Page 1 of 5 2014-IACC-0478

Impact of Plug-In Electric Vehicle Battery Charging on a Distribution System

S. K. Bunga A. H. Eltom N. Sisworahardjo

Student Member, IEEE Senior Member, IEEE
University of Tennessee at Chattanooga University of Tennessee at Chattanooga
615 McCallie Ave 615 McCallie Ave
Chattanooga, TN, 37403 Chattanooga, TN, 37403
sharmilabunga@gmail.com Ahmed-Eltom@utc.edu
Abstract -- Plug-in Electric Vehicles (PEV) battery chargers
are mostly connected to the low-voltage grid for charging. Their
increased penetration coupled with uncoordinated charging
could impact the distribution system in terms of unbalanced
voltage and transformer overloading. Although PEV battery
charging is increasing, impact on the distribution system is not
fully investigated. This paper focuses on unbalanced voltage
caused by uneven distribution of PEV penetration among the
phases. Using real data provided by utility, a distribution system
has been modeled and tested using Matlab-Simulink. PEV
penetration levels at 10-80% are studied, unbalanced voltage is
calculated, and transformer overloading is observed. As PEV
adoption is expected to increase, the impact on the distribution
system will increase. Coordinated or smart charging of PEVs will
be essential for consumers and utilities.
Index Terms-- Plug-in electric vehicle, battery charging,
unbalanced voltage, and transformer overloading.
I. INTRODUCTION
Traditional dependence on fossil fuels has undoubtedly
made the transportation a primary contributor to the
greenhouse emissions. As passenger vehicle is accounting for
more than half of the total electrification of transportation
energy requirement, it is the major consumer of the energy [1].
However the technological strides that are made and being
made in the field of plug-in electric vehicles (PEVs) are
becoming central to combating greenhouse gas emissions,
namely the Battery Powered Electric Vehicle (BEV) and Plug-
in Hybrid Electric Vehicle (PHEV). The promising direction
of PEVs for transportation is two-fold; PEVs can reduce the
reliance on fossil fuels hence limiting the greenhouse gas
emissions, as well, electricity as an energy source for
passenger cars is less expensive per mile than fossil fuels [2].
Lately there has been a widespread PEV adoption by
consumers, which is only expected to increase manifold in near
future [3, 4]. This is expected to add a significant uneven load
among the phases at the primary and secondary distribution
networks and overload the transformers. Essentially this put
more stress on old inherently unbalanced system designed
decades ago based on end users loads at that time. In addition
to overloading at primary and secondary circuit, PEVs could
cause significant unbalanced voltage among phases. This is
due to unequal single and double phase charging by level 1 and
Member, IEEE
University of Tennessee at Chattanooga
615 McCallie Ave
Chattanooga, TN, 37403
nurh@ieee.org
level 2 chargers. This will have a direct impact on induction
machine performance [5], and may result in increased line
losses, harmonics and other power quality problems that could
damage and or negatively impact the performance of utility
and end-users equipment [6, 7].
Essentially such an adoption would put more stress on
electric utilities particularly at the distribution level as more
and more consumers expect to use more energy in the form of
electricity.
The goal set by the Obama administration is one million
plug-in electric vehicles (PEVs) on the road by 2015 in the US.
This goal is being supported by government incentives and
stimulus investments to accelerate market acceptance,
including grants and loans to manufacturer and tax credits to
consumers [8]. Projected market penetration of PHEVs could
reach a maximum of 10% of new vehicle market share by
2015, 35% by 2020, and 50% by 2025 [9]. In addition, various
U.S. state governments have passed their own laws to promote
PEVs by providing grants for electrical vehicle research,
funding for PEV charging infrastructure, and loans or tax
credits for green technology related businesses.
PEV chargers are devices that transmit electric energy from
grid into the PEV batteries. Currently there are three levels of
battery charging techniques available to recharge the PEV
batteries, classified as Level 1, Level 2 and Level 3 chargers
[10]. Level 1 and 2 are single phase and Level 3 is three phase.
Level 1 chargers are standard 120 V/16 A wall outlet and has
a corresponding charger rating of 1.8 kW and are primarily
located at homes [11]. Level 2 chargers have higher voltages,
typically 208-240 V and draws more current up to 80 A. They
could be charged at home or at public chargers. They have
shorter charging time than Level 1 chargers. Level 3 is a three-
phase battery charger and has a maximum capacity of 96 kW,
rated at 208-600 V/400 A. Generally level 3 chargers are not
available in residential areas, because this type of chargers
draw much more current than Level 1 and 2 chargers, and
hence require very short period of charging time [12, 13]. Of
these three levels, only Level 1 and 2 chargers could impact a
distribution system in terms of unbalanced voltage and
transformer overloading. This study is focused on impact of
Level 2 chargers on the residential distribution grid.
From the consumer point of view, the PEV batteries have to

978-1-4799-2288-8/14/$31.00 2014 IEEE

 2014-IACC-0478 Page 2 of 5

be charged so the driver can drive off with a fully-charged
battery. This brings in the question of place and time for
charging these batteries. There are two main places where the
PEV batteries can be recharged: either on a corporate or public
car park, or at home. Irrespective of the location,
uncoordinated power consumption that can result from this
charging activity on a local scale can lead to grid problems.
The charging of PEVs has an impact on the distribution grid
because these vehicles consume a large amount of electrical
energy and this demand of electrical power can lead to extra-
large and undesirable peaks in the energy consumption. The
impact of these extra single phase electrical loads can be
analysed in terms of power losses and unbalanced voltages
[14].
the metered customers connected to it. The service transformer
serves the customers at 480V, 240V and 120V. The modelled
system is comprised of total 44 service transformers; of which
41 transformers are single phase (7200-120/240V) and 3
transformers are the three phase transformers. Single phase
transformers ratings range from 15-100 kVA and three phase
transformers sizes ranging from 15-500 kVA. As shown in the
Fig 2, the studied system comprises of a three phase source, V-
I measurements, subsystem sections 1 and 2 and scopes.
V LOAD
VCR1
UTC777
X Y
Section2
A55661

From the distribution system operator point of view, the X Y X Y X Y

power losses during charging are of an economic concern and Section1
B51432
R
C55453 D56914
transformer and feeder overloads are of a reliability and safety CAP1
REG1

concern [15]. In addition, power quality (i.e., voltage profile,

unbalanced voltage, and harmonics) is essential to the
H91668
distribution grid operator as well as to grid customers. Voltage X Y
P PCRs ALTB
deviations and harmonics are a definite power quality concern E52785
[16]. They are reported to cause equipment damage and
LOAD
deterioration of efficiency [5, 6].
Using real data provided by a utility, the impact of X Y
LOAD
uncoordinated PEVs battery charging on a 12.47 kV F55966
distribution system in terms of maximum voltage deviation, CAP2
unbalanced voltage at various locations, and transformers
PCRs ALTA
overloading is investigated. Calculating voltage unbalance G58157
over a distribution grid due to wider consumption of electricity
Fig. 1. One line diagram of the studied system.
for charging PEVs is central in finding a solution such as
smart or coordinated charging.
The remaining of the paper is organized as follows: Section Discrete,
RMS Voltage A5566
3-ph voltage To Workspace
II presents the model of distribution system. In Section III, case Ts = 5e-005 s.

study and simulation are performed and discussed, with RMS

3-ph current
Current A5566
To Workspace2
conclusion in Section IV.
Vabc
A
A a Iabc
II. MODELING OF A DISTRIBUTION SYSTEM A
Conn1 Conn2
Vabc_1
N B B b B a

A. Distribution System Model C

C c b Conn3 Conn4
3-Phase C
A 12.47 kV distribution system in this study was modelled Voltage Source 3-Phase c
Conn5 Conn6
and tested using Matlab/Simulink/SimPowerSystem (Version 161KV
Transformer V-I @ A5566
Subsystem_Sec1
7.12 R2011b). The existing distribution system consists of 161/12.47 KV
15 MVA
residential metered and primary metered customers. Like most
Vabc
distribution systems, the modelled system is a radial system, Vabc_2 3-Phase
RLC Load Iabc
A

receiving power from a three phase source at 12.47 kV. Fig. 1 A a B

b
shows a one line diagram of a distribution system provided by B
C c
C
Vabc_3
a power distributor for this study. V-I @
Primary
While the distribution system consists of several sections, Voltage_B51432 RMS
meter
To Workspace1 RMS2
this study is based on the first two sections of the system for Vabc
A Conn17 Conn15
the household metered customers (as shown in Fig. 1). The Current_B51432 RMS Iabc

distribution system feeds power to the metered customers To Workspace3 RMS3

A
a B Conn20 Conn16

through a main feeder. The main feeder line consists of several B b

Conn21 Conn19
branch lines. The main feeder has overhead (OH) lines and the 3-Phase
C c
C

branch lines consists of overhead (OH) lines and underground RLC Load1 V-I @ B51432
Subsystem_Sec2

(UG) cables. Each service transformer serves the demand of Fig. 2. Matlab/Simulink model of the studied system.

978-1-4799-2288-8/14/$31.00 2014 IEEE

Page 3 of 5 2014-IACC-0478

B. Distribution System Load
Fifteen minutes intervals load data of four days (01/09/2013
- 01/12/2013) for the distribution system are provided by the
power distributor. The load data measured at the substation
include all residential metered customers and primary metered
customers served by the system. Primary metered customers
are major customers who purchase bulk energy. However the
primary metered customers are not considered in the system
model for this study because of lack of data for those
customers. Figs. 3 and 4 shown are the active and reactive load
profiles of the distribution system respectively.
represented by the difference of load data at CB1 and CB2.
Out of four days (01/09/2013 - 01/12/2013) load profile
data, one day (01/10/2013) load data is selected for testing the
distribution system model and the simulation results are
compared with real load data for that particular day. The
assessment of the impact of PEV battery charging on the
distribution system during peak and off peak load demand was
performed. From the load data at CB1, it is indicated that the
peak demand occurred at around 2.30 PM and off peak demand
at 6.30 AM on 01/10/2013 (Figs. 3 and 4).
Section 1

T/F_B
4.5 3_Ph T/F
T/F_A
3_Ph T/F
T/F_A
4.0 Distribution
UG Cable
T/F_A L4 12_T/F &
38 Cust.
Substation 3_Ph T/F T/F_A
T/F_A Main Feeder
UG Cable UG Cable
3.5
4_T/F &
Active Power (M W )

CB1 4_T/F &

72 Cust.
CB2 Load
12 Cust. T/F_A
161/12.47 KV T/F_A T/F_C T/F_A Cap. Bank
3.0 15 MVA
L1 10_T/F &
L3
16 Cust. T/F_B
L2
2.5
1/9/2013 Section 2
2.0
1/10/2013 Fig. 5. Simplified Diagram of Distribution System Model.
1/11/2013
1.5
1/12/2013 III. CASE STUDY AND SIMULATION RESULTS
1.0
0:08 3:08 6:08 9:08 12:09 15:09 18:09 21:09 As shown in Fig. 5, the distribution system model has 177
Time (Hour) customers in total, of which 174 customers are connected on
Fig. 3. Total Active Power for 4 days. single-phase and 3 customers are connected on three-phase for
their electricity consumption and are served by the 161/12.47
0.6
kV distribution substation. In order to assess the impact of
0.5
PEV battery charging on the distribution system, various
locations were selected to measure the three-phase voltages
0.4 based on the number of customers connected to the 12.47 kV
Reactive Power (M VAR)

1/9/2013 distribution system.

0.3
0.2
0.1
0.0
0:08 3:08 6:08 9:08
Time (Hour)
Fig. 4. Total Reactive Power for 4 days.
C. Distribution System Load Data Analysis
By having total active and reactive load data for the entire
system and knowing residential metered customers energy
usage for the first two sections, calculations are made to
estimate the reactive power consumed by the metered
customers for those two sections of the system. Intelligent
switches A55661 and B51432 in Fig. 1 represents circuit
breakers CB1 and CB2 respectively in the distribution system
model, Fig. 5. The data at CB1 represents the load data for the
entire system, whereas the data at CB2 represents the load data
for the later part of the system. Given that the load data for the
sections 1 and 2 that are part of this study, essentially
1/10/2013 As depicted in Fig. 5, L1, L2, L3, and L4 indicate the
1/11/2013 locations where the three-phase voltage and current are
1/12/2013
measured. The selection of the locations is based on the fact
that each of these locations has more than 10 customers
connected to the sub branch line. Three-phase measurements
were observed at CB1 and CB2 to identify voltage variations
12:09 15:09 18:09 21:09 at the beginning and at the end of the distribution system
model.
A. Model Testing Without PEV Penetration
As a benchmark from which PEV impacts can be compared,
the distribution system model was simulated without PEV
battery charging. As indicated in Table 1, the simulated system
without any PEV penetration is very similar to the actual
distribution system in terms of phase voltages and currents.
Hence the simulated system without PEV penetration is used
as a benchmark in this study.
B. Unbalanced Voltage Calculation
A single-phase level 2 charger can have significant impact
on the distribution system, which is measured in terms of
unbalanced voltage. According to IEEE standards, unbalanced
voltage is defined as the ratio of maximum voltage deviation

978-1-4799-2288-8/14/$31.00 2014 IEEE

 2014-IACC-0478 Page 4 of 5

from the average phase voltage to the average phase voltage added to Phase b for further penetration levels. At 80% PEV
[17, 18]. It is also known as the percentage unbalanced voltage penetration level, phase a (79 out of 81 customers), and phase
(%VU), and is calculated as c (12 out of 12 customers) are almost maxed out the number of
PEVs can be added to the distribution system, therefore further
Max Volt. Dev. from Avg. V levels of penetration (90%, 100%) cannot be simulated in this
%VU = 100 (1) study.
Avg. V
D. Impact of PEV Penetration on Simulated System
The IEEE uses the same definition of unbalanced voltage as
The percentage of unbalanced voltage is the primary metric
NEMA, the only difference being that the IEEE uses phase
that was used to assess the impact of increased levels of PEV
voltages rather than line-to-line voltages. In this study,
penetration (10%, 30%, 50%, 60%, 70%, and 80%). From Fig.
percentage unbalanced voltage calculations are performed
6, it can be observed that for the peak and off-peak demands,
according to IEEE standards (1).
the percentage of unbalanced voltage is constantly increased
TABLE I
with every level of increase in PEV penetration. The observed
SIMULATED AND ACTUAL PHASE VOLTAGES AND CURRENTS AT PEAK AND voltage unbalance in the simulated system can be attributed to
OFF-PEAK DEMANDS OF THE DISTRIBUTION SYSTEM WITHOUT PEV uneven distribution of single-phase loads over the three phases
PENETRATION
[5].
Vl-n (kV) Phase currents (A) In the simulation that was conducted on all-phase of the
Load CB# Status
Va Vb Vc Ia Ib Ic distribution system model with varying penetration from 10 to
CB1 Simulated 7.11 7.11 7.11 189 187 178 80%, the highest unbalanced voltage is 1.23%. This is slightly
Actual 7.14 7.11 7.09 197 195 186
Peak
CB2 Simulated 7.06 7.08 7.08 60 60 60
higher than the acceptable percentage of unbalanced voltage,
Actual 7.12 7.09 7.09 58 79 52 i.e., 1% for a motor terminal as per NEMA motor standard [7].
CB1 Simulated 7.13 7.14 7.14 130 127 117 An excessive level of unbalanced voltage causes a lot of ill
Off- Actual 7.12 7.10 7.06 135 131 123 effects on induction motors. The ill effects of voltage
Peak CB2 Simulated 7.10 7.12 7.12 49 49 49
Actual 7.10 7.08 7.06 48 62 43 unbalance in induction motors include overheating rotor bars,
lower torque, and poor efficiency. Under unbalanced
conditions the negative sequence of motor current produces
C. Model Testing With PEV Penetration retarding torque. This retarding torque must be countered by
There are 174 single-phase residential customers in the increased torque in the direction of rotation and hence
distribution system model. Out of these 174 customers, 81 increased current levels in the motor. Due to skin-effect, the
customers draw power from phase a, 81 customers from phase effective rotor resistance to these currents is much larger than
b, and 12 customers from phase c. The distribution system the rotor positive sequence resistance, which causes negative
model was simulated for various levels of PEV penetration and sequence currents produce more rotor heat than positive
unbalanced voltages were measured. sequence current. This heating will reduce the motor life and
The PEV penetrations are as follows: may cause motor failure [6].
10% PEV Penetration - 17 Customers (8 PEVs were added
to the phase a, 8 PEVs were added to the phase b, and 1 1.35
PEV was added to the phase c). 1.20 Peak CB1
30% PEV Penetration - 51 Customers (24 PEVs were 1.05 O-Peak CB1
added to the phase a, 24 PEVs were added to the phase b,
Unbalanced Voltage (%)

Peak CB2
0.90
and 3 PEVs were added to the phase c). O-Peak CB2
0.75
50% PEV Penetration - 87 Customers (40 PEVs were
0.60
added to the phase a, 41 PEV were added to the phase b,
and 6 PEVs were added to the phase c). 0.45

60% PEV Penetration - 104 Customers (49 were added to 0.30

the phase a, 48 PEVs were added to the phase b, and 7 0.15
PEVs were added to the phase c). 0.00
70% PEV Penetration - 122 Customers (64 PEVs were 0% 20% 40% 60% 80% 100%
added to the phase a, 48 PEVs were added to the phase b, PEV Penetration Level (%)

and 10 PEVs were added to the phase c).
80% PEV Penetration - 139 Customers (79 PEVs were
added to the phase a, 48 PEVs were added to the phase b,
and 12 PEVs were added to the phase c).
At 60% PEV penetration, overloading of phase b
transformers was observed. Hence no additional customers are
Fig. 6. Percent unbalanced voltage vs. level of PEV Penetration.
For comparison purpose an additional study was conducted
with 81 PEVs were connected to phase a and no PEV
penetration on the other two phases. It is observed that the
highest unbalanced voltage for this case was 1.5%.

978-1-4799-2288-8/14/$31.00 2014 IEEE

Page 5 of 5 2014-IACC-0478

IV. CONCLUSION [11] E. Sortomme, M. Hindi, S. MacPherson and S. Venkata, "Coordinated

This study examines the impacts of PEV battery charging
on a 12.47 kV distribution system at various penetration levels
of PEVs. This study is focused on percentage unbalanced
voltage and service transformers overloading. It was found that
percentage unbalanced voltage is negatively impacted by
increased PEV penetration levels in the residential distribution
grid. Transformer overload was observed for one of the phases
of simulated distribution system at a higher penetration level
of PEVs.
Since these challenges will be inevitable with projected
adoption of PEVs, smart or coordinated charging, grid
reinforcements, optimization of electric grid will become
indispensable to meet the new energy requirements without
straining the infrastructure.
Addressing the challenges that have been mentioned earlier,
future work will be focus on developing simulation test bed
based on real-time digital simulator and investigate the impact
of PEV penetration on the end users equipment in particular
Charging of Plug-In Hybrid Electric Vehicles to Minimize Distribution
System Losses," Smart Grid, IEEE Transactions on, vol. 2, no. 1, pp.
198-205, 2011.
[12] Z. Darabi and M. Ferdowsi, "Aggregated Impact of Plug-in Hybrid
Electric Vehicles on Electricity Demand Profile," Sustainable Energy,
IEEE Transactions on, vol. 2, no. 4, pp. 501-508, 2011.
[13] C. Pang, P. Dutta and M. Kezunovic, "BEVs/PHEVs as Dispersed
Energy Storage for V2B Uses in the Smart Grid," Smart Grid, IEEE
Transactions on, vol. 3, no. 1, pp. 473-482, 2012.
[14] K. Clement-Nyns, E. Haesen and J. Driesen, "The Impact of Charging
Plug-In Hybrid Electric Vehicles on a Residential Distribution Grid,"
Power Systems, IEEE Transactions on, vol. 25, no. 1, pp. 371-380,
2010.
[15] C. Farmer, P. Hines, J. Dowds and S. Blumsack, "Modeling the Impact
of Increasing PHEV Loads on the Distribution Infrastructure," in
System Sciences (HICSS), 2010 43rd Hawaii International Conference
on, 2010.
[16] F. Geth, N. Leemput, J. Van Roy, J. Buscher, R. Ponnette and J.
Driesen, "Voltage droop charging of electric vehicles in a residential
distribution feeder," in Innovative Smart Grid Technologies (ISGT
Europe), 2012 3rd IEEE PES International Conference and Exhibition
on, 2012.

and distribution network performance in general. [17] K. Lee, G. Venkataramanan and T. Jahns, "Modeling Effects of Voltage
Unbalances in Industrial Distribution Systems With Adjustable-Speed
Drives," Industry Applications, IEEE Transactions on, vol. 44, no. 5,
V. REFERENCES pp. 1322-1332, 2008.
[18] P. Pillay, P. Hofmann and M. Manyage, "Derating of Induction Motors
Operating with a Combination of Unblanced Voltages and Over-Or
[1] G. Putrus, P. Suwanapingkarl, D. Johnston, E. Bentley and M. under-Voltages," Power Engineering Review, IEEE, vol. 22, no. 7, pp.
Narayana, "Impact of electric vehicles on power distribution networks," 51-51, 2002.
in 2009 IEEE Vehicle Power and Propulsion Conference, VPPC '09.,
2009.
[2] C. H. Stephan and J. Sullivan, "Environmental and Energy Implications
of Plug-In Hybrid-Electric Vehicles," Environmental Science \&
Technology, vol. 42, no. 4, pp. 1185-1190, 2008.
[3] M. Brenna, F. Foiadelli and D. Zaninelli, "Power and energy estimation
for Plug-in electric vehicles recharge in metropolitan area," in Electrical
Power Quality and Utilisation (EPQU), 2011 11th International
Conference on, 2011.
[4] N. Rotering and M. Ilic, "Optimal Charge Control of Plug-In Hybrid
Electric Vehicles in Deregulated Electricity Markets," Power Systems,
IEEE Transactions on, vol. 26, no. 3, pp. 1021-1029, 2011.
[5] A. Eltom and M. Aziz, "The economics of energy efficient motors
during unbalanced voltage conditions," in Power Engineering Society
Inaugural Conference and Exposition in Africa, 2005 IEEE, 2005.
[6] A. Eltom and A. Demirbas, "Motor system energy efficiency in the
nylon industry: A comparison of PWM and square wave inverters," in
Electric Machines and Drives Conference, 2009. IEMDC '09. IEEE
International, 2009.
[7] N. Moharari, Microprocessor Based Protection of Induction Motors
Using Thermal, Mechanical, and Skin Effect Electrical Models to
Predict Motor Temperature Rise, University of Tennessee at
Chattanooga, Engineering, 1989.
[8] J. T. Salihi, "1974 Energy Requirements for Electric Cars and Their
Impact on Electric Power Generation and Distribution Systems,"
Industry Applications, IEEE Transactions on, Vols. IA-20, no. 4, pp.
1095-1095, 1984.
[9] J. Taylor, A. Maitra, M. Alexander, D. Brooks and M. Duvall,
"Evaluation of the impact of plug-in electric vehicle loading on
distribution system operations," in Power Energy Society General
Meeting, 2009. PES '09. IEEE, 2009.
[10] S. Haghbin, S. Lundmark, M. Alakula and O. Carlson, "Grid-Connected
Integrated Battery Chargers in Vehicle Applications: Review and New
Solution," Industrial Electronics, IEEE Transactions on, vol. 60, no. 2,
pp. 459-473, 2013.

978-1-4799-2288-8/14/$31.00 2014 IEEE