ibu ion sys em used
o alida ion pu poses.
Del ino mic ocon olle wi h a sampling pe iod o 50 μs. De ails
on he modeling and con ol o he VSCPs can be ound in
Appendix A,aswellasin[23].
B. Expe imen al Tes bed
The expe imen al es bed is a scaled-down e sion o he
CIGRE Eu opean benchma k MV eede , p esen ed in [22] and
se up in he labo a o y o Uni e sidad de Se illa. The o iginal
dis ibu ion sys em consis s o wo adial eede s, wi h leng hs
o 15 km (Feede 1) and 8 km (Feede 2), espec i ely. In he
p esen s udy, only Feede 1 o he o iginal g id has been emu-
la ed, wi h i s espec i e single-line diag am p esen ed in Fig. 4.
This subsys em comp ises 11 buses, hos ing wo cons an -powe
loads and he h ee VSCPs o Sec ion IV-A. The scaled-down
dis ibu ion sys em has been designed by con e ing he base
magni udes o he o iginal g id: om 20 kV and 10 MVA, o
400 V and 100 kVA, espec i ely; allowing o he ep oduc ion
o he beha io o he o iginal dis ibu ion sys em in e ms o
ol age d ops, powe lows, and losses, in an LV labo a o y. Fu -
he mo e, a Rega on con ollable ol age sou ce is connec ed
a Bus 0 o decouple he scaled-down expe imen al dis ibu ion
sys em om he LV supply g id o he labo a o y. Mo e speci i-
cally, he Rega on con ollable ol age sou ce aims o main ain
a cons an ol age a Bus 0, emula ing a s i , ups eam sys em.
Fig. 5(a) shows he b anches o he wo eede s and he
cabine s wi h he de ices used o emula e loads and CIRESs.
The in e io o such a cabine is shown in Fig. 5(b). De ails
on he pa ame e s o he scaled-down MV dis ibu ion sys em
and he a chi ec u e o he eal- ime con ol o he es bed a e
p o ided in [22]. The line and ans o me pa ame e s o he
es bed a e ound in Sec ion A o Appendix A, a he end o he
manusc ip .
Fig. 5. Labo a o y scaled-down MV dis ibu ion sys em: (a) Dis ibu ion
sys em o e iew, (b) Cabine in e io .
Fig. 6. Ac i e (a) and eac i e (b) powe p o iles o he loads a Bus 5 and Bus
8 o Scena io E2.
C. Expe imen al Scena ios
To expe imen ally e alua e he pe o mance o RRL in e ms
o RVC mi iga ion, wo di e en scena ios a e conside ed, each
co esponding o di e en powe p o iles o he VSCPs and
di e en consump ion pa e ns o he loads.
In Scena io E1, an a i icial s ep p o ile wi h a du a ion o
300-s and a maximum powe o 7.5 kW is used o emula e he
ola ili y o he CIRES p ima y ene gy sou ce. The p o ile is
assigned as inpu o he VSCPs. The loads a e con olled o
abso b 7 kW a uni y powe ac o .
In Scena io E2, he VSCPs a e assigned a eal, 300-s p o ile
wi h a peak powe o 5 kW. This p o ile co esponds o he 300-s
in e al ha exhibi s he mos se e e powe a ia ions in he PV
measu emen s o [16]. Mo eo e , in his scena io, he loads a e
assigned he ac i e powe consump ion p o iles o Fig. 6(a).
These p o iles de i e om he daily p o iles designa ed by
CIGRE Task Fo ce C06.04.02 [24] o dis ibu ion sys em loads,
and ha e been e-scaled o a 300-s in e al. Simila p o iles a e
conside ed o he eac i e powe consump ion, as shown in
Fig. 6(b).
DIMOULIAS e al.: EXPERIMENTAL EVALUATION AND COORDINATED DEPLOYMENT OF RRL AGAINST RVC IN DISTRIBUTION SYSTEMS 1843
Fig. 7. Ou pu ac i e powe p o ile o VSCP2 o Cases 1-3: (a) Scena io E1;
(b) Scena io E2.
Fo bo h scena ios, h ee di e en cases a e examined: no
RRL con ol applied o he VSCPs (Case 1), he RRL scheme
o [16] wi h RRlim=0.01 pu/s o all VSCPs (Case 2), and he
RRLschemeo [16] wi h RRlim=0.005 pu/s o all VSCPs (Case
3). Case 1 is he base case ha highligh s he RVC-inducing
e ec o CIRES ola ili y. The RRlim alues o Cases 2-3 a e
exp essedinpe uni (pu) wi h espec o he a ingo eachVSCP
and a e ela i ely elaxed, co esponding o 60 %/min and 30
%/min, espec i ely, hus being signi ican ly less s ic han he
commonly sugges ed limi o 10 %/min [17],[19],[25]. These
lax RRlim alues a e selec ed o p e en he excessi e s essing
o he supe capaci o and i s dc/dc con e e . The achie ed
smoo hing e ec is indica ed in Fig. 7(a) and (b), by he ou pu
ac i e powe o VSCP2, o Scena ios E1 and E2, espec i ely.
The RVC-supp essing capabili y o he implemen ed RRL
con ol is assessed acco ding o he IEEE 1547:2018 and IEC
61000-4-30:2015-Class A S anda ds. The wo S anda ds di e
ega ding he equi ed measu ing sampling a e. Mo e speci i-
cally, p o ided ha he IEC S anda d is based on he U ms(1/2)
concep , measu emen s wi h a sampling pe iod o up o 10ms
a e equi ed. Con e sely, he equi ed sampling pe iod is unde-
ined in IEEE 1547:2018. The only speci ica ion conce ns he
examina ion o he a e o change o ol age (RoCoV) o e an
1-s pe iod. The e o e, any sampling pe iod o up o 0.5s would
su ice. Due o he di e en esolu ions o he ol age me e s
connec eda eachbuso he es bed, he ollowingmeasu emen s
a e a ailable: (a) ms measu emen s, wi h a esolu ion o 0.5s,
o Buses 2-11; (b)ins an aneous measu emen s, sampled e e y
50 μs, o Bus 1. Consequen ly, he RRL e ec is assessed
acco ding o IEEE 1547:2018 o Buses 2-11, and as pe he
IEC S anda d o Bus 1.
D. Resul s Acco ding o IEEE 1547:2018
Resul s a e p esen ed o Buses 5, 6, 8, 9, and 11, whe e he
loads and VSCPs a e connec ed, as depic ed in Fig. 4.
Fig. 8. Scena io E1: Vol age p o iles wi h espec o ime: (a) Bus 8 - Case 1;
(b) Bus 8 - Case 3; (c) Bus 6 - Case 1; (d) Bus 6 - Case 3.
TABLE I
SCENARIO E1: RVCSATLOAD BUSES ACCORDING TO THE IEEE STANDARD
TABLE II
SCENARIO E1: RVCSATVSCP BUSES ACCORDING TO THE IEEE STANDARD
1) Scena io E1: The capabili y o he RRL algo i hm o
supp essRVCphenomenais isualizedinFig.8,wi h he ol age
p o iles o Buses 6 and 8 o Cases 1 and 3. By compa ing
Fig. 8(a) and (c) wi h Fig. 8(b) and (d), espec i ely, i is a es ed
ha he RRL applica ion supp esses he de ec ed RVC e en s o
Case 1, signi ican ly imp o ing he smoo hness o he ol age
p o iles. Mo eo e , as shown in Fig. 8, he VSCP bus (Bus 6) ex-
hibi s mo e equen and signi ican ol age a ia ions compa ed
o he load bus (Bus 8). This is due o he highe o al impedance
in he pa h be ween Bus 0 and Bus 6. Speci ically, as indica ed
by Tables IX and X, he o al impedance in he pa h be ween Bus
0 and Bus 6 is 281.1+j262.9 mΩ, whe eas he o al impedance
be ween Bus 0 and Bus 8 is only 211.1+j246.9 mΩ. This highe
o al impedance, and mo e impo an ly, he signi ican ly highe
o al esis ance in he pa h o Bus 6, ende s he ol age a Bus 6
mo e sensi i e o ac i e powe a ia ions [26], esul ing in mo e
se e e ol age luc ua ions.
The o e all e ec o he RRL con ol is agg ega ed o all
buses and cases in Tables Iand II. The impac is quan i ied
h ough he o al numbe o RVC e en s, N, and he maximum
1844 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 61, NO. 2, MARCH/APRIL 2025
Fig. 9. Scena io E2: Vol age a Bus 6.
Fig. 10. Scena io E2: Numbe o RVC e en s pe VSCP bus and case. The
maximum RoCoV (%/s) o each bus and case is shown abo e he ba s.
RoCoV, dV
d (%/s), o hese e en s. As shown in Table I, o
Cases 2-3, he applica ion o RRL, unde bo h RRlim alues,
elimina es all RVC e en s a load buses. Rega ding VSCP buses,
he esul s o Table II showcase ha he applica ion o RRL wi h
RRlim =0.01 pu/s (Case 2) achie es up o a 60% educ ion
in he o al numbe o e en s and up o a 40% educ ion in
he maximum RoCoV, compa ed o Case 1. E en ually, he use
o RRlim =0.005 pu/s (Case 3) comple ely supp esses RVCs,
es o ing ol age quali y.
2) Scena io E2: Fo Scena io E2, he RVC-supp essing e -
ec o he employed RRL con ol is showcased in Fig. 9, whe e
he ol age o Bus 6 o Case 1 and Case 3 is depic ed. As shown,
he RRL algo i hm e ec i ely smoo hs he ol age, mi iga ing
i s luc ua ions and ab up a ia ions. The o e all pe o mance
o he RRL algo i hm is demons a ed in Fig. 10, by he o al
numbe o induced RVCs a VSCP buses and he maximum
RoCoV (%/s) pe case. As obse ed, he applica ion o RRL
wi h RRlim =0.01 pu/s (Case 2) signi ican ly imp o es ol age
quali y.E en ually, o Case3,RVCsa ecomple elysupp essed,
and he maximum RoCoV alues (%/s) a e con ained below 1
%/s. Rega ding load buses, no RVCs we e epo ed.
The es s o Scena io E1 and Scena io E2 ha e expe imen ally
alida ed he e ec i eness o RRL con ol as an RVC-mi iga ion
measu e unde di e se condi ions o CIRES powe ola ili yand
load consump ion. Compa a i ely, he condi ions o Scena io
E2 esul ed in ewe and less se e e RVC e en s han hose
o Scena io E1. This is a ibu ed o he mo e g adual powe
a ia ions o he VSCPs and he in luence o he loads, whose
ac i e and eac i e powe a ia ions pa ially balance he CIRES
ola ili y.
E. Resul s Acco ding o IEC 61000-4-30:2015-Class a
The RVC-supp essing capabili y o he RRL con ol is u he
es ed unde he IEC de ini ion, examining he changes in he
Fig. 11. RVC e en a Bus 1: (a) Case 1 - h =2%; (b) Case 1 - h =1%;
(c) Case 3 - h =1%. In (a) and (b), ed do s indica e he s a ing and ending
ins an s o RVC e en s.
ol age o Bus 1 o Scena io E1. No e ha he condi ions o
Scena io E1 ha e been selec ed, as hey co espond o mo e
se e e powe and ol age a ia ions. P io o discussing he
pe o mance o he RRL con ol, i is impo an o highligh
he s ong dependency o he de ec ion p ocess and he RVC
du a ion on he use -de ined h eshold, h(%). This dependency
is obse ed in Fig. 11, whe e a ce ain RVC e en is depic ed.
Speci ically, Fig. 11(a) and (b) ep esen he same de ec ed
e en , bu o h =2% and h =1%, espec i ely. In he la e
case, due o he s ic e h eshold, he RVC e en is igge ed
ea lie , esul ing in a longe du a ion ΔT. Fo he sake o
be e compa ison, he s a ime, s=116.55 s, o he e en o
h =2%, is also s amped on Fig. 11(b). Rega ding he e ec
o RRL, applying he conside ed echnique o all VSCPs wi h
RRlim =0.005 pu/s supp esses he e en , con aining ol age
be ween he uppe (UB) and lowe bounda y (LB).
The o e all esul s o he 300-s analysis pe iod and he h ee
dis inc cases a e summa ized in Table III, which p esen s he
o al numbe o e en s and he maximum epo ed alues o he
cha ac e is ic RVC pa ame e s. Rega ding Case 1, he esul s
highligh he e ec o h on he numbe o RVCs, as Nis
almos doubled i h is se o 1%. Fu he mo e, he RRL scheme
e ec i elymi iga esRVCs, educing hei numbe ,du a ion,and
in ensi y. Mo e speci ically, o Case 2, he numbe o e en s is
educed by 75%, compa ed o Case 1, o bo h h alues. Fo
DIMOULIAS e al.: EXPERIMENTAL EVALUATION AND COORDINATED DEPLOYMENT OF RRL AGAINST RVC IN DISTRIBUTION SYSTEMS 1845
TABLE III
RVCSATBUS 1ACCORDING TO THE IEC STANDARD
Algo i hm 2: B eakdown o he Pa ame e -Acquisi ion
S age.
1: Selec a wo s -case p o ile o he p ima y ene gy
sou ce.
2: De e mine ep esen a i e buses o he dis ibu ion
sys em unde s udy.
3: Calcula e he ol age magni ude-ac i e powe
sensi i i y ac o s be ween he buses o S ep-2 and he
CIRES buses.
4: Fo he p o ile o S ep-1, de ec he induced RVCs a
he selec ed buses o S ep-2, as pe he IEC S anda d.
5: Fo each bus o S ep-2, iden i y he mos se e e RVC
and s o e he M alues o U ms(1/2) be o e i s
occu ence.
Case 3, he achie ed educ ion eaches 92%, owing o he s ic e
RRlim alue.
V. PROPOSED METHOD FOR EFFICIENT DEPLOYMENT OF RRL
AGAINST RVCS
Ha ing alida ed he RVC-supp essing e ec o RRL
schemes in Sec ion IV, his sec ion p oposes a new sys em-
planning me hod o he coo dina ed mi iga ion o RVCs in
ac i e dis ibu ion sys ems. Mo e speci ically, o a gi en dis i-
bu ionsys em, heme hodde e mines heCIRESuni s oope a e
unde RRL con ol, as well as he necessa y RRlim alues, o
mi iga e RVCs unde minimum echnical equi emen s, such as
henumbe o CIRESuni swi hRRLcon olo he o al equi ed
ESS capaci y. No e ha he me hod add esses RVCs acco ding
o he IEC S anda d, due o i s mo e comp ehensi e guidelines.
The me hod consis s o wo s ages: pa ame e acquisi ion and
o mula ion o he op imiza ion p oblem.
A. Pa ame e -Acquisi ion S age
The solu ion o he me hod de i es om an op imiza ion
p oblem. Fo he o mula ion o he op imiza ion p oblem, a
pa ame e -acquisi ion s age needs o p ecede. Mo e speci ically,
his s age p o ides a amewo k o add essing a wo s -case
scena io in e ms o RVCs. As he me hod is sol ed o line o
sys em-planning pu poses, he pa ame e -acquisi ion s age is e-
qui ed obeexecu edonly i one o he ollowingchanges occu s
in he dis ibu ion sys em: (a)a new CIRES is connec ed; (b)an
exis ing CIRES is decommissioned; (c)a change occu s ela ed
o he lines o ans o me s o he g id. The algo i hmic s eps o
he pa ame e -acquisi ion s age, as in ended o be ollowed by
DSOs, a e summa ized in Algo i hm 2and ho oughly discussed
below:
S ep-1: Toensu e hep e en iono RVCsunde allcondi ions,
he RRL alloca ion needs o be dimensioned acco ding o a
wo s -case, ola ile inpu p o ile, co esponding o he p ima y
ene gy sou ce o CIRESs. I a ailable, DSOs can de i e ola ile
p o iles by analyzing his o ical da ase s wi h measu emen s o
CIRESac i epowe om hei dis ibu ionsys emo byconsul -
ing his o ical da a o wea he condi ions, e.g., [27],[28], om
he speci ic loca ion o hei dis ibu ion sys em. Al e na i ely,
DSOs can selec benchma k ola ile p o iles a ailable in he
echnical li e a u e, such as hose o [9],[16]. I should be no ed
ha he compac size o dis ibu ion sys ems allows o he
assump ion o uni o m wea he condi ions ac oss hem, e.g.,
sola i adiance, ensu ing ha all CIRES uni s a e subjec o
he same inpu p o ile. Mo eo e , his uni o mi y ep esen s he
wo s -case scena io in e ms o he induced ol age a ia ions,
as he changes in powe o all CIRES uni s will be synch onized,
and hus, hei e ec on ol age maximized.
S ep-2: The s a egy aims o ensu e he o al supp ession o
RVC e en s o all buses o a dis ibu ion sys em. Ne e heless,
i is imp ac ical o examine all buses while implemen ing he
me hod. The e o e, he s a egy is based on he easonable se-
lec ion o some ep esen a i e buses, acco ding o he ollowing
logic: I he eme gence o RVCs is p e en ed o hose buses,
hen i is sa e o assume ha i has been p e en ed o all ne wo k
buses. These ep esen a i e buses can be selec ed by consul ing
he ol age- o-ac i e powe sensi i i y ma ix o he dis ibu ion
sys em and employing ne wo k pa i ioning echniques [29],
[30].
S ep-3: Subsequen ly, he ol age- o-ac i e powe sensi i -
i y ac o s be ween he ep esen a i e buses o S ep-2 and he
CIRES buses a e calcula ed. The calcula ion can ei he be pe -
o med iaapowe sys emsanalysisso wa eo acco ding oap-
p op ia e scien i ic wo ks, e.g. [26],[31],[32]. I should be no ed
ha he calcula ion o he sensi i i y ac o s only equi es he
knowledge o he opology, as well as o he line and ans o me
impedanceso hedis ibu ionsys em,whichis usuallya ailable
o he DSOs. Mo eo e , in case he p ecise dis ibu ion sys em
pa ame e s a e no ully known, measu emen -based echniques,
e.g., [33],[34],[35], can be employed o es ima e he equi ed
pa ame e s, and h ough hem, he sensi i i y ma ix.
S ep-4: Execu ing ms simula ions on he examined dis ibu-
ion sys em unde he p o ile o S ep-1, he RVCs a he buses o
S ep-2 a e iden i ied. The powe o he es dis ibu ion sys em
componen s, e.g., loads, is no al e ed, hence RVCs only depend
on CIRES ola ili y.
S ep-5: The mos ab up RVC e en o each bus is iden i ied.
Ab up ness is de ined as he di e ence be ween he ms ol age
immedia ely be o e and a e he onse o an e en . This e en
1846 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 61, NO. 2, MARCH/APRIL 2025
co esponds o he mos ab up ac i e powe change and, ea-
sonably, exhibi s he s ic es equi emen s o mi iga ion. Thus,
i is a ional o assume ha he success ul supp ession o his
e en implies he supp ession o all RVCs. Fo his e en , he M
alues o U ms(1/2) be o e i s occu ence a e s o ed.
In summa y, he p o ile o S ep-1, he sensi i i y ac o s o
S ep-3, and he M ol age alues pe examined bus o S ep-5 a e
he pa ame e s equi ed o he o mula ion o he op imiza ion
p oblem, which is p esen ed in Sec ion V-B.
B. Op imiza ion P oblem Fo mula ion
Le e aging he pa ame e s ob ained in Sec ion V-A, an op i-
miza ion p oblem is o mula ed o e icien ly alloca e he RRL
unc ionali ies among CIRES uni s, o he mi iga ion o RVCs.
The p oblem is cen ally sol ed by he DSO a hei p emises
and is o mula ed as a mixed-in ege linea p og amming p ob-
lem o ensu e low compu a ional complexi y and acili a e i s
execu ion.
Mo e speci ically, le NCIRES be he se o buses wi h
CIRESs, N be he se o ep esen a i e buses, and K=
{1,2,...,M}be an in ege se , wi h pa ame e Mde ined
in Sec ion II-A. The ol age- o-ac i e powe sensi i i y ac o
be ween a bus i∈N
and bus j∈N
CIRES is deno ed as ΔUi
ΔPj,
while he a ing o he CIRES uni a bus jis Sj. Mo eo e ,
Uk
pas ,i deno es he k- h p eceding ol age alue be o e he mos
se e e RVC e en a bus i, wi h k∈K,i∈N
.
The op imiza ion p oblem ea s he Uk
pas ,i alues as known
pa ame e s and de e mines he equi ed amp a es o he CIRES
uni s o henex hal -cycle,such ha he occu enceo hisRVC
e en would ha e been p e en ed. Elabo a ing, he mos se e e
RVC e en a each bus i∈N
would be a oided, i he ms
ol age alue o he nex hal -cycle we e such ha all Uk
pas ,i
alues, and he new alue, lied wi hin he h eshold o hei
a i hme ic mean alue. The e o e, le U∗
ideno e his desi ed
ol age alue a bus i∈N
o he nex hal -cycle, so ha he
RVC e en is no igge ed. This desi ed U∗
iwill esul in a
alue o he a i hme ic mean o he M+1 hal -cycle ol age
alues, deno ed as U∗
mean,i. E en ually, in o de o he desi ed
U∗
i alue o be achie ed and he RVC e en o be p e en ed, he
amp a e o he CIRESs o he nex hal cycle should be RR∗
j,
j∈N
CIRES. Based on he abo e, he op imiza ion p oblem is
ma hema ically o mula ed as ollows:
min (RR∗
j),j ∈N
CIRES (1)
s. .
(1− h)·U∗
mean,i ≤Uk
pas ,i ≤(1+ h)·U∗
mean,i,i∈N ,k∈K
(2)
(1− h)·U∗
mean,i ≤U∗
i≤(1+ h)·U∗
mean,i,i∈N
(3)
U∗
mean,i =1
M+1(U∗
i+
M
k=1
Uk
pas ,i),i∈N
(4)
U∗
i=UM
pas ,i +
j∈NCIRES
ΔUi
ΔPj
·Sj·RR∗
j·Δ hc,i∈N
(5)
RR∗
j=g(RR),j ∈N
CIRES (6)
Equa ion(1)exp esses heobjec i e unc iono he op imiza ion
p oblem, which is a unc ion o he equi ed amp a es. Eq. (2)
and (3) ep esen he condi ions o he p e en ion o he RVC
e en , while (4) is he de ini ion o he U∗
mean,i.In(2) and (3),
h is he conside ed pe cen age alue o he h eshold. Finally,
(5) ep esen s he linea ized equa ion ha co ela es he ms
ol age o he nex hal -cycle, U∗
i, wi h he ms ol age alue
o he cu en hal -cycle, UM
pas ,i, wi h espec o he sensi i i y
ac o s and he amp a es o CIRES uni s. In (5),Δ hc is equal
o he hal -cycle du a ion, while RR∗
jis exp essed in pu/s wi h
espec o Sj.Eq.(6) de e mines RR∗
j om a pe missible ange
o amp a e alues, ep esen ed by ec o RR.
Equa ions(1) and(6) a eabs ac ly exp essed,as hei speci ic
o m depends on he pa icula e sion o he op imiza ion
p oblem. In he p esen s udy, h ee dis inc e sions o he
op imiza ion p oblem a e add essed, as ollows:
P oblem 1: Conside ing a ixed RRlim alue, minimize he
numbe o CIRES uni s ope a ing unde RRL con ol.
P oblem 2: Conside ing a ixed RRlim alue, minimize he
o al equi ed ESS capaci y.
P oblem 3: Conside ing a wide ange o applicable RRlim
alues, minimize he o al equi ed ESS capaci y.
Elabo a ing on he conside ed p oblems:
P oblem 1: He e,CIRES uni scanei he ope a ewi hou RRL
con ol o unde RRL con ol wi h RRlim. Thus, he pe missible
ange o alues o RRjis RR =[RRlim,RR
in], wi h RRin
being he known, inpu amp a e o he CIRESs o he nex
hal -cycle, as dic a ed by he selec ed ola ile p o ile. The e o e,
(6) co esponds o:
RR∗
j=(1−b∗
j)·RRin +b∗
j·RRlim,j ∈N
CIRES (7)
A bina y a iable, b∗
j, is in oduced o each CIRES uni . I
b∗
j=1, he co esponding uni is decided o ope a e unde he
RRL scheme and be in ol ed in he RVC mi iga ion. Thus, he
objec i e unc ion becomes:
min
j∈NCIRES
b∗
j(8)
P oblem 2: This o mula ion p ese es (7) bu akes in o accoun
he di e en powe a ings among CIRES uni s, which, in u n,
a ec he ene gy a ings o hei co esponding ESSs. Mo e
speci ically, o he same inpu powe p o ile and unde he same
RRlim, he equi ed maximum ene gy capaci y o an ESS is
p opo ional o he powe a ing o i s associa ed CIRES uni .
The e o e, he objec i e unc ion becomes:
min
j∈NCIRES
b∗
j· j(9)
DIMOULIAS e al.: EXPERIMENTAL EVALUATION AND COORDINATED DEPLOYMENT OF RRL AGAINST RVC IN DISTRIBUTION SYSTEMS 1847
He e, jis he a ed-powe a io be ween he CIRES uni a bus j
and he CIRES uni wi h he lowes a ed powe . Thus, (9) min-
imizes he equi emen s in agg ega e ESS capaci y, assigning
weigh s o CIRESs acco ding o hei powe a ing.
P oblem 3: This is he mos gene alized o mula ion, whe e
a disc e e ange o RRlim alues is assumed. The RRlim alues
and he RRin alue comp ise se NRRL and a e deno ed as
RRm,m∈N
RRL. Each uni could ope a e unde one o hese
alues o wi hou RRL con ol. This is exp essed by (10),
h ough he usage o bina y a iables, bm∗
j.
m∈NRRL
bm∗
j=1,j ∈N
CIRES (10)
The e o e, (6) becomes (11), as ollows:
RR∗
j=
m∈NRRL
bm∗
j·RRm,j ∈N
CIRES (11)
The op imiza ion c i e ion is, simila ly o P oblem 2, he min-
imiza ion o he o al equi ed ESS capaci y. In his p oblem,
he ene gy capaci y equi emen s o an ESS depend on bo h he
a ing o he associa ed CIRES and he RRlim alue. To es ima e
he e ec o RRlim on ene gy a ing, he ene gy p o ile co e-
sponding o each examined RRlim is heo e ically calcula ed,
o he CIRES wi h he lowes a ed powe . The esul ing ene gy
usage,alongwi h he CIRES a ings, p oduce weigh ac o s ha
ep esen he ene gy capaci y equi emen o CIRESs. These
ac o s a e deno ed as cm
j,j∈N
CIRES,m∈N
RRL. Hence,
(1) becomes:
min
j∈NCIRES
(
m∈NRRL
bm∗
j·cm
j)(12)
VI. VALIDATION OF THE PROPOSED METHOD
The alida ion o he p oposed me hod is pe o med as ol-
lows: Ini ially, he sys em unde s udy is p esen ed, o which
he pa ame e -acquisi ion s age is execu ed. Subsequen ly, he
me hod is applied, de i ing he op imized solu ion o he
deploymen o RRL con ol. Finally, he e ec i eness o he
de i ed solu ion in supp essing he RVCs o he sys em unde
s udy is es ed ia dynamic ms simula ions.
A. Sys em Unde S udy
Thep oposed s a egyisapplied o heIEEEEu opean LV es
eede [36]. This dis ibu ion sys em consis s o 906 buses and
55 loads. In his s udy, he loads a e modeled as h ee-phase,
cons an -powe , and ohmic, wi h a consump ion o 3kW. The
sho ci cui capaci y o he ups eam eede is 63 MVA. No e
ha his dis ibu ion sys em is selec ed, ins ead o he CIGRE
MV eede o Sec ion IV, as i s ex ended size and complexi y
ende i mo e sui able o demons a ing he e ec i eness o he
p oposed me hod in alloca ing he RRL con ol among mul iple
CIRESs.
To simula e scena ios o high CIRES pene a ion, 32 PV uni s
a e dispe sed ac oss he dis ibu ion sys em. Fo he same PV
loca ions, wo dis inc scena ios a e examined, co esponding
o di e en a ed powe alues pe PV uni . The wo scena ios,
TABLE IV
PV RATINGS PER SCENARIO
TABLE V
RVCSOFTHEBASE CASE (ΔUMAX AND ΔUSS IN %)
alongwi h he opology o he dis ibu ionsys em,a e isualized
in Fig. 12. As depic ed, in Scena io S1, he 16 mos emo e
PVs ha e a 20-kWp a ing, while he 16 mos p oximal o he
subs a ion PVs ha e a 10-kWp a ing. Scena io S2 ega ds he
opposi e condi ion. The a ed powe and he loca ion o he PVs
o bo h scena ios a e mapped in Table IV.
B. Execu ion o Pa ame e -Acquisi ion S age
He e, he pa ame e -acquisi ion s age is execu ed, as ollows:
S ep 1: The ola ili y o he PV p ima y ene gy sou ce is
modeledby heinpu ac i epowe p o ileo Fig.13.Speci ically,
o de i e his p o ile, he amp a es o he eal, daily PV p o ile
o [16] we e analyzed, and he 600-s window wi h he mos
se e e powe a ia ions was isola ed. The co esponding inpu
amp a es a ealso depic edinFig.13. Thep o iles a eexp essed
in pu o apply o dis inc PV a ings.
S ep 2: Fo he examined dis ibu ion sys em, Buses 619,
861, and 882 a e selec ed as ep esen a i e. As obse ed in
Fig. 12, hese buses a e emo ely loca ed in dis inc a eas o
he dis ibu ion sys em.
S ep 3: The ol age- o-powe sensi i i y ac o s a e calcula ed
h ough he DIgSILENT Powe ac o y so wa e.
S ep 4: Using Powe ac o y, ms simula ions a e execu ed
o in es iga e he induced RVCs unde he p o ile o Fig. 13
a he selec ed buses. The simula ion ime s ep is 10ms. The
induced RVCs o Scena ios S1, S2 a e agg ega ed in Table V,
as pe he IEC S anda d. Speci ically, he o al numbe o e en s,
N, and he maximum alues o ΔUmax and ΔUss, o hose N
e en s, a e eco ded. The lowe alue o h =1% is conside ed,
o gene a e a base case wi h mul iple RVC e en s. As shown,
al hough he o al PV capaci y in bo h scena ios is 480 kWp,
RVCs in Scena io S1 a e mo e equen and o highe in ensi y,
due o he highe a ing o he PVs a he emo e buses. O e all,
up o 9 RVCs a e eco ded, o a 600-s pe iod, i.e., 1/6 o an
hou . As a compa ison, he IEEE 1547-2018 S anda d limi s
he allowable numbe o RVCs o only 2 ≤N≤10 pe hou .
