Dynamic Simulation of EV Fast Charging with
Integration of Renewables
Yao Chen
ABB Corporate Research China
Power Systems Department
Beijing, China
[email protected]
Hector Zelaya De La Parra
ABB Corporate Research Sweden
Power Technologies
Vasteras, Sweden
[email protected]
Fabian Hess
Product Group EV Charging
Infrastructure
ABB Switzerland
Turgi, Switzerland
[email protected]
Abstract: This paper presents the results of a dynamic simulation
of various scenarios of Electric Vehicle (EV) fast charging in
combination with renewable generation. Based on a hypothetical
distribution network, the possible benefits from coordinated
controls of fast charging and renewable generation are studied
under grid connection and islanding condition. For grid
connection condition, strong and weak grid scenarios are both
simulated; for islanding condition, energy storage requirements
are considered. The simulation results show that with proper
control strategies, fast charging infrastructure can help to
compensate active and reactive power in different conditions, so
as to improve the operation performance of power systems with
renewable energy integration.
Keywords: EV charging infrastructure; DC fast charging;
renewable generation; islanding operation; supervisory controls;
grid impact analysis
I. INTRODUCTION
World governments put high priority on electric vehicle
industry development in recent years as EV is regarded to be
one of the most promising ways to achieve the target of cutting
the vehicle emissions. Take China for example, which has the
second largest vehicle market after the U.S., the government
forecasts 500,000 EVs on road by end of 2015, and is targeting
5 million EVs by 2020 [1]. In order to provide charging service
for massive EVs, the existing power grids need to be upgraded
by building up charging infrastructure. As announced by State
Grid Company of China, they plan to set up 1000 charging
stations and 240,000 charging poles by end of 2015, and the
overall investment is estimated to be over 20 billion RMB [2].
The large scale charging infrastructure, left without proper
regulation, might lead to new challenges for power system
operation, for example change of power flow patterns, increase
of peak load, power quality problems [3-4]. The technologies
to realize a smooth Vehicle to Grid (V2G) integration become
the foci of proactive studies [5-8] and are being introduced into
some of the utilities pilot projects for demonstration.
However, these studies tend to treat EV charging stations
separately and few of them look into the dynamics of the
system when applying a supervisory control strategy. Papers
[9-10] start to look at the possible dynamic control for fast
charging stations, but still lacking supervisory control
strategies and comprehensive demonstration.
In this paper, an EV charging infrastructure is not
considered as an isolated part of the power system, but in
combination of renewable generation which is also under rapid
development nowadays, in order to understand the interactions
in between and the coordination requirements to maintain the
normal operation of power grids. Firstly, a dynamic system
model of a hypothetical distribution network has been
developed using Matlab/Simulink environment. This also
includes fast charging stations comprised of several poles and
renewable generation to investigate the interaction of fast
charging stations when connected to the vicinity of wind or
solar power stations. Secondly, several conditions of interest
are analyzed, either due to low wind conditions combined with
weak or strong grid or under islanding condition when the
charging stations can contribute with active and reactive power
injection into the system. The paper investigates those different
scenarios with particular attention to active and reactive power
requirements and bus voltages and frequency behavior.
Thirdly, the control strategies of fast charging in coordination
with renewable generation are developed and implemented in
the simulation model. The performance improvements are
demonstrated by the simulation results.
II. MODELLING
A. Power System (Grid) Model
The power system model is based on Matlab/Simulink as
has already been reported in [9]. The model objective is to
investigate EV fast charging interaction with the grid. The
single line diagram is shown in Fig. 1, where Node 799
represents the Point of Common Coupling (PCC) of the
distribution network, the network parameters and the loading
condition of each nodes, either symmetrical or asymmetrical,
are set according to IEEE 37 Node Test Feeder [11]. As
indicated in Fig. 1, N=3 fast charging stations with a total
capacity of 500 kW each are connected to Node 712, Node 720
and Node 709 respectively. Each charging station may consist
of M=5~10 DC charging points with rated capacity 50kW [12]
to 100kW and above [13]. A distributed renewable generation
is connected to Node 708 with fluctuating output. Besides,
there is a synchronous generator connected to Node 703 which
is stand-by in grid-connected operation and will be switched in
when islanding occurs.
Alternatively, a setup of N>25 highly distributed smaller
charging systems with an aggregated capacity of 500kW can be
considered, the rated capacity of each charging point could be
in the range of 10~50kW. However this scenario is not
modeled in this paper because it is equivalent to the N=3
scenario from a coordinated control strategy point of view.
799
701
742
705 702
720
704 713
707
722
703
744 729
728
727
706
725
718
714
730
731
709
708 732
775 733
736
734 710
735
737 738 711 741
740
724
712
Figure 1. Power system topology diagram
B. Fast Charging Station Model
The fast charging station model consists of several blocks:
battery and management system, line-side active rectifier
capable of four-quadrant operation, bi-directional DC/DC
converter allowing battery charging and discharging, and
corresponding converter control systems.
Two different models can be developed: (i) detailed model
with the two power electronics converters which allows
detailed analysis of the harmonics and power factor correction
available in the system; (ii) phasor model that simplifies the
complexity of the modulation strategy for system level
investigations. The latter one is adopted in the results presented
in this paper as shown in Fig. 2.
3
C
2
B
1
A
Ambient Temperature
Vehi cl e Inputs
V/I Measurement
f(x)=0
Sol ver
Confi gurati on
PS S
Eabc
Iabc
P
Q
Si gnal s
& Scopes
P/Q Measurement
m1
Idc
T
e
m
p
+-
Confi gurabl e
Battery
Batt_Idc
Trip
m
A
B
C
Four-quadrant
DC Fast Charger
AC/DC + DC/DC
Figure 2. Fast charging station phasor-model
C. Modelling of Renewable Sources
The renewables sources of generation have been modeled
as simple power sources in the system as the purpose of the
investigation is to study system effects during step changes.
The output of renewable generation will vary according to
predefined tables to simulate its fluctuation characteristics.
Based on which, the distributive fast charging stations will be
controlled so as to optimize the operation of the network.
III. GRID-CONNECTED SYSTEM
The following two scenarios are simulated for grid-
connected operation by changing the short-circuit level and the
X/R ratio (reactance at the nominal frequency 50Hz over
resistance) of the external grid.
- Strong Grid Connection: system short-circuit level =
100MVA, X/R ratio = 7;
- Weak Grid Connection: system short-circuit level =
50MVA, X/R ratio = 3.
During grid-connected operation, the frequency is
determined by the large external grid and is therefore
considered to be constant within the studied distribution
network. However, the voltage profile might be different from
node to node depending on the power flow. The change of
loading level and the fluctuation of renewable generation can
both lead to voltage variation which may exceed the acceptable
range in certain cases. Therefore the main issue to address is
how to utilize the four-quadrant operation capability of fast
charger for reactive power regulation and voltage support.
A. Control Strategy
Given a step change in the renewable power available, the
bus voltages in the system are affected with a corresponding
variation from nominal operation. The control strategy consists
of applying the required reactive power in order to maintain
zero voltage variations. The supervisory control at high level
verifies first the availability of power on each charging station
and it then decides the reactive contribution per station. Since
there are many system voltages available in the system, the
controller uses the bus further away from the PCC point of the
network as representative of the bus voltages in the system. In
the simulation, the voltage signal of Node 709 is selected as the
control objective. The reactive power injection is then
calculated to compensate for this worst case node in the
EV
EV
EV
RE
SG
system. A block diagram of the control function is shown in
Fig. 3.
Figure 3. Control strategy flow chart for grid-connected scenario
B. Results
The results of the grid-connected system are shown in
Fig.4 and Fig. 5, corresponding to strong grid connection and
weak grid connection respectively. In the simulation, the
loading level of the network is about 2.5MVA. The total
charging power of the three fast charging stations is about
550kW. Renewable generation varies during a time interval of
1.5s between a level of 1.5MW and 0.3MW to simulate the
fluctuation of wind speed or PV output due to clouds.
The results show that the DC fast charging stations can take
care of the reactive power needs for the simulated scenarios.
Under both scenarios, it can be noticed that with the
coordinated reactive power control of fast charging stations,
the voltage drop at each observed nodes can be reduced by
50% when step changes of renewable generation occur. This is
a simple and elegant solution that emphasizes the advantages
of a fully controlled four-quadrant AC/DC converter on the
chargers front end, eliminating the need for external
equipment such as SVC.
1.5 2 2.5 3 3.5 4 4.5
0.88
0.89
0.9
0.91
0.92
0.93
0.94
Voltage profiles without coordinated control under strong grid connection condition
Time/s
V
o
l
t
a
g
e
/
V
Node 701
Node 702
Node 703
Node 730
Node 709
1.5 2 2.5 3 3.5 4 4.5
0.88
0.89
0.9
0.91
0.92
0.93
0.94
Voltage profiles with coordinated control under strong grid connection condition
Time/s
V
o
l
t
a
g
e
/
V
Node 701
Node 702
Node 703
Node 730
Node 709
Figure 4. Voltage profile comparison under strong grid connection condition
1.5 2 2.5 3 3.5 4 4.5
0.85
0.86
0.87
0.88
0.89
0.9
0.91
Time/s
V
o
l
t
a
g
e
/
V
Voltage profiles without coordinated control under weak grid connection condition
Node 701
Node 702
Node 703
Node 730
Node 709
1.5 2 2.5 3 3.5 4 4.5
0.86
0.87
0.88
0.89
0.9
0.91
0.92
0.93
Time/s
V
o
l
t
a
g
e
/
V
Voltage profiles with coordinated control under weak grid connection condition
Node 701
Node 702
Node 703
Node 730
Node 709
Figure 5. Voltage profile comparison under weak grid connection condition
IV. ISLANDING OPERATION
During islanding operation, the synchronous generator as
mentioned above will be switched in to control the frequency
and voltage of the islanded network. However, the dynamic
changes from renewable generation will produce transients that
might have an effect on the voltage stability and the frequency
of the system. In this section, the possible performance
improvement by coordinate active and reactive power control
of EV charging stations will be studied. The same power
system diagram of Fig. 1 applies here.
A. Control Strategy
The control strategy during islanding is different to the
previous case. The concern here is to maintain the balance
between the active and reactive power. The control strategy
also include detection of the islanding condition, this can be
achieved by one of several available methods, e.g. natural
frequency detection, THD (total harmonic distortion)
measurement, automatic disconnection signal, etc.
As long as islanding operation is detected, besides reactive
power compensation, the supervisory controller will also
enable coordinated active power control of the distributed fast
charging stations. It is assumed that the fast charging stations
has a certain battery storage capability available. The
supervisory controller will verify the available energy storage
capacity of each station, measure the output variation of
renewable generation, and decide whether to charge or
discharge the energy storage system and the corresponding
charge/discharge power according to the built-in logic. The
block diagram of the control function is shown in Fig. 6.
Figure 6. Control strategy flow chart for islanding scenario
B. Simulation Results
The results of the islanding system simulation are shown in
Fig. 7 and Fig. 8, which are the speed and the terminal voltage
of the synchronous generator under the fluctuation of
renewable generation with or without the coordinated active
and reactive power control of EV charging stations. In the
simulation, the capacity of the synchronous generator is
4MVA. The total energy storage capacity of the three fast
charging stations is 1MVA. Renewable generation varies
during a time interval of 0.8s between a level of 0.8MW and
2.5MW to simulate the possible wind power output during
gust.
13 13.5 14 14.5 15 15.5 16 16.5 17 17.5 18
0.99
0.995
1
1.005
1.01
1.015
1.02
Time/s
S
p
e
e
d
/
p
u
Generator speed under renewable output fluction during islanding operation
w/o control
w/t control
Figure 7. Generator speed during islanding operation under renewable
generation fluctuation
13 13.5 14 14.5 15 15.5 16 16.5 17 17.5 18
0.88
0.9
0.92
0.94
0.96
Time/s
V
o
l
t
a
g
e
/
p
u
Generator terminal voltage under renewable output fluctuation during islanding operation
w/o control
w/t control
Figure 8. Generator terminal voltage during islanding operation under
renewable generation fluctuation
The results show that the fast charging stations can help to
reduce the impact of fluctuating renewable generation on
system frequency and voltage substantially during islanding
operation. The amplitude of frequency variation of the
generator is reduced by 50%. The voltage stabilizing effect is
even more obvious, not only is the voltage variation reduced
during renewable generation fluctuation period, but also the
oscillation damping after the recovery of renewable generation
is much faster with the developed control strategy.
V. CONCLUSIONS
This paper presents the simulation results of a hypothetical
smart grid system composed by several sources of renewable
generation that interact together with fast charging stations for
both conditions: grid-connected and islanding. It has been
shown that the four-quadrant operation of the DC fast
charging station has the advantage of being able to
compensate for active and reactive power in different
conditions of the power system. Two supervisory control
strategies have been studied to manage the flow of active and
reactive power in the system. The performance improvements
under both grid-connected and islanding operation have been
demonstrated by the simulation results.
REFERENCES
[1] Chinas EV Industry Leaps Forward. http://www.chinadaily.com.cn/busi
ness/2010-09/03/content_11237059.htm
[2] SGCC plans to invest 20 billion for electric vehile charging infrastructur
e during the Twelveth Five-Year-Plan. http://www.escn.com.cn/2011/12
30/91538_2.html
[3] L. Z. Xu, G. Y. Yang, Z. Xu, et al. Impacts of Electric Vehicle
Charging on Distribution Networks in Denmark, Automation of
Electric Power Systems, Vol. 35, No. 14, pp. 18-23.
[4] H. L. Li, X. M. Bai. Impacts of Electric Vehicles Charging on
Distribution Grid, Automation of Electric Power Systems, Vol. 35, No.
17, pp. 38-43.
[5] H. Y. Han. The Study on the Control Strategy of V2G Participating
Peak Regulation and Frequency Regulation of the Grid, Master Thesis,
Beijing Jiaotong University, June 2011.
[6] R. S. Li, X. L. Wang, F. Q. Zhou, et al. The system of electric vehicle
intelligence charge station with smart power flow control, Power
System Protection and Control, Vol. 38, No. 21, pp. 87-90.
[7] K. Yunus, H. Zelaya De La Parra, M. Reza, Distribution Grid Impact
of Plug-In Electric Vehicles Charging at Fast Charging Stations Using
Stochastic Charging Model, European Power Elctronics Conference,
Birmingham, England, 2011.
[8] E. Pouresmaeil, D. Montesinos-Miracle,o. Gomis-Bellmunt, Control
scheme of a three level H-bridge converter for interfacing between
renewable energy resources and the AC grid, European Power
Elctronics Conference, Birmingham, England, 2011.
[9] Y. Chen, A. Oudalov, J. S. Wang, "Integration of Electric Vehicle
Charging System into Distribution Network," the 8th International
Conference Power Electronics, Jeju, Korea, 2011.
[10] Y. Chen, B. Su, J. S. Wang. "Integration of Electric Vehicles into Grid
with Bi-directional Power Conversion Technology," IET Annual Power
Symposium, HongKong, China, 2011.
[11] IEEE PES Distribution System Analysis Subcommittee. IEEE 37 Node
Test Feeder. http://ewh.ieee.org/soc/pes/dsacom/testfeeders/index.html.
[12] Technical Specifications of Quick Charger for the Electric Vehicle.
CHAdeMO Association, April 2010.
[13] Connection set for conductive charging of electric vehicles - Part3: DC
charging coupler. Draft national standard in China, December 2011.
