Design, modeling and simulation of a PV power plant
Abstract
The rise of the photovoltaic energy during the last decades and mainly during the last years has many indicators; the fast growth of the global production capacity, the innovations related to the solar energy technologies or the continuous adaptation of the law code to fit a wider variety of scenarios and conditions. Additionally the recent improvements have given Large Scale PV power plant (LS-PVPP) the capability to work as baseload power plants relieving dirtier sources of energies. However one of the biggest disadvantages in front of conventional energy sources is the cost. In the last years PV technologies have achieved higher cost-effective ratio and now are able to compete against non-renewable production methods. For this reason the present work is centered in the development of a tool for integration of LS-PVPP in the actual energy distribution system that allows the user to optimize and study the design and functionality of the plant. A MATLAB based program and different functions have been developed to perform a power flow analysis. Different grid codes requirements have been analyzed for LS-PVPP. Approaches from different years are taken into account and the tendency of unifying some parts of these grid codes to achieve higher levels of energy share is discussed too. Examples of control implementation to meet these requirements are also included. The example used to test the Power flow equations (PFE) solver is composed of central inverters. According to [2] this configuration is the most cost-effective nowadays despite the string inverters are close and present a series of benefits related with the versatility and control
Full text
Uni e si a Poli `ecnica de Ca alunya
Escola T`
ecnica Supe io d’Enginye ia Indus ial
de Ba celona. ETSEIB
Design, modeling and
simula ion o a PV powe
plan .
Bachelo ’s inal p ojec
Au o : Pablo Poza Fe agu
Supe iso : O iol Gomis Bellmun
Junio 2020
Summa y
The ise o he pho o ol aic ene gy du ing he las decades and mainly du ing he las
yea s has many indica o s; he as g ow h o he global p oduc ion capaci y, he inno-
a ions ela ed o he sola ene gy echnologies o he con inuous adap a ion o he law
code o i a wide a ie y o scena ios and condi ions.
Addi ionally he ecen imp o emen s ha e gi en La ge Scale PV powe plan (LS-PVPP)
he capabili y o wo k as baseload powe plan s elie ing di ie sou ces o ene gies. How-
e e one o he bigges disad an ages in on o con en ional ene gy sou ces is he cos .
In he las yea s PV echnologies ha e achie ed highe cos -e ec i e a io and now a e
able o compe e agains non- enewable p oduc ion me hods.
Fo his eason he p esen wo k is cen e ed in he de elopmen o a ool o in eg a ion
o LS-PVPP in he ac ual ene gy dis ibu ion sys em ha allows he use o op imize
and s udy he design and unc ionali y o he plan .
A MATLAB based p og am and di e en unc ions ha e been de eloped o pe o m a
powe low analysis. Di e en g id codes equi emen s ha e been analyzed o LS-PVPP.
App oaches om di e en yea s a e aken in o accoun and he endency o uni ying
some pa s o hese g id codes o achie e highe le els o ene gy sha e is discussed oo.
Examples o con ol implemen a ion o mee hese equi emen s a e also included.
The example used o es he Powe low equa ions (PFE) sol e is composed o cen al
in e e s. Acco ding o [2] his con igu a ion is he mos cos -e ec i e nowadays despi e
he s ing in e e s a e close and p esen a se ies o bene i s ela ed wi h he e sa ili y
and con ol.
Con en s
1 P e ase 8
1.1 Objec i e .................................... 8
1.2 Mo i a ion ................................... 8
2 Powe low analysis 9
2.1 P ocedu e.................................... 9
2.2 Powe low equa ion basis . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Fo mula ion................................... 11
2.3.1 Bus admi ance ma ix . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Limi a ions and conside a ions . . . . . . . . . . . . . . . . . . . . . . . . 12
2.5 Solu ionme hods. ............................... 12
2.6 De ini ions.................................... 13
2.6.1 Slack bus................................. 13
2.6.2 Load bus; PQ ype. . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.6.3 Vol age con olled bus; PV ype. . . . . . . . . . . . . . . . . . . . 14
2.7 The New on-Raphson powe low solu ion. . . . . . . . . . . . . . . . . . . 14
2.7.1 P oblem gene al o m . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.7.2 Mul i-Va iable NRA . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.8 Load lows o he sys em . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3 Valida ing he esul s 18
3.1 Conside a ions.................................. 18
3.1.1 Limi a ions ............................... 18
3.1.2 Dis ibu ion sys em limi a ions . . . . . . . . . . . . . . . . . . . . 19
1
3.1.3 Pa ame e s unde s udy. . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 Examplecase................................... 20
3.2.1 Pe uni alues ............................. 21
3.2.2 Ini ial alues and supposi ions . . . . . . . . . . . . . . . . . . . . 21
3.3 Resul compa ison e sus MATPOWER . . . . . . . . . . . . . . . . . . . 23
3.3.1 Fi s compa ison: Ac i e and eac i e powe demand a ia ion wi h
PQbuses................................. 23
3.3.2 Second compa ison: Ac i e and eac i e powe gene a ion a ia ion
wi hPQbuses.............................. 26
3.3.3 Thi d compa ision: Ac i e powe gene a ion and g id ol age a i-
a ionwi hPVbuses........................... 28
3.4 Analysis o maximum and minimum ol age de ia ions and loses. . . . . . 30
3.4.1 Pa icula case: di e en alues o each con e e ou pu . . . . . 30
4 G id code equi emen s 33
4.1 In oduc ion................................... 33
4.2 S a ic egula ion ................................ 35
4.2.1 Powe ac o egula ion . . . . . . . . . . . . . . . . . . . . . . . . 35
4.2.2 Powe Cu ailmen . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.2.3 Ac i e powe ese es . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.2.4 Vol age ange and Con ol . . . . . . . . . . . . . . . . . . . . . . . 37
4.2.5 Remo e Vol age Con ol . . . . . . . . . . . . . . . . . . . . . . . . 37
4.2.6 F equency................................ 37
4.2.7 Flicke .................................. 38
4.3 G id suppo ; eac i e powe compensa ion . . . . . . . . . . . . . . . . . 38
5 En i onmen al impac analysis 42
5.1 La ge scale PV powe plan . . . . . . . . . . . . . . . . . . . . . . . . . . 42
5.1.1 Cons uc ion .............................. 43
5.1.2 Ope a ion................................ 43
5.1.3 Decommission.............................. 44
5.1.4 Po en ial en i onmen al consequences . . . . . . . . . . . . . . . . 44
5.1.5 Po en ial ecological consequences . . . . . . . . . . . . . . . . . . . 44
5.1.6 T ansmission sys em impac . . . . . . . . . . . . . . . . . . . . . . 45
5.1.7 Conclusions............................... 45
2
6 P ojec s budge 47
6.1 La ge Scale PV powe plan . . . . . . . . . . . . . . . . . . . . . . . . . . 47
6.2 Ancilla yse ices................................ 48
6.3 In oduc ion................................... 51
6.4 Sc ip s...................................... 51
6.5 Fi s compa ison powe low esul s . . . . . . . . . . . . . . . . . . . . . . 52
6.6 Second compa ison powe low esul s . . . . . . . . . . . . . . . . . . . . 54
6.7 Thi d compa ison powe low esul s . . . . . . . . . . . . . . . . . . . . . 55
3
Lis o Figu es
2.1 Example o a 5 bus sys em wi h he impedance o he lines. . . . . . . . . 10
3.1 Rela ion be ween he Pdc and Vdc and main poin s o analyze. . . . . . . . 19
3.2 Example case wi h designa ed buses. . . . . . . . . . . . . . . . . . . . . . 20
3.3 Ele en h bus alues (1s case) . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.4 Second bus alues (1s case) . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.5 Se en h bus alues (1s case) . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.6 Ele en h bus alues (2nd case) . . . . . . . . . . . . . . . . . . . . . . . . 27
3.7 Second bus alues (2nd case) . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.8 Se en h bus alues (2nd case) . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.9 Ele en h bus alues (3 d case) . . . . . . . . . . . . . . . . . . . . . . . . 29
3.10 Second bus alues (3 d case) . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.11 Se en h bus alues (3 d case) . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.12 Example scheme wi h ol age magni ude and phase alues . . . . . . . . . 31
3.13 Example scheme wi h ol age magni ude and phase alues . . . . . . . . . 31
3.14 Real and Imagina y losses changing he g id ol age om 0.9 o 1.1 . . . 32
4.1 Powe ins alled in Spain om 2007 o 2018. . . . . . . . . . . . . . . . . . 34
4.2 Summa y o he o al ene gy managed by he ancilla y se ices in he
spanishpeninsula. ............................... 34
4.3 Reac i e powe equi emen s o Pue o Rico, Sou h A ica and China,
Ge manyandRomania. ............................ 35
4.4 Powe cu ailmen equi emen s . . . . . . . . . . . . . . . . . . . . . . . . 36
4.5 Summa y o g id codes equi emen s o LS-PVPP . . . . . . . . . . . . . 38
4.6 Scheme o he connec ion poin s con ex ualized in he cu en example . . 39
4
4.7 F om le o igh ; ou pu PoC and g id PoC phaso diag ams when PF is
se ouni y.................................... 40
6.1 B eakdown o ancilla y se ices cos in he a e age inal p ice o ene gy in
hepeninsula sys em ............................. 49
6.2 Ac i e and eac i e powe demand (1s case) . . . . . . . . . . . . . . . . 52
6.3 Ac i e and eac i e powe gene a ion (1s case) . . . . . . . . . . . . . . . 52
6.4 Imagina y and eal losses (1s case) . . . . . . . . . . . . . . . . . . . . . . 53
6.5 Ac i e and eac i e powe demand (2nd case) . . . . . . . . . . . . . . . . 54
6.6 Ac i e and eac i e powe gene a ion (2nd case) . . . . . . . . . . . . . . 54
6.7 Ac i e and eac i e powe gene a ion (2nd case) . . . . . . . . . . . . . . 54
6.8 Ac i e and eac i e powe demand (3 d case) . . . . . . . . . . . . . . . . 55
6.9 Ac i e and eac i e powe gene a ion (3 d case) . . . . . . . . . . . . . . . 55
6.10 Ac i e and eac i e powe gene a ion (3 d case) . . . . . . . . . . . . . . . 55
5
Lis o Tables
3.1 P esen wo k sys em pa ame e s. . . . . . . . . . . . . . . . . . . . . . . . 21
3.2 Example’s pa ame e s in pu. . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3 Requi ed pa ame e s o each bus. . . . . . . . . . . . . . . . . . . . . . . 22
3.4 Requi ed pa ame e s o each bus. . . . . . . . . . . . . . . . . . . . . . . 22
3.5 Selec ed a iables and anges . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.6 Selec ed a iables and anges . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.7 Selec ed a iables and anges . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.8 Ini ial alues o he con e e ou pu buses. . . . . . . . . . . . . . . . . . 30
5.1 Compa ison o li e cycle emissions o sola echnologies and con en ional
ca bon-in ensi e sys ems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
6.1 PV plan CAPEX cos s and cu en p ojec budge . . . . . . . . . . . . . 48
6.2 PVplan OPEX ................................ 48
6
Ac onyms
DES Dis ibu ed ene gy sis ems. 19, 36, 37
ENSTO-E Eu opean Ne wo k o T ansmission Sys em Ope a o s o Elec ici y. 33
eSCR Equi alen sho ci cui a io. 40
LS-PVPP La ge Scale PV powe plan . 1, 4, 34, 37–40, 42–45, 47
MPP maximum powe poin . 19
MPPT Maximum powe poin acke . 30
NRA New hon-Raphson algo i hm. 14, 16, 17, 24, 26, 51
PF Powe ac o . 5, 35, 38, 40
PFC Powe ac o con ol. 39
PFE Powe low equa ions. 1, 23, 28, 34
PoC poin o connec ion. 5, 38–41
R G Ne wo k Code on Requi emen s o G id Connec ion applicable o all Gene a o s.
33
SCR Sho ci cui a io. 39, 40
TSO T ansmission sys em ope a o . 33, 36
7
Design, modeling and simula ion o a PV powe plan 14
2.6.2 Load bus; PQ ype.
This classi ica ion e e s o he buses whe e he eal and eac i e powe a e speci ied. I
is ecommended o designa e any bus wi h injec ed complex powe as a load bus. The
ol age magni ude and angle adap o he si ua ions c ea ed by he powe injec ions.
I has o be men ioned ha he ol ages ha e limi a ions o ensu e smoo h and secu e
ol age anges o he de ices connec ed o he g id. Usually hese anges go om 90 o
110% o he nominal ol age.
2.6.3 Vol age con olled bus; PV ype.
As i s name desc ibes he ol age is con olled adding a ol age con olle module ha
se s he ol age in a na ow ange ha can be unde s ood as cons an . In addi ion he eal
powe is also speci ied. Consequen ly he eac i e powe is a a iable wi h a limi a ions
wi h uppe and lowe bounds. In a eal si ua ion a PV bus always ha e a gene a o ha
mee s he equi emen s o be a a iable sou ce o eac i e powe .
2.7 The New on-Raphson powe low solu ion.
2.7.1 P oblem gene al o m
In his sec ion he New on-Raphson algo i hm is desc ibed; in he i s pa he p esen-
a ion and gene al o m o he p oblem a e in oduced. In he second pa he ma icidal
powe low solu ion is discussed.
In p ac ice, he bus appa en ne powe Siis a pa ame e mo e accessible han he
bus cu en s Ii he e o e NRA eo ganizes he equa ions ollowing he nex p ocedu e.
Reg ouping e ms he gene ic admi ance exp ession is ob ained:
Yij =|Yij|∠θij =|Yij|cos θij +j|Yij |sin θij =Gij +jBij (2.12)
Using he Gauss powe low (2.8) S∗
ican be exp essed as:
S∗
i=V∗
iIi=V∗
i
n
X
j=1
(YijVj) (2.13)
Now eo ganizing equa ion (2.8) in co ela ion wi h he gene ic admi ance eo ganiza ion
(2.12) and he appa en powe conjuga e S∗
iequa ion (2.13) esul s he nex exp ession:
S∗
i=Pi−jQi=
n
X
j=1
|YijViVj|∠(θij +δj−δi) (2.14)
Finally 2nequa ions a e ob ained ollowing he gene ic o m o :
Pi=|Vi|2Gii +
n
X
j=1,
j6=i
|YijViVj|cos (θij +δj−δi) (2.15)
Design, modeling and simula ion o a PV powe plan 15
Qi=|Vi|2Bii +
n
X
j=1,
j6=i
|YijViVj|sin (θij +δj−δi) (2.16)
The key idea behind he New on-Raphson algo i hm is o use a sequen ial linea iza ion
whe e unc ion depending on xis equaled o ze o. S a ing om he ini ial guess e e y
s ep an inc emen o ∆xis de ined:
(ˆx) = 0 (2.17)
∆x( )III =x−x( )(2.18)
Rep esen ing (x) by a Taylo se ies whe e he componen s o 2nd o de o highe a e
neglec ed. This can be done and pe o ms well because he equa ion equals o ze o. The
exp essions emains now as:
(ˆx) = (x( )) + d (x( ))
dx ∆x( )(2.19)
Finally i can be sol ed o ∆x( )and s a a new i e a ion se ing he alue o x( +1) as
he new one.
∆x( )=−d (x( ))
dx −1
(x( )) (2.20)
This algo i hm has a quad a ic con e gence meaning he e o dec eases quickly when
app oaching he solu ion and he esul depends on he ini ial guesses ha can be se o
he ol ages. Two a e conside ed; x(0) = 0 o x(0) =−1. The i e a ion s op when he
absolu e alue o he unc ion wi h he cu en x(n)is lowe han a ole ance, se by he
use .
2.7.2 Mul i-Va iable NRA
The i s change o no ice is ha in he mul i- a iable case he a iables and unc ions ˆx,
(ˆx) and he e o e he inc emen ∆xbecome ec o s con aining he pa ame e s o each
bus. In addi ion he de i a i e componen is now a ma ix known as he Jacobian. This
ma ix con ains he pa ial de i a i e o each unc ion (ˆx) o e e y a iable xn.
J(x) =
∂ 1(x)
∂x1
∂ 1(x)
∂x2. . . ∂ 1(x)
∂xn
∂ 2(x)
∂x1
∂ 2(x)
∂x2. . . ∂ 2(x)
∂xn
.
.
..
.
.....
.
.
∂ n(x)
∂x1
∂ n(x)
∂x2. . . ∂ n(x)
∂xn
(2.21)
The sys em has a ma ix s uc u e he e o e he ope a ions mus be pe o med in acco -
dance. Following he solu ion scheme o p e ious sec ions he jacobian is in e ed o sol e
o he inc emen ∆x.
III : ep esen s he cu en s ep
Design, modeling and simula ion o a PV powe plan 16
The NRA applica ion o powe low need a e o mula ion o he powe equa ions associ-
a ed o each bus. Taking in o accoun he equa ions (2.12) and he conjuga e o (2.14)
and de ining he nex exp essions:
Vi=|Vi|ejθi=|Vi|∠θi
ejθ = cos θ+jsin θ
θij =θi−θj
The ollowing equa ion can be ob ained:
Si=
n
X
j=1
|Vi||Vj|ejθij (Gij −jBij) (2.22)
And esol ing in o he eal and imagina y pa s he equa ions used in he algo i hm can
be ound.
Pi=
n
X
j=1
|Vi||Vj|(Gij cos θij +Bij sin θij) = PGi −PDi (2.23)
Qi=
n
X
j=1
|Vi||Vj|(Gij sin θij −Bij cos θij) = QGi −QDi (2.24)
Wi h he i s bus designa ed as he slack bus and se ing i s ol age magni ude and angle
he es o he ol ages and phases a e ound ollowing he nex scheme:
x=
θ2
.
.
.
θn
|V2|
.
.
.
|Vn
(x) =
P2(x)−PG2+PD2
.
.
.
Pn(x)−PGn +PDn
Q2(x)−QG2+QD2
.
.
.
Qn(x)−QGn +QDn
A he beginning o he i e a ion he numbe o s eps is se o 0, = 0 and he ini ial
guesses a e in oduced. Each s ep he numbe is inc easing in one, = +1 The s opping
c i e ia is he same as seen be o e o a gi en ole ance, .
Design, modeling and simula ion o a PV powe plan 17
PV cases
As long as he PV buses a e ixed in ol age magni ude he e’s no need o include he
unknown in he sys em o w i e he eac i e powe balance equa ions; i a ies o main ain
he ixed ol age wi hin a easonable limi s. Howe e hose can be included w i ing he
ol age cons ain as |Vi|−Vise poin = 0. I hey a e no included in he sys em he sc ip
used in his wo k con ains a simple limi iola ion es . I checks i he limi s in oduced
in he bus da a a e iola ed. I he bounds a e o e passed he p og am displays a message
indica ing which bus is ha ing issues and he magni ude o hese.
2.8 Load lows o he sys em
Once he NRA con e ged he Vol age magni udes and angles a e known and he line lows
and losses a e ob ained.
Fi s o all he Line Cu en Flows a e calcula ed in each b anch using Ohm’s law in bo h
di ec ions o he b anch. The esul ing exp essions a e shown nex : The om bus- o bus
cu en exp ession is ob ained:
Iij =−(Vi−Vjab anch)Yij /a2
b anch +bb anch/a2
b anchVi(2.25)
and he e e se di ec ion exp ession esul s in:
Iji =−(Vj−Viab anch)Yji/a2
b anch +bb anchVj(2.26)
Whe e ais he ap se ing and is se o 1 o all simula ions. This ap se ing is a
unc ionali y ound in some ans o me s. I allows he ans o me o ha e a iable u n
a ios o keep he changes o he p ima y side. The g ound admi ance is ep esen ed by
b. In his wo k he g ound admi ance is he hal o he suscep ance p esen in any o
he wo buses con o ming he line. Secondly, he Line Powe Flows in MVA and he line
losses a e ob ained using he appa en powe equa ions:
Sij =ViI∗
ij ∗Sbase (2.27)
Lij =Sij +Sji (2.28)
Finally, he Bus Powe Injec ions esul om applying he equa ion (2.27) in addi ion o
he ini ial bus cu en appa en powe .
S∗
i=
Busn
X
j=1
V∗
iVjYij (2.29)
Whe e he eal pa is he ac i e powe Pi= eal(Si) and he imagina y is he eac i e
powe Qi=−imag(Si).
Chap e 3
Valida ing he esul s
3.1 Conside a ions.
In his chap e he p ocess o alida e he esul s is desc ibed and esul s o di e en cases
a e discussed.
The p ocess o alida ing consis in conside ing di e en cases and unning hem in a
consolida ed powe low algo i hm. In his wo k MATPOWER is se as he e e ence.
The di e ences be ween he inal esul s in bo h p og ams a e discussed wi h he pu pose
o explaining hem.
Once he eliabili y o he p esen wo k’s p og am is se di e en scena ios a e es ed
ocusing on desc ibing he beha iou o he sys em. The me hodology o he analysis is
based on a ying he alue o signi ican pa ame e s o he sys em in a ange ha include
ealis ic and common si ua ions and obse e he eac ion o he powe lows o he sys em.
The sc ip con ains di e en limi a ions ha a e commen ed in he nex sec ions. Some
o hem a e he eac i e powe limi a ion (Qmax, Qmin) when wo king wi h PV buses o
he capabili y cu e o he con e e s.
3.1.1 Limi a ions
In eal si ua ions he e a e a lo o physical limi a ions ha can be easily missconside ed.
The aim o his sec ion is o ci e all he limi a ions ha ha e been aken in o accoun
and i i is possible de ine a p ocedu e o deal wi h each one.
Capabili y cu e o he con e e
In he pape [2] he dependency o he in e e ’s capabili y cu e on ambien empe a u e,
sola i adiance, he dc ol age a ia ion and he in e e ope a ion. The con e e used
o pe o m he analysis ound in [2] is an in e e o med by wo s ages; he i s one is
18
Design, modeling and simula ion o a PV powe plan 19
a dc-dc s age used o s ep up he ol age o keep wo king on he maximum powe poin
(MPP). The second one make he p ope con e sion o an ac signal.
The i s wo pa ame e s, ambien empe a u e and sola i adiance, a e en i onmen al
due o i he alues can no be chosen. The es o he pa ame e s a e elec ical and
can be modi ied wi h obus p ocedu es. The dc ol age a ia ion is limi ed in he lowe
limi a ion (Vmin) by he minimum ol age o keep he ac side unde he nominal alues.
These nominal alues a e dic a ed by he g id code applied. The uppe limi a ion (Vmax)
in se by he open ci cui ol age o each panel imes he numbe o panels connec ed in
se ies. In he igu e (3.1) he ela ion be ween he Vdc and Pdc and he main poin s unde
s udy a e ep esen ed.
Figu e 3.1: Rela ion be ween he Pdc and Vdc and main poin s o analyze.
3.1.2 Dis ibu ion sys em limi a ions
To ensu e a co ec and smoo h pene a ion o he DES, especially sola ene gy based
gene a ion sys ems, many limi a ions a e imposed. Depending on he de ini ion o he
buses, PQ o PV he emaining pa ame e s ha e limi a ions. In i s case he PQ achie e
he gi en powe alues by egula ing he ol age angle and magni ude. The bounda ies
and hei s ic ness a e se depending on he local poin o he sys em and i s possible
implica ions on he de ices connec ed o i . In sec ion 4 he egula ion and he examples
o some g id codes can be ound. In second case and analogously he P V buses main ain
he ol age cons an by egula ing he eac i e powe . Thus, hose a e a ached o he
eac i e powe limi a ions ha a e explained in sec ion 4
3.1.3 Pa ame e s unde s udy.
The indica o s chosen o s udy and alida e his wo k’s esul s a e lis ed nex .
•Se en h bus angle and magni ude (V7 hbus): he capaci o is connec ed be ween he
g ound and he se en h bus he e o e i is included is he analysis.
Design, modeling and simula ion o a PV powe plan 20
•Ele en h bus angle and magni ude (V11 hbus): bo h o he su ace sc ip and linea
sc ip impose he same changes in each o he ou p a ays (buses 11,12,13,14).
Thus, he las bus included is he ele en h bus due o he balanced and symme ic
g id in he p esen example.
•Second bus angle and magni ude (V2ndbus): he second bus ol age is chosen due o
he p oximi y o he connec ion poin .
•To al ac i e and eac i e powe demand.
•To al ac i e and eac i e powe gene a ion.
•Real and imagina y losses.
3.2 Example case.
Figu e 3.2: Example case wi h designa ed buses.
The igu e 3.2 illus a es he sys em unde s udy and he posi ion o he buses. I emains
he same o all he buses du ing all he scena ios.
The nex able 3.1 con ains he alue o he di e en elemen s om he sys em
Design, modeling and simula ion o a PV powe plan 21
Table 3.1: P esen wo k sys em pa ame e s.
Elemen Value
Rc 8E-03 ohms
Xc (4,28E-05 H) 1,345E-2 ohms
Xcap1 (4,20E-04 F) 7,5788 ohms
Rcab 1,475E-02 ohms
Xcab (1,42E-04 H) 4,446E-2 ohms
R 1,60E-05 ohms
L (1,45E-04 H) 4,5553E-02 ohms
R hi 0,01 ohms
L hi (9,55E-05 H) 3E-2 ohms
Rg 5,33E-02 ohms
Lg (1,697653E-03 H) 5,33E-1 ohms
3.2.1 Pe uni alues
The exis ence o ans o me s which ha e di e en a io in he sys em make pe uni
alues a g ea ool o wo k wi h. In o de o change he alues o pe uni he base alues
shall be de ined. The base appa en powe emains he same in all he ans o me sides.
The base ol age is he one se by he ans o me side whe e he impedance is loca ed.
Finally he ela ion be ween his base pa ame e s and he cu en and impedance bases
is se by he Ohm’s law which is ep esen ed nex :
Sbase =VbaseIbase =V2
base
Zbase
(3.1)
Wi h he base alues he con e sionIo all he alues is pe o med esul ing in he nex
able:
Table 3.2: Example’s pa ame e s in pu.
Zbase1Zc(0,2 + 0,33615j) pu
Zcap (-5,2778E-03j) pu
Z i (1,6E-07 + j4,5553E-04) pu
Zbase2Zcab (1,475E-04 + j4,461E-04) pu
Z h i (1E-04 + j1,152E-04) pu
Zbase3Zg(5,333E-06 + j5,3333E-04) pu
3.2.2 Ini ial alues and supposi ions
As i is discussed in he Powe Flow analysis sec ion he sc ip need some o he pa ame e s
o s a a new case. In one hand in e ms o he bus da a he equi emen s a e lis ed in
IZbase1= 0,04 ohms, Zbase2= 100 ohms and Zbase3= 10 kohms.
Design, modeling and simula ion o a PV powe plan 22
able (3.3).
Table 3.3: Requi ed pa ame e s o each bus.
Bus Bus id
Type 1- Slack bus, 2- PV bus and 3- PQ bus.
Vsp Vol age speci ied a he beginning o bus i.
he a The V phase o bus i.
PGiReal powe gene a ed a each bus.
QGiReac i e powe gene a ed a bus i.
PLiAc i e powe as load in bus i.
QLiReac i e powe as load in bus i.
Qmin In e io limi s o eac i e powe .
Qmax Supe io limi o eac i e powe .
In he i s cases he ollowing supposi ions a e assumed:
•The su ace sc ip is a me hod o alida ing he esul s agains MATPOWER and
he e a e no inal conclusions ex ac ed.
•The i s bus is always he slack bus. In he cu en si ua ion he g id bus is se as
bus 1 and se as angle e e ence.
In he o he hand he alue o he lines is equi ed. The pa ame e s included a e shown
in able 3.4.
Table 3.4: Requi ed pa ame e s o each bus.
F om bus Bus id o he beginning o he line
To bus Bus id o he end o he line
R [pu] Resis ance o he line.
X [pu] Reluc ance o he line.
B/2 G ound admi ance
X’me TAP ap alue.
The supposi ions o he line pa ame e s a e:
•The Qmax and Qmin, in case o choosing P V buses, a e se o -99 MVA and 99
MVA espec i ely ha mean he e a e no limi a ions o hose a e conside ed wide
enough o he sys ems powe magni ude.
•The ap alueII used o adjus he eac i e powe in he bus e minals is always
equal o 1 in o de o simpli y he ini ial e i ying ope a ions and he exis ence o
powe elec onics.
II he ap alue is a capabili y some ans o me s ha e o change he u ns a io. I is used o manage
he eac i e powe . Nowadays he powe elec onics a e mo e e icien and much less cos -e ec i e.
Design, modeling and simula ion o a PV powe plan 23
•The g ound admi ance e e ed o he admi ance o he shun elemen s connec ed
o he bus. The alue e e s o he hal o he o al suscep ance (B) o he node due
o he alue is epea ed in he wo b anches ha include he pa icula bus.
3.3 Resul compa ison e sus MATPOWER
Wi h he aim o alida ing ha he p esen wo k’s PFE sc ip is obus enough o ex ac
conclusions o i , di e en cases a e es ed. Two MATLAB based p og ams a e included
in his wo k: he su ace sc ip and he linea sc ip .
The linea sc ip pe o ms mul iple simula ions in bo h p og ams; PFE and MATPOWER
changing one o he ini ial pa ame e s and plo s he esul s o he a iables o in e es
discussed in sec ion 3.1.3 in a 2D g aphic. I is used o e alua e he e ec s o changing
one o he inpu s o he sys em wi h he aim o inding ou i s beha iou .
The su ace sc ip pe o ms mul iple simula ions changing wo o he ini ial alues and
plo s su aces ep esen ing all he combina ions o a pa icula ange o he wo a iables.
I is used o alida e he esul s e sus MATPOWER. Bo h o hem a e limi ed due o
he impossibili y o applying di e en condi ions on each PV a ay. Howe e some cases
will be pe o med manually, applying di e en condi ions in some o he PV a ays (buses
11, 12, 13, 14). The su ace sc ip is use ul o iden i y in e es poin s such as minimum
o maximum alues depending on he wo a iables changing.
In he nex sec ions a summa y o he cases and he esul s o each p og am a e discussed
ollowing he same s uc u e:
1. Pa icula case ini ial alues and conside a ions.
2. Resul compa ison.
3.3.1 Fi s compa ison: Ac i e and eac i e powe demand a i-
a ion wi h PQ buses.
Ini ial alues and conside a ions
The eason o choosing he ollowing pa ame e s is discussed in he sec ion 3.1.3. This
case is con o med o es he capabili y o he PFE sol e o wo k wi h PQ buses.
The alues used o he simula ion a e shown in he nex able and a e e e enced o he
buses 11, 12, 13 and 14. In he eal sys em hese buses ep esen he PV a ays in o he
collec ion g id:
Resul s
Fi s ly he esul s o he ol age and angle o he buses o in e es a e displayed in igu es
3.3, 3.4, 3.5. The o iginal colo s a e black wi h 60% o opaci y and g een. As he esul s
Design, modeling and simula ion o a PV powe plan 30
3.4 Analysis o maximum and minimum ol age de i-
a ions and loses.
In his sec ion indi idual simula ions and a e pe o med wi h he aim o de ining he
beha iou o he sys em. Fo he nex simula ions all he con e e ou pu s a e se as
PV ype buses as he majo i y o de ices need a ange o ol ages o wo k co ec ly and
he magni ude o he changes in he sys em’s ol ages a e commonly wide han hose
anges. In sec ion 4.3 a new ype o bus is de ined whe e he eac i e powe is a unc ion
o he ol age wi h he aim o educing he losses and he e o e he s ess o he sys em
among o he bene i s.
3.4.1 Pa icula case: di e en alues o each con e e ou pu
In his case he ol age o he con e e s (buses 11, 12, 13 ,14) a e se o di e en alues
o simula e a eal si ua ion whe e he di e ences be ween componen s and condi ions se
di e en e iciencies in each PV a ay. Fo example; di e en dus densi ies in he panels
o shades ha educe he e iciency o he Maximum powe poin acke (MPPT)III.
The ini ial da a o he buses is summa ized in he able 3.8 Using he sc ip Indi idual
Table 3.8: Ini ial alues o he con e e ou pu buses.
Bus id Vol age [p.u.] P demand [MW] P gene a ion [MW]
11 1.02 0.65 0.43
12 1.01 0.59 0.44
13 1.03 0.62 0.41
14 1.09 0.66 0.40
case he esul s a e ob ained and ep esen ed in igu e 3.12. I can be obse ed ha he
maximum ol age de ia ion in pe uni , excep ing he ol age egula ed buses, is ound
in bus 5 wi h an inc emen o 0,0161% which ans o med o eal alues is an inc emen
o 322 V. The minimum ol age de ia ion is ound in bus 2 wi h a pe uni inc emen o
0,0094% and eal alue o 188 V.
The losses and injec ed appa en powe s in each bus a e illus a ed in igu e 3.13. The
eal and imagina y losses a e simila o he magni udes o hesu ace sc ip o he same
g id ol age alue and ac i e powe gene a ion igu e 6.10. In ac i he linea sc ip is
used o e alua e he ela ion be ween he g id ol age and bo h imagina y and eal losses
using he same ini ial da a. The esul s om igu e 3.14 co espond o a a ia ion in g id
ol age om 0.9 o 1.1 p.u. I can be concluded ha bo h losses p esen a minimum when
he g id ol age is 1.
IIIMPPT: is a de ice ha se he ol age in he a ay o mee he maximum ac i e powe ou pu . I
has o wo k wi h he lowe ol age on he a ay he e o e i one panel is pe o ming poo ly he es o
he a ay is a ec ed. I is conside ed as a ype o loss.
Design, modeling and simula ion o a PV powe plan 31
Figu e 3.12: Example scheme wi h ol age magni ude and phase alues
Figu e 3.13: Example scheme wi h ol age magni ude and phase alues
Design, modeling and simula ion o a PV powe plan 32
Figu e 3.14: Real and Imagina y losses changing he g id ol age om 0.9 o 1.1
Chap e 4
G id code equi emen s
4.1 In oduc ion
A G id Code is a echnical documen con aining he ules go e ning he ope a ion, main e-
nance and de elopmen o he ansmission sys em. I is necessa y due o he exis ence o
di e en ene gy supplie s ha wo k wi h he same T ansmission sys em ope a o (TSO).
In he g id code all he equi emen s a e speci ied and discussed o ensu e s abili y and
quali y o elec ical sys ems in con inuous ope a ion as well as du ing aul ’s o ex ao -
dina y e en s. G id codes a e di e en and depend on he coun y hey a e con o med.
Each coun y elabo a es one and he di e ences be ween hem a e mo i a ed by he cha -
ac e is ics o he coun y and he pa icula needs c ea ed by he local elec ical sys em.
Howe e acco ding o [7] he e is a endency o ha monize g id codes in di e en EU
coun ies wi h documen s such as: Ne wo k Code on Requi emen s o G id Connec ion
applicable o all Gene a o s (R G) d a ed by Eu opean Ne wo k o T ansmission Sys-
em Ope a o s o Elec ici y (ENSTO-E). These documen s a e used as amewo k while
de eloping g id codes.
The inc easing pene a ion o enewable ene gies du ing he las decades ha e o ced im-
p o emen s in he echnology and sys em’s beha iou con ol. These changes a e mainly
mo i a ed by he in insic cha ac e is ics o enewable ene gies, such as he dependency
on he en i onmen and he in e mi en beha iou o he na u al ene gy esou ces. These
a ia ions can conclude in undesi ed aul s c ea ing damages o he ansmission sys em
elemen s and all he equipmen connec ed o he ne wo k. The g id codes a e cons an ly
upda ed o ensu e smoo h and co ec unc ioning in all si ua ions.
Fo example he g id code o a la ge coun y wi h a good numbe o sou ces and a
low pene a ion o enewable ene gies mus ha e less s ic condi ions o ansmission
sys em equency han a li le island which has an equal p opo ion o enewable and
non- enewable ene gy sou ces.
The e a e wo ypes o equi emen s: s a ic and dynamic. S a ic equi emen s e e o
he con inuous ope a ion s a e and he dynamic collec he p ocedu es o ollow du ing
aul sequences and dis u bances. The g id codes include wo s case scena ios o co e
33
Design, modeling and simula ion o a PV powe plan 34
all he possible e en s ha can ake place including s ic measu es as he comple e
disconnec ion.
The enewable ene gies p esen a big lexibili y o pene a e in he ma ke as hey can be
in oduced as wind a ms, PV plan s, Hyd o-powe ed s a ions... The sola ene gy is one
o he esou ces ha a e expe iencing a as g ow h du ing he las decades. In he pape
[6] p esen ed in MPDI jou nal i is exposed ha he powe ins alled capaci y in Spain
o sola ene gy inc eased o e a 5 GW om 2007 o 2018 wi h a posi i e endency. The
e olu ion is ep esen ed in igu e 4.1.
Figu e 4.1: Powe ins alled in Spain om 2007 o 2018.
In he p esen wo k he g id code equi emen s o a La ge Scale PV powe plan (LS-
PVPP) a e discussed and con ex ualized o di e en g id codes. The pa icula solu ion
p esen ed in [8] o compensa ing eac i e powe is in oduced in he p esen wo k PFE
sol e .
Acco ding o [11] he amoun o ene gy managed ia he peninsula sys em’s ancilla y
se ices in GWh is illus a ed in igu e 4.2.
Figu e 4.2: Summa y o he o al ene gy managed by he ancilla y se ices in he spanish
peninsula.
Design, modeling and simula ion o a PV powe plan 35
4.2 S a ic egula ion
In gene al he s a ic egula ion o enewable ene gies depend on he esou ce used and
i s in insic cha ac e is ics. Fo example o sola and wind sou ces powe cu ailmen is
needed due o he unp edic able a ailabili y and he as and wide a ia ions hey can
su e due o en i onmen al condi ions. Powe cu ailmen limi s he powe ex ac ion in
wo di e en si ua ions: when he powe ex ac ed is g ea e han he nominal alue o
he plan in peak gene a ion e en s o when he demand is lowe han he p oduc ion.
In [4] an analysis o he connec ion equi emen s o wind powe gene a ion uni s o six
g id codes is ca ied ou .
4.2.1 Powe ac o egula ion
The powe ac o is he a io o eac i e o appa en powe consump ion, hus i s equi e-
men s and he eac i e powe a e simila in he di e en G id Codes. Any induc ance
wi h no capaci o s a ached consumes eac i e powe . Consequen ly his eac i e powe
consump ion mus be p oduced somewhe e in he g id. One hing o conside is ha he
dis ibu ion o eac i e powe is cos in ensi e and he eac i e powe is no p o i able
al hough i is used in some cases o suppo he g id du ing pa icula e en s. In [5] an
example o di e en G id Codes powe ac o / eac i e powe equi emen s is p esen ed:
Figu e 4.3: Reac i e powe equi emen s o Pue o Rico, Sou h A ica and China, Ge -
many and Romania.
The mos es ic i e PF equi emen in igu e (4.3) is he ed line e e ing he G id code
om Sou h A ica. F om 100% o 20% o p oduc ion he maximum PF a e 0.975 leading
and 0.975 lagging.
One o he limi a ions commen ed in sec ion 3.1.1 a e he eac i e powe bounds o PV
buses. Those limi a ions should be es ablished in o de o mee he G id Code guidelines
Design, modeling and simula ion o a PV powe plan 36
o eac i e powe consump ion.
4.2.2 Powe Cu ailmen
Powe cu ailmen in oduces he possible educ ion o gene a ed ac i e powe depending
on g id equi emen s. Two common si ua ions whe e powe cu ailmen is e ec i e a e:
•The possible o e loading a peak gene a ion hou s o na u al esou ces as wind o
sola ene gy. Due o he inhe en cha ac e is ics o hose.
•When he demand is lowe han he gene a ed ac i e powe wi hou powe cu ail-
men .
The o e loading is egula ed by he TSO because i would cause an inc emen o he
equency, he e o e an inc emen in he ine ia o he ansmission sys em.
The powe cu ailmen is no ou lined p ecisely in mos o he G id Codes and i de-
pends on en i onmen al condi ions, ins alled capaci y and economics. Acco ding o [4]
an indica o o apply his equi emen is he equency su plus. In Eu opean coun ies,
wi h a g id equency o 50 Hz, he bound o s a applying powe cu ailmen was 50,5
Hz. When he equency de ia ion dec eases he ees ablishmen o he ac i e powe has
i s own limi a ions. I should be less han he 10% o he ne wo k capaci y pe minu e.
Figu e 4.4 ex ac ed om [4] illus a es he ope a ion condi ions when he ne wo k is
o e loading.
Figu e 4.4: Powe cu ailmen equi emen s
4.2.3 Ac i e powe ese es
The endency o in oducing mo e and mo e Dis ibu ed ene gy sis ems (DES) based
on enewable ene gies in exis ing dis ibu ion ne wo ks gene a es new p oblems. The
di icul y o p edic ion and in e mi ency o enewable ene gy sou ces in addi ion o he
high densi y o DES lead o ol age and equency issues in he dis ibu ion ne wo k.
Fo example he p oximi y o PV plan s due o he land es ic ions submi hem all o
he same en i onmen al condi ions. I has o be no iced ha he mos common peaks o
sola p oduc ion a e ound in he middle o he day whe e he demand is low. The ene gy
su plus is edi ec ed in o he g id howe e in a dis ibu ion ne wo k wi h a high densi y
Design, modeling and simula ion o a PV powe plan 37
o DES i can conclude in bidi ec ional powe lows, o e loads o o e ol ages on some
b anches o ans o me s.
To ix his issues and ensu e a smoo h pene a ion o he DES in he dis ibu ion ne wo ks
wo solu ions a e ound. The i s one is he use o ba e ies as ene gy s o age sys ems
o supply he ac i e powe demand in comp omised si ua ions. Al hough i has been
expe imen ally implemen ed p esen ing good esul s i is s ill cos -p ohibi i e and is no
cos -compe i i e i compa ed wi h o he sou ces. The main p oblem is he cu en ene gy
s o age echnology ha p esen s a low e iciency and a high main enance and p oduc ion
cos s a his powe a ings. The al e na i e me hod is he ac i e powe egula ion. I is
based on he powe cu ailmen . In a si ua ion whe e he ene gy p oduc ion can each
he nominal alues he ac i e powe ou pu is egula ed o o e a ound an pe cen age
o he maximum capaci y. The emaining capaci y is se as ac i e powe ese es. This
app oach is easie o implemen bu i causes a educ ion in he e iciency o he plan .
4.2.4 Vol age ange and Con ol
The PV plan mus be able o un a a ed ol age plus he speci ied ol age ange. The
a ed ol age and he ol age ange a ies depending on he ansmission sys em, which
depends on he coun y. The ol age anges a he di e en a ed alues, o limi ed
pe iods o ime du ing aul o pa icula e en s, a e also speci ied. Acco ding o [4] he
equi emen o all nominal alues in con inuous ope a ion a ies a ound ±10% o he
nominal alue.
4.2.5 Remo e Vol age Con ol
Addi ionally in mos o he g id codes i is equi ed ha he LS-PVPP o powe s a ion
con ains a closed loop ol age egula ion sys em.
The ol age egula ion ollow he basis o he PV bus de ini ion whe e a ol age se poin
is speci ied and egula ed by con inuously modula ing he eac i e powe ou pu in he
accep able anges. In addi ion he ol age se poin shall be wi hin he limi s.
The ime esponse o he ol age egula ion sys em is also speci ied in some o he g id
codes analyzed in [4].
4.2.6 F equency
The gene a o ’s plan mus be capable o wo king in acco dance wi h speci ied equency
anges. The majo i y o he g id codes p esen ed [4] mus be able o un con inuous
ope a ion in equencies be ween 47,5 Hz and 52 Hz. Time delays migh be necessa y in
some ange o equencies. The p ocedu e in case o b eaking he limi s is o educe he
ac i e powe ou pu and emain connec ed. Disconnec ion is only pe mi ed a equencies
below 47 Hz and abo e 53 Hz and wi h no ime delay.
Design, modeling and simula ion o a PV powe plan 38
4.2.7 Flicke
A licke is a single, apid change o he oo mean squa e ol age. The connec ion o
di e en elemen s such as capaci o s, lines, cables, ans o me s and o he elemen s may
cause ansmission sys em s ep changes.
I mus be men ioned ha a lo o g id codes do no men ion licke ’s, acco ding o [4]
he Danish g id code speci ies he nex p ocedu e.
Di e en h esholds a e se depending on he ime in e al whe e he licke magni ude
is e alua ed. The sho e m licke is a weigh ed a e age o he licke con ibu ion o e
en minu es. The long e m licke is de ined o e 2 hou .
Finally in igu e 4.5, ex ac ed om [5], a summa y o he equi emen s o a LS-PVPP
is illus a ed o a be e unde s anding.
Figu e 4.5: Summa y o g id codes equi emen s o LS-PVPP
4.3 G id suppo ; eac i e powe compensa ion
As he p esen wo k is cen e ed in enewable ene gies and hey a e based on na u al
esou ces usually he gene a ion plan s a e loca ed in emo e a eas. Consequen ly he
ansmission leng hs a e long enough o conside he ansmission line impedance signi -
ican . As he ansmission lines a e mos ly composed o induc ance elemen s, which a e
pa o he imagina y domain, and ans o me s.
The in e e is synch onized o a poin o connec ion (PoC) and he eedback measu e-
men s a e aken a his poin . As he cu en ed o he PoC lows h ough he g id
impedance a quan i y o eac i e powe is c ea ed al hough he in e e is wo king a
uni y PF. Thus, he ac i e powe o he PV in e e becomes coupled wi h he eac-
i e powe seen by he g id. This undesi ed eac i e powe causes a bad beha iou ha
includes an inc ease in ansmission losses and consequen ly educes he maximum ans-
mission capaci y, comp omises sys em s abili y and s ains he g id.
The undesi ed eac i e powe depends on he g id cu en and g id impedance. I he
impedance is high he PoC cu en has a signi ican impac on he local ol ages. A g id
Design, modeling and simula ion o a PV powe plan 39
Figu e 4.6: Scheme o he connec ion poin s con ex ualized in he cu en example
wi h a high impedance is de ined as a weak g id. A pa ame e o measu e he deg ee o
weakness is he Sho ci cui a io (SCR) which is de ined by he nex exp ession:
SCR =V2
n
SnZg
(4.1)
Whe e Vnis he nominal ol age, Snis he nominal appa en powe (usually in dis i-
bu ion gene a ion sys ems is he alue o he in e e due o he ans o me s ha e he
same powe a io) and Zgis he ansmission sys em impedance. Once he alue o he
SCR is less han 5 i can be conside ed weak. As i is commen ed in p e ious sec ions
he enewable ene gy gene a ing plan s a e loca ed a emo e loca ions he impedance
become signi ican , hus he g id is conside ed weak.
Acco ding o [8] he eac i e powe has been managed using synch onous gene a o s,
capaci o banks as he p esen wo k case examples o wi h FACTSIde ices. The e-
ac i e powe o an in e e is ela i ely easy o con ol wi hou he need o addi ional
in es men s.
The eac i e powe con ol has wo goals: ol age egula ion and eac i e powe op imiza-
ion o a comp omise be ween hem. In ac hey can no be op imized simul aneously as
hey a e ela ed. In [8] and in he p esen wo k he ol age egula ion is no conside ed
as he buses a e se as PV buses and a no i ica ion pops up i he Q limi s a e iola ed
wi h he cu en ol age se poin .
Acco ding o he au ho s o [8] he mos common app oach o con ol eac i e powe in
LS-PVPP is he Powe ac o con ol (PFC) whe e he eac i e powe is p opo ional o
he ac i e powe eed. Con en ional PFC assumes he ac i e powe - eac i e powe a io
(Q/P) cons an al hough he eac i e powe injec ion om in e e a ec s he PoC ol age
changing he commen ed Q/P a io and he e o e he PFC achie es op imal decoupling
only when he in e e is ope a ing a nominal alues. In enewable gene a ion sys ems he
ac i e powe ou pu changes d as ically along wi h he en i onmen al condi ions he e o e
i adi ional PFC is applied i leads o subop imal eac i e powe compensa ion.
IFACTS: lexible al e na ing cu en ansmission sys em: sys em o s a ic componen s used o he
AC ansmission used o enhance con ol and inc ease powe ans e capabili y o he ne wo k. I is
usually based on powe elec onics.
Design, modeling and simula ion o a PV powe plan 46
deep s udy on he loca ion, a mi iga ion plan and a co ec law code ha p o ec s he
mo e sensi i e a eas can conclude in many en i onmen al and social bene i s.
Chap e 6
P ojec s budge
The p esen wo k is cen e ed in he op imiza ion o a PV plan by including a ool
ha allows he use o ake ad an age o eal ime in o ma ion o educe losses and
choose he bes se ings a each si ua ion. The cos o implemen a ion is ela i ely low
in compa ison o he o al budge o c ea ing an en i e PV powe plan and i can ha e
posi i e economic e ec s as consequences. Di e en concep ion cos s pe ene gy uni a e
p esen ed. In i s place he CAPEXI(capi al expendi u e, able 6.1) da a o a LS-
PVPP p ojec a e p esen ed. In second place OPEXII (Ope a ional expendi u es, able
6.2) de i ed om he con inuous ope a ion o he PV plan a e abula ed. Finally a small
e iew o he economic impac o he ancilla y se ices is p esen ed acco ding o Red
El´ec ica de Espa˜na.
6.1 La ge Scale PV powe plan
The p ices shown on able 6.1 a e ex ac ed om [2]. The da a is ex ac ed om
Bloombe g New Ene gy Finance which is a eliable ene gy ela ed da abase. The ba e y
sys em commen ed in sec ion 4 is included o illus a e he high cos pe kWh o his
me hodology. The miscellaneous cos include an a e age land p ince and he enginee ing,
he de i ed cos o legal pe missions and he cons uc ion among o he s.
Whe e EUR/Wnis he cos pe wa o nominal powe . The Balance o he sys em (BOS)
encompasses all he componen s o a pho o ol aic sys em o he han pho o ol aic panels
and he in e e s. This concep includes wi ing, swi ches, moun ing sys em and some
mechanism used o inc ease he e iciency o he PV powe plan such as he maximum
powe poin acke (MPPT) o he GPS sola acke ha changes he o ien a ion o he
panels depending on he posi ion o he Sun.
In he able 6.2 he con inuous ope a ions cos o a yea o p oduc ion is p esen ed. The
PV echnologies need ene gy o s a con e ing ene gy and he in e mi ency o he sola
ICAPEX; he in e sion cos o s a unning he PV plan .
IIOPEX; he ope a ion and main enance cos o unning he PV plan .
47
Design, modeling and simula ion o a PV powe plan 48
Table 6.1: PV plan CAPEX cos s and cu en p ojec budge
CAPEX componen Cos pe Wa Cu en p ojec
Modules 0,31 EUR/Wn1,24 M. EUR
Cen al in e e s 0,03 EUR/WAC 0,12 M. EUR
Balance o he sys em 0,17 EUR/WAC 0,68 M. EUR
Miscellaneous cos 0,30 EUR/WAC 1,2 M. EUR
Powe cu ailmen wi h ba e ies (op ional) 400 EUR/kWh -
sou ce o he nigh pe iods c ea e si ua ions whe e he consumed ene gy is highe han
he p oduc ion. The dus emo al, panel cleaning, main enance p oduc s and sala ies a e
included. The acke sys em ope a ion cos is isola ed o illus a e he associa ed cos
o implemen ing a sola acking sys em.
Table 6.2: PV plan OPEX
OPEX componen Cos pe MW
OPEX PV plan ope a ion 9500 EUR/MW
OPEX acke ope a ion (i included) 900 EUR/MW
6.2 Ancilla y se ices
The g id codes equi emen s ha e an impac on he a e age inal p ice o ene gy. In he
epo o Red El´ec ica de Espa˜na o 2018 abou he ancilla y se ices [11] a b eakdown
o he componen s con o ming ha p ice is pe o med. The igu e 6.1 illus a es i . The
g id suppo equi emen s seen in sec ion 4 in oduce a educ ion on he inal p ice. The
magni ude o i s educ ion is small in compa ison wi h he inal p ice bu he ad an ages
in ene gy managemen in e ms o eliabili y and secu i y make hei applica ion essen ial.
Design, modeling and simula ion o a PV powe plan 49
Figu e 6.1: B eakdown o ancilla y se ices cos in he a e age inal p ice o ene gy in he
peninsula sys em
Conclusions and u he wo k
Du ing he las yea s sola ene gy has pene a ed in he ene gy ma ke as an al e na i e
o ossil uel-based ene gy. I has been possible due o he ad ances in powe elec onics
mixed wi h he au oma ic con ol me hodologies ha pe mi ed he mi iga ion o he
unin ended beha iou o enewable ene gies. The balance o he en i onmen al e ec s
b ough by hese enewable ene gies is mos ly posi i e howe e in sec ion 5 is shown
ha he e ec s on he en i onmen a e di e se and s ongly dependen on he si ua ion
and he ecosys em sensibili y. In some cases hose consequences a e unknown and can
conclude in a pe manen change in he su oundings o he ins alla ion wi h impo an
losses in biodi e si y and na u al weal h. In consequence a deep s udy on he biodi e si y
and p esen ecosys em shall be pe o med in o de o mi iga e hose nega i e e ec s. The
soil and physic en i onmen a e also a ec ed as a pa o he ene gy hey we e usually
ecei ing o de elop na u al p ocesses is used o p oduce ene gy o human consump ion
causing changes in he local empe a u e and soil nu ien s de elopmen .
In he cu en wo k a p og am based on [3] o sol e he powe lows o a gi en sys em
is de eloped and alida ed e sus MATPOWER. I models he buses using adi ional
me hods as PQ and PV buses. A summa y o he mos common s a ic egula ion p esen
in mos o he g id codes is included. A pa icula solu ion o mi iga e he nega i e e ec s
o he unin ended eac i e powe on la ge dis ibu ion sys ems based on egula ing he
amoun o eac i e powe as a unc ion o he ol age is also included. In u he wo k he
implemen a ion o ha bus model and mo e g id code equi emen s can be pe o med.
Al hough he good esul s achie ed by ene gy s o age sys ems and he new de elopmen s
he implemen a ion cos is s ill high in o de o be compe i i e. Thus, he ac i e powe
ese es me hodology discussed in sec ion 4.2.3 leads o a e iciency dec ease.
50
Annex
6.3 In oduc ion
In his chap e di e en images and sc ip in o ma ion om his wo k can be ound.
6.4 Sc ip s
A b ie desc ip ion o he sc ip s used in he p esen p ojec is included nex . The sc ip s
can be ound in he .zip ile annexed o his p ojec .
•The sc ip s named busda as and lineda as a e ex ac ed om [3] and a e used o
in oduce each case ini ial da a. The in o ma ion abou he da a in oduc ion is
discussed in sec ion 3.2.2.
•yma ix is used o c ea e he admi ance ma ix.
•p eq and p eq2 a e he co e o he NRA. Thei esul s a e he ol age and phase
alues o he buses. The di e ence be ween hem is he way o displaying he esul s.
•load lows1 and load lows2 a e used o pe o m he load low analysis. They a e
called om p eq and p eq2 espec i ely.
•The Su ace sc ip is used o alida e he esul s agains MATPOWER and iden i y
pa icula poin s i needed. I a ies wo pa ame e s o he sys em and plo he
esul s in di e en su aces o bo h p og ams. The di e en pa ame e s ha can
be changed a e: ac i e and eac i e powe demand (Pl,Ql), ac i e and eac i e
powe gene a ion (Pg,Qg), all con e e s ol age a he same a e (only PV buses;
V11,V12,V13 and V14) and he g id ol age (V1).
•The linea sc ip is made o a wo dimension compa ison o s udy he beha iou
in a mo e isola ed manne han in he Su ace sc ip .
•The linea p e sc ip pe o m simula ions o changing one pa ame e and plo s he
esul s in wo dimension g aphics. The a ailable pa ame e s o change a e he same
om su ace sc ip .
51
Design, modeling and simula ion o a PV powe plan 52
•The indi idual case sc ip pe o m one simula ion o he selec ed case and displays
an accu a e summa y o he alues o he sys em. A b eakdown o he b anches
and buses alues as well as he o als a e included. In addi ion in cases including
PV buses he limi iola ions a e also displayed.
6.5 Fi s compa ison powe low esul s
(a) Ac i e powe demand in KW (b) Reac i e powe demand in KVA
Figu e 6.2: Ac i e and eac i e powe demand (1s case)
(a) Ac i e powe gene a ion in KW (b) Reac i e powe gene a ion in KVA
Figu e 6.3: Ac i e and eac i e powe gene a ion (1s case)
Design, modeling and simula ion o a PV powe plan 53
(a) Real losses in KW (b) Imagina y losses in kVA
Figu e 6.4: Imagina y and eal losses (1s case)
Design, modeling and simula ion o a PV powe plan 54
6.6 Second compa ison powe low esul s
(a) Ac i e powe demand in KW (b) Reaci e p`owe demand in KVA
Figu e 6.5: Ac i e and eac i e powe demand (2nd case)
(a) Ac i e powe gene a ion in KW (b) Reaci e p`owe gene a ion in KVA
Figu e 6.6: Ac i e and eac i e powe gene a ion (2nd case)
(a) Real losses (b) Imagina y losses in kVA
Figu e 6.7: Ac i e and eac i e powe gene a ion (2nd case)
Design, modeling and simula ion o a PV powe plan 55
6.7 Thi d compa ison powe low esul s
(a) Ac i e powe demand in KW (b) Reac i e powe demand in KVA
Figu e 6.8: Ac i e and eac i e powe demand (3 d case)
(a) Ac i e powe gene a ion in KW (b) Reac i e powe gene a ion in KVA
Figu e 6.9: Ac i e and eac i e powe gene a ion (3 d case)
(a) Real losses (b) Imagina y losses
Figu e 6.10: Ac i e and eac i e powe gene a ion (3 d case)