Journal of Mechanics Engineering and Automation 5 (2015) 616-622 doi: 10.17265/2159-5275/2015.11.004

Analysis of Influence and Contribution on Distribution

System under Widespread PVs and EVs

Shoji Kawasaki and Shunsuke Fukami

Department of Electronics and Bioinformatics, School of Science and Technology, Meiji University, Kanagawa 214-8571, Japan

Abstract: In recent years, against a background of an environmental problem and resource problem, the introduction of RES (renewable energy source) such as wind power generation and PV (photovoltaic generation), EV (electric vehicle), and PHEV (Plug-in hybrid electric vehicle) has been expanding. However, various problems have an ongoing discussion. When the production of electricity by RESs exceeds the power consumption, it is possible to cause a steep variation of point voltage and a deviation from a proper voltage range in a distribution system to which RESs are interconnected. When EVs and PHEVs have spread to the distribution system, a new peak power-demand and a steep voltage drop might occur in the midnight charging time zone in case the electricity charges are low. In this paper, the authors analyze the effects on the distribution system under widespread PVs, EVs, and PHEVs. In addition, the authors propose an improvement plan and analyze about the influence and contribution. Key words: Photovoltaic generation, electric vehicle, reverse power flow, battery charge, plug-in hybrid electric vehicle.

1. Introduction

In recent years, the number of RES such as PV and

wind power generation system that interconnects to a

distribution system has been increasing to reduce the

influence on the environment in Japan. And the

number of an automobile powered by electricity such

as EV and PHEV has been increasing too. The output

of PV is influenced by change of weather conditions

and causes a rapid fluctuation. When the production of

electricity by RESs exceeds the power consumption, it

is possible to cause a reverse power flow, steep

change of point voltage, deviation from a proper

voltage range, etc. In addition, when EVs and PHEVs

have spread to the distribution system, a new

power-demand and steep voltage drop might occur in

the midnight charging time zone in case the electricity

charges are low in Japan.

In this study, firstly, the authors build a detailed

analysis model of general distribution system in Japan,

and analyze the effects on the distribution system

Corresponding author: Shoji Kawasaki, senior assistant

professor, research fields: optimization control of distribution network and electric power quality improvement of distribution network.

introduced a demand response which changes the

electric power usage according to the situation if the

electric power supply and demand under widespread

of PVs, EVs, and PHEVs. Next, the authors propose a

new charging schedule for EVs and PHEVs, and

analyze the influence and contribution on the

distribution system introduced the proposed schedule

under a variety of conditions. Finally, the authors

summarize the results of analyses mentioned above as

an index, and indicate it as a part of electric power

system planning.

2. Analyses Condition

2.1 Distribution System Model

Fig. 1 shows the used analytical model. The model

was assumed the general three-phase (phase U, phase

V and phase W) unbalanced distribution system of

residential area in Japan. The electric power is

supplied to loads through a distribution substation

from a higher system like Fig. 1. The line length

between the nodes is 0.5 km (total line length is 3.5

km), and each line is assumed the aluminum wire

(AL-OE120). Also, this feeder is composed of 9 loads,

D DAVID PUBLISHING

An

Fig. 1 Distri

9 pole transf

assume 100

to the secon

transformers

2.2 Power-D

Fig. 2 sh

residential a

patterns pow

load period

other is a c

demand dec

heavy load p

and 1.6 MW

connected

imbalance fa

light load pe

2.3 Model of

Fig. 3 sho

clear day. T

a.m., and th

power factor

And the in

system are a

of PVs.

2.4 Distribu

When the

system is co

that it affec

greatly. The

the distribut

and area C)

alysis of Influ

ibution system

formers, and

V or 200 V r

dary distribu

s.

Demand Curv

hows the us

area [1]. In th

wer-demand c

in case electr

urve of light

reases. The to

period and in

W, respectiv

to the feed

actor of feede

eriod is 1.31%

f PVs

ows used PV

The PV outpu

he peak outpu

r for the inter

fluence and

analyzed by

tion Ratio of

e number of P

oncentrating

cts the power

erefore, in th

ion system in

as shown in

uence and Co

m model of resid

9 PV system

residential lo

tion system t

ve

sed power-de

his study, the

curve. One is

ric demand in

t load period

otal load cap

n light load p

vely. The nu

der is 2,30

er in heavy lo

% and 1.21%,

V output of o

ut approache

ut is 3 kW. I

rconnection o

contribution

changing the

f PVs

PVs spread to

certain area,

r flow of dis

his study, the

nto three area

Fig. 4, and v

ontribution o

dential area.

s. The used lo

oad are conne

thorough the

emand curve

authors use t

a curve of he

ncreases, and

d in case ele

acity of feede

period is 2.8 M

umber of ho

0. The vol

oad period an

, respectively

one house of

es the peak a

In this study,

of PV system

n on distribu

e penetration

o the distribu

, it is consid

stribution sys

e authors div

as (area A, are

erified the po

n Distribution

oads

ected

pole

e of

the 2

eavy

d the

ctric

er in

MW

uses

ltage

nd in

y.

f the

at 12

, the

m is 1.

ution

rate

ution

dered

stem

vided

ea B,

ower

Fig.

Fig.

Fig.

Tab

flow

as s

ana

2.5

T

EV

n System und

. 2 Power-dem

. 3 PV output

. 4 Divided ar

ble 1 Distribu

w under the f

shown in Tab

alyzed by the

Simulation C

Table 2 show

s and PHEVs

Case 1

Case 2Case 3Case 4

der Widespre

mand of distrib

t per house.

reas in distribu

ution ratio of P

four cases of

ble 1. In this

phase differe

Condition of E

s the details o

s [2]. The intr

Area A

1

11

2::

::

ead PVs and E

bution system.

ution system.

PVs.

f distribution

study, the p

ences between

EVs and PHE

of simulation

roduction rat

Area B11

12

::::

EVs 617

.

ratio of PVs

power flow is

n the loads.

EVs

n condition of

e of EVs and

Area C

1112

7

s

s

f

d

An

618

Table 2 Sim

Table 3 Num

PHEVs is 3

Japanese go

EVs and PH

of houses, E

In this stu

by a norma

home can s

electric vehi

where, Hcar[

home, Hhome

is rate of

transportatio

2.6 Charging

In this stu

the followin

are as follow

Charging

charging sch

Charging

Charging

all EVs and

electricity ch

started from

Fig. 5 sho

rate of car u

Charging meth

Battery Capac

1

2

3

4

5

6

7

8

9

Total

LoadNumber

alysis of Influ

mulation condit

mber of EVs a

30%, which

overnment by

EVs is 3 to 1

EVs and PHEV

udy, all EVs

al charger, so

start chargin

icles stay hom

1 1

[%] is hourly

e is hourly rat

car users o

on.

g Schedule fo

udy, EVs and

ng two kinds

ws:

schedule

hedule”;

schedule 2: “

schedule 1 b

d PHEVs stay

harges are lo

m this time.

ows the rate o

users at 11 p.m

od

ityPHEV

EV

EV

PHEV

The number of hou

207

299

230

253

230

276

253

299

2300

253

uence and Co

tion of EV and

nd PHEVs at e

is the spread

y 2030 [3]. A

[4]. Table 3 s

Vs which eac

and PHEVs

o only electr

ng. The auth

me or not Eq.

y rate of elect

te of people s

occupied to

or EVs and PH

PHEVs char

of charging

1: “Night

“Split chargin

based on the

y home at 11

ow in Japan, a

of people wh

m. [5, 6]. As

24kWh

4.4kWh

3kW (const

2kW (const

V

V

uses The number of

47

68

52

57

52

62

57

57

68

520

ontribution o

d PHEV.

each load.

d target valu

And the ratio

shows the num

ch load has.

s charge batte

ric vehicles

hors set whe

(1).

100

tric vehicles

stay home, an

main means

HEVs

rge the battery

schedules, w

simultaneo

ng schedule”.

e assumption

p.m. in case

and the charg

ho stay home

shown in Fig

tant power charge)

tant power charge)

EVs The number of

20

19

19

15

22

17

19

17

22

170

n Distribution

ue of

o of

mber

eries

stay

ether

(1)

stay

nd A

s of

y by

which

ously

.

that

e the

ge is

and

g. 5,

Hho

Hca

perc

p.m

In

at 1

the

PHE

othe

F

of c

Hho

Hca

perc

a.m

sam

cha

2.7

E

and

and

EV

dist

and

cha

Fig.p.m

Fig.

PHEVs

n System und

ome at 11 p.m.

r which calc

centage of E

m.

n charging sc

12 a.m. when

charge is st

EVs charge

ers start charg

Fig. 6 shows

car users at

ome at 12 a.m.

r which calc

centage of E

m. and the oth

me schedule

arge.

Configuratio

Electric vehic

d there are di

d PHEVs. Th

s and PHE

tinctively. Fig

d PHEVs fo

arging [7].

. 5 Rate of pm.).

. 6 Rate of pe

der Widespre

is 87%, A at

culated by Eq

EVs and PHE

chedule 2, EV

n PV output

tarted from

surplus powe

ging at 11 p.m

the rate of p

12 a.m. [5,

is 28%, A at

culated by Eq

EVs and PHE

hers start ch

as 1. All E

on of Remaini

cle owners h

ifferences in

herefore, the

EVs start c

gs. 7 and 8 s

or the remai

people at hom

eople at home a

ead PVs and E

11 p.m. is 45

q. (1) is 94.

EVs start ch

Vs and PHEV

approaches th

this time. A

er produced b

m.

people stay ho

6]. As show

12 a.m. is 45

q. (1) is 67.

EVs start ch

arging at 11

EVs and PH

ing Battery

have various

the mileage

remaining b

charging sho

shows percen

ning battery

me and of car

and of car user

EVs

5.7%. Hence,

1%, and this

harging at 11

Vs stay home

he peak, and

And EVs and

by PVs. The

ome and rate

wn in Fig. 6,

5.7%. Hence,

1%, and this

harging at 12

p.m. by the

EVs do full

life patterns,

e on the EVs

battery when

ould be set

ntage of EVs

y when start

r users (at 11

rs (at 12 a.m.).

,

s

e

d

d

e

e

,

,

s

2

e

l

,

s

n

t

s

t

1

.

An

Fig. 7 Percecharging.

Fig. 8 Percestart charging

3. VerificPHEVs Ch

Figs. 9 a

residential a

period by ea

the power-d

simultaneou

power-dema

schedule co

charging s

increases in

Fig. 11 s

Figs. 12-14

charging sch

the night si

1,500 kW h

hand, the

schedule is

simultaneou

charging tim

In Figs. 1

night simult

alysis of Influ

entage of EVs f

entage of PHEg.

cation of Iharging

nd 10 show

area of heavy

ach charging

emand rises r

usly chargin

and is suppr

ompared wit

schedule. In

the daytime.

shows the po

show the lin

hedule. In Fi

multaneously

higher than it

power-dema

about 900 kW

usly charging

me zone.

2-14, the line

taneously cha

uence and Co

for remaining

EVs for remai

Influences

the power-d

y load perio

g schedule. In

rapidly at 11

ng schedule.

ressed by th

th the night

nstead, the

ower-demand

ne voltage at

ig. 11, the po

y charging sc

by without E

and by the

W lower tha

g schedule b

e voltage of e

arging schedu

ontribution o

battery when

ning battery w

by EVs a

demand curv

d and light

n Figs. 9 and

p.m. by the n

. The rise

he split char

t simultaneo

power-dem

d at 11 p.m.

11 p.m. by e

ower-demand

chedule is ab

EV. On the o

split char

an it by the n

by dividing

each phase by

ule is about f

n Distribution

start

when

and

ve of

load

d 10,

night

of

ging

ously

mand

and

each

d by

bout

other

ging

night

the

y the

from

150

othe

sch

EV

F

cha

pow

by

PHE

Fig.load

Fig.load

Fig.

n System und

0 V to 200 V

er hand, the

edule is only

by dividing t

From these

arging by EV

wer-demand a

dividing the

EVs, it was p

. 9 Power-ded period).

. 10 Power-dd period).

. 11 Power-de

der Widespre

V lower comp

e line voltag

y about 50 V

the charging

results, th

Vs and PHEV

and the steep

e charging

possible to mi

emand curve

demand curve

emand (at 11 p

ead PVs and E

pared with no

ge by the sp

lower comp

time zone.

e night sim

Vs occur the

voltage drop

time zone o

itigate these i

of residential

e of residenti

p.m.).

EVs 619

o EV. On the

plit charging

ared with no

multaneously

new peak of

p. In addition,

of EVs and

influences.

l area (heavy

al area (light

9

e

g

o

y

f

,

d

y

t

An

620

Fig. 12 Line

Fig. 13 Line

Fig. 14 Line

alysis of Influ

(a

evoltage of HV

(

evoltage of HV

(a

evoltage of HV

uence and Co

a) Heavy load p

V system at 11 p

(a) Heavy load

V system at 11 p

a) Heavy load p

V system at 11 p

ontribution o

period

p.m. (phase UV

period

p.m. (phase VW

period

p.m. (phase W

n Distribution

V).

W).

WU).

n System und

der Widespre

(b) Light loa

(b) Light load

(b) Light loa

ead PVs and E

ad period

d period

ad period

EVs

An

Fig. 15 Penereverse power

Fig. 16 Penereverse power

4. Verifica

Figs. 15 a

12 a.m. whic

In Figs. 1

batteries at

the penetrat

reverse pow

from 7% to

absorbed the

5. Conclus

In this stu

system mod

PVs, EVs an

contribution

of EVs and

alysis of Influ

etration rate or flow (heavy l

etration rate or flow (light lo

ation of Rev

and 16 show

ch can avoid

15 and 16, sin

12 a.m. acco

tion rate of

wer flow in e

o 10%. It i

e surplus pow

sions

udy, the auth

del of residen

nd PHEVs, an

ns by the two

PHEVs. In t

uence and Co

of PVs at 12 a.mload period).

of PVs at 12 a.moad period).

verse Powe

the penetratio

the reverse p

nce EVs and

ording to cha

f PVs which

ach case has

s because E

wer produced

ors structured

ntial area und

nd verified th

o kinds of ch

the charging

ontribution o

m. which can a

m. which can a

er Flow

on rate of PV

power flow.

d PHEVs char

arging schedu

h can avoid

s expanded ab

EVs and PH

by PVs.

d the distribu

der widesprea

he influences

harging sched

schedule 2,

n Distribution

avoid

avoid

Vs at

rged

ule2,

the

bout

HEVs

ution

ad of

and

dules

EVs

and

PHE

cha

tim

hom

and

PHE

othe

cha

and

othe

pow

mit

In

PVs

surp

und

occ

the

exp

und

occ

the

exp

pen

F

zon

suit

volt

and

by a

Re

[1]

[2]

[3]

n System und

d PHEVs base

EVs stay ho

arges are low

e. In the char

me at 12 a.m.

d the charge i

EVs charge

ers start char

arging schedu

d occurred st

er hand, by

wer-demand

igated.

n addition, th

s and verified

plus power o

der the chargi

urred when th

other hand, u

panded to 27

der the chargi

urred when t

other hand,

panded to 1

netration rate

From the abo

ne of electric

tably, the ra

tage drop are

d PHEVs can

absorbing sur

ferences

Institute of Report. 2008Management System—ConInterface. RepKoyanagi, F.Vehicle’s DeIEEJ Trans. PNext-GeneratMinistry o

der Widespre

ed on the ass

me at 11 p.m

w, and the ch

rging schedul

when PV ou

s started from

surplus powe

rging at 11

ule 1, the po

teep voltage

the chargin

rise and st

he authors ch

d the reverse

f PVs. In hea

ing schedule

he penetratio

under the cha

7.9%. In ligh

ing schedule

the penetratio

under charg

8.9%. And

about from 7

ove results,

vehicles like

apid power-

e mitigated. A

n contribute t

rplus power p

Electrical Eng8. Study on Et in Autonomntrol Method aport Number R., and Uriu, Y.emand and Its PE, IEEJ 117 (tion Vehicle of Economy,

ead PVs and E

sumption that

m. in case th

harge is start

le 2, EVs and

utput approach

m this time. A

er produced b

p.m. As a re

wer-demand

drop at 11

ng schedule

teep voltage

hanged the p

power flow

avy load peri

1, the reverse

on rate exceed

arging schedu

ht load perio

1, the reverse

on rate excee

ging schedule

the rate ex

7% to 10% in

when the ch

e EVs and PH

-demand rise

And the char

to the distrib

produced by P

gineers of JapaEnergy Deman

mous Demand nd Effect by D

R07013. 1997. “ModelImpact on the

1): 41-6. Strategy. 20

, Trade a

EVs 621

t all EVs and

he electricity

ted from this

d PHEVs stay

hes the peak,

And EVs and

by PVs. The

esult, by the

rose rapidly

p.m. On the

2, the rapid

drop were

enetration of

produced by

iod of Case 1

e power flow

ds 16.4%. On

ule 2, the rate

d of Case 1

e power flow

eds 7.6%. On

e 2, the rate

xpanded the

each case.

harging time

HEVs is split

e and steep

rging of EVs

ution system

PVs.

an Technologynd and Supply

Area PowerDemand Supply

ling of Electrice Daily Load.”

010. Japaneseand Industry.

d

y

s

y

,

d

e

e

y

e

d

e

f

y

w

n

e

w

n

e

e

e

t

p

s

m

y y r y

c ”

e .

Analysis of Influence and Contribution on Distribution System under Widespread PVs and EVs

622

http://www.meti.go.jp/committee/summary/0004630/pdf/20100412002-3.pdf.

[4] FUJI KEIZAI CO., LTD, HEV, EV-Related Market thorough Analysis Survey. 2013. http://www.group.fuji-keizai.co.jp/press/pdf/130301_13018.pdf.

[5] NHK Broadcasting Culture Research Institute Report. 2010. Survey of How the Nation`s People Use Their Time.

https://www.nhk.or.jp/bunken/summary/yoron/lifetime/pdf/110223.pdf.

[6] Ministry of Land, Infrastructure, Transport and Tourism. 2012. National Survey of City Traffic Characteristics. http://www.mlit.go.jp/common/001032141.pdf.

[7] Implemental Report of the Demonstration Experiment of EVs and PHVs in Saitama Prefecture. 2013. https://www.pref.saitama.lg.jp/a0502/documents/613045.