Title

Author(s)

Citation

Issued Date

URL

Rights

Frequency regulation for a power system with wind power and

battery energy storage

Liang, L; Zhong, J; Jiao, Z

The 2012 IEEE International Conference on Power System
Technology (POWERCON 2012), Auckland, New Zealand, 30
October-2 November 2012. In Conference Proceedings, 2012, p.
1-6
2012

http://hdl.handle.net/10722/189865

International Conference on Power Systems Technology

Proceedings. Copyright IEEE.

Frequency Regulation for a Power System

with Wind Power and Battery Energy Storage
Liang Liang, Student Member, IEEE, Jin Zhong, Senior Member, IEEE, and Zaibin Jiao, Member,
IEEE

Abstract-With the increase of wind power penetration to a

notable level in power systems, the requirement on frequency
regulation services has increased accordingly in recent decades.
Due

to

the

conventional

limited

ramp

generators,

rate

simp ly

and

economic

increase

the

factors

of

capacity

of

conventional generators may not be an effective solution for

providing

frequency

regulation

services

quickly.

Large-scale

battery energy storage system has been applied as a promising

solution

for

frequency

control

in

some

renewable

energy

integration p rojects. In this paper, a control method is proposed

for

battery

energy

storage

system

to

provide

frequency

regulation service. The control strategy to the energy storage

system

is

indicated

in

the

classical

two-area

AGC control

diagram. The effects of the proposed battery control strategy for

power system frequency control have been simulated and tested.
The simulation results show that battery systems could be used as
a tool in quick frequency control. The results also show the
possibility of introducing battery energy storage system into real
wind power generation projects.

Index Terms-AGC, battery energy storage system, frequency

regulation, optimal sizing, and renewable energy integration

I.

INTRODUCTION

UE to the "zero emission" characteristic of wind power

generation, it has been regarded as one of the solutions to
the global energy issues [1-5]. With the fast development
of wind power generation technologies, the rated power of a
single turbine has increased to 10MW in 2011. It is possible
that in the near future rated power of an off-shore wind farm
could exceed IGW, which is similar to the largest
conventional generators. In this case, the power output
fluctuation of such a large wind fann may cause serious
frequency stability problem to the system. Consequently,
negative effects of wind energy system could not be neglected
as previously. The connection between wind energy and
power grid should be more flexible and stable.
Renewable energy integration issues have been widely
discussed from various aspects. A system-wide approach is
proposed for distributed generations based on the new IEC
communication standard in [6]. In [7], a new control strategy

This work was partly supported by grants from the Research Grant
Council of Hong Kong (Project No. HKUIO/CRF/IO), HKU Strategic
Research Theme and the University Development Funding (Initiative on
Clean Energy & Environment), and the State Key Laboratory of Power
Systems of Tsinghua University,Beijing,China.
Liang Liang, Jin Zhong and Zaibin Jiao are with the Department of
Electrical and Electronic Engineering, The University of Hong Kong, Hong
Kong (e-mails:lliang@eee.hku.hk;jzhong@eee.hku.hk;zbjiao@eee.hku.hk).

for DFIG system is proposed to reduce the reserve

requirement for power system. These works could optimize
the allocation of system reserves. However, with the fast
increasing of wind power capacity, more conventional
generators need to be operated under frequency regulation
modes. New technologies are expected to carry on the task of
frequency regulation task for power grid.
Battery energy storage systems could inject or with draw
energy from the power grid through power conversion systems
in seconds. The rated power and rated capacity of battery
energy storage have reached hundred MWh levels with recent
technology development. The large number of electrical
vehicles in the future could provide distributed energy
storages for power systems. It was proposed that battery
energy storage system could be used for power system
frequency regulation. A study on frequency regulation in an
isolated power system has been carried out in [8]. The work
supports the idea that battery energy storage system could be
applied for frequency regulation. An optimization problem for
EV taking part in the frequency regulation service has been
studied in [9]. An optimal control method of EV was proposed
in this paper to provide frequency regulation service in order
to obtain the maximal revenue. An optimal dispatching
method for battery energy storage was studied in [10, 11]. The
work focused on how to reduce power fluctuations of the wind
farm outputs. In [12, 13], the battery energy storage system
has been introduced into the control loop of AGC system. The
dead band effect and generation rated constrains are
considered. The work is focused on the application of energy
storage system in the load frequency control.
In this paper, the application of battery for frequency
regulation is studied. A control strategy is proposed for the
battery energy storage system for the frequency control
purpose. The results are obtained for different parameters
based on the proposed method.
The organization of this paper is as follows. In Section II,
the control diagram of AGC and battery energy storage system
is presented. The historical data of wind power generation is
provided in Section III. The simulation results of frequency
regulation is presented and analyzed in section IV. Section V
concludes.
II. AGC CONTROL SYSTEM WITH BATTERY ENERGY STORAGE
The classical two-area AGC control system with tie-line
control has been developed and discussed since decades ago.
The frequency control system for a multi-area power system
978-1-4673-2868-5/12/$31.00 2012 IEEE

was proposed in [14]. The method has been widely applied in

the modem power system. In [15], the dynamic models of two
major turbines, steam turbine and hydro turbine, have been
studied. The simulation accuracy of frequency regulation
could be improved by applying dynamic models of generation
turbines. With widely applications of lead-acid batteries in
various industries, the possibility of introducing battery energy
storage into power industry has been discussed [12]. Due to
the limitation of present battery technologies, the application
area of battery discussed in [12] was limited in load frequency
control.
In this paper, we propose a control diagram to include
battery energy storage system in power system frequency
control. The modified diagram of classical two-area AGC
control with energy storage system is shown in Fig.l. Symbols
;\.. PI and;\.. P2 represent the power fluctuation, which could be
caused by renewable energy generation or load fluctuations.
Symbol;\.. Ps represents the output power from energy storage
system. Renewable energy generation could be considered in
Area 1 and Area 2 through;\.. Pland;\.. P2
Based on previous works, the proposed control system of
battery energy storage system is shown in Fig. 2. The control
system includes three loops. 1) Item

Ks
P

T ChargeS+ 1

to be 50%. 2) Symbol Ks represents the feedback gain of the

frequency fluctuation in the power grid. Item ;\.. PCaW ;\.. P2
represents the compensated power, which is caused by ramp
rate limitations of generators. 3) Item

1 +sTConvl

represents

the effect of the 1 siorder delay of DC-AC power converter in

the battery energy storage system.

represents

an auto balancing charging loop for the battery energy storage

system. This loop could control the remaining energy of the
energy storage around in the initial state, which is normally set

1 + sTConvl

Fig. 2 Control Diagram of Battery Energy Storage System

III.

OPERATION DATA OF WIND POWER GENERATION

To study frequency regulation issues in a system with

large capacity of wind power, the historical wind power
generation data, which is shown in Fig. 3, is used.
It is shown in the figure that the wind farm output has a
large power fluctuation. To study the impact of wind power

BI

-!
-

RI

..
...

.I

Kll
S

Peall

p
,
,
,
-y +L
I

1
l+sTgl

-----.

1
1 +sT[)

Ms

+.

..

Kpi

1 +sTpl

r;

,Ir

Al2

-!

r--

,Ir

.I

AI2

..

K1 2
s

-)L
I

B2

1 +STg2

1 +STT2

Fig. I Control Diagram of Battery Energy Storage System

+t-jll

..

Kp2
1 +sTp2

r--

on power grid operation at different wind power penetration

levels, the wind power profile in Fig. 3 is uniformed into
per unit (p.u.) values.
1

nll1

0.8

,;
2 0.6

o
Cl.

00.4

I
10

0.2

I
10

12
14
Time (Day)

Fig. 6 Frequency deviations with energy storage system

16

lL

18

20

22

24

Fig. 3 The historical data of wind fann

IV.

12

Time (Days)

The output power of battery energy storage system is

shown in Fig. 7. The maximum output power of energy
storage is less than 0.8p.u..The capacity of battery energy
storage system provided in Fig. 8 shows that the maximum
capacity of energy storage system is less than O.4p.u.hour.

FREQUENCY CONTROL CONSIDERING BATTERY ENERGY

STORAGE AND WIND POWER FLUCTUATION

A. Case 1

In Case 1, we will study the AGC system with wind

power generation and battery energy storages located in one
of the two areas.
Based on the block diagram described in Fig. 2, AGC
with wind generation system in one area is shown in Fig. 4.
The wind power signal is added at L'. P I. The control system
of the battery energy storage system is the same as in Fig. 2.
10

12

Time (Days)

Fig. 7 Output power of battery energy storage system

,

Fig. 4 The control diagram of AGC with wind power in one area

Applying the historical wind power data shown in Fig. 3,

the simulation results of frequency regulations are shown in
Fig. 5 and Fig. 6.

.1.

'I

10

12

Time (Days)

Fig. 8 Capacity of battery energy storage system

The simulation results of Fig. 7 and Fig. 8 show that with

a small capacity, a fast energy storage system could greatly
reduce frequency deviations caused by wind power
generations located in the same AGC control area.
10

12

Time (Days)

Fig. 5 Frequency deviations without energy storage

It is found in Fig. 5 that the highest frequency deviation

without battery energy system is more than 1.5 Hz during
the period of 24 days. In Fig. 6, it shows that the fast
response characteristic of battery energy storage system
could reduce the maximum frequency deviation to less than
0.12Hz, which is less than 10% of the deviation without
energy storage.

B. Case 2

In Case 2, we will study the scenario that wind power

generations are located at different AGC control areas as
battery energy storages.
The block diagram of the control system is shown in Fig.
1. Wind power generation is added in Area 2 at the point of
L'. P2in Fig. 1. The energy storage system is located in Area 1.
The simulation results are shown in Fig. 9 to Fig. 12.

4
"
,.
"

':jld,, II

",11' "r

"'
"'

"

I'

11,1

11K

11'

"

Il.l

If

"
Tm.{D>Iys)

"

"

"' ,

"

"

"

"

a) Result of Area 1
b)Result of Area 2
Fig. 9 Frequency deviations without energy storage

""
""
.

.... ".

'

"

" ,.
-.

., r

rl

j,

I ffiN.u

.Ii

u. !.."

"

"

"

"

"

12

Time (Days)

Fig. 12 Capacity of Energy Storage System in two area system

C. Case 3

,,

..

10

-<l"

"

a) Result of Area 1
b)Result of Area 2
Fig. I OFrequency deviations with energy storage system

Fig. 9 shows the frequency fluctuations of two areas

without battery energy storages, and wind power is in Area
2. The highest frequency deviations of two areas have
exceeded 0.6Hz and 0.8Hz, respectively. This is lower than
the frequency deviation in Case 1. The tie line, which
connects two areas, could provide extra power for frequency
regulation from the other area. Fig. 10 shows the simulation
result with battery energy storage system installed in Area 1.
The frequency deviation is to less than 0.02 Hz and 0.15 Hz,
respectively. The fast response characteristic of energy
storage system could reduce power deviations in both areas.

In Case 3, we will study the scenario that wind power

generations are located in the same control area as battery
energy storages.
The energy storage and wind power are located in Area 1 in
Fig. 1. The results of frequency fluctuations without battery
energy storage system are the same as Fig. 9, which just
switches the sequence of Area 1 and Area 2.
..

..

.
'

,.

..

. "'

..

...01 ".L

">,,. '''.

""

JdI"

.
. ! "I

L I:""

"

""

'

, .liIiiI

II"

""

.al

'''. '"

..I

'

.
. ..Iii
,,r"

L ""

1'1 If

'"
"

""

"

"

"

"

"

"

a) Result of Area 1
b)
Result of Area 2
Fig. 13Frequency deviations with energy storage system

"'"1---t-....-t-...l- . -;\",.. ---7.--;';- .. .,. .--;;-Io .--!

. -.cTIT\O(D.)

..,.

,.
"""""Po.... 'OfB.Ul)'Er>t'l1YSto'.... S'.lOm(pU)

a) Output Power in Time domain b) Distribution of Output Power

Fig. II The output power of energy storage system

The power output of the energy storage in Area 1 is

shown in Fig. 11. The Fig. I I(a) shows that the maximum
output power of the energy storage system under proposed
control method is less than 0.7p.u.. The Fig. I I(b) shows the
distribution of output power. The deviation is distributed
between 0.2 p.u., which is possible to be implemented
with modem battery technologies.
The capacity requirement of the energy storage system is
shown in Fig. 12. The required capacity of battery energy
storage system is less than 0.06p.u. ho in total. This means
that if the energy storage system could maintain its output
power at rated power for 4 minutes, the system could satisfy
the capacity requirement for frequency regulation in a two
area system during a time period of 24 days.
ur

The results in Fig. 13 show that the battery energy storage

system could provide a better compensation result if the
battery energy could be placed at the same area with wind
power generation system. The frequency fluctuations in both
areas could be controlled under 0.05Hz, which equals to half
of the maximum frequency deviation in Fig. 10.
'"
'"
"

": lui

'" I"" "r'

"'

lil ,.. J
\1 ,.

lI:

"

'"

".
"

..
"
Tm.(D)

"

..

..,.
"'
0
"
"
O'-'l>'-'l'ofB_ryE..,.'IlyS",'.... &;._(pu,1

a) Output Power in Time domain

b) Distribution of Output Power
Fig. 14 The output power of energy storage system

The result of battery energy storage system output is shown

in Fig. 14, which is similar with the results shown in Fig. 11.
The maximum output power has reached to 0.9 p.u., while the
distribution of output power is almost the same as the result
shown in Fig. 11.

VI. ApPENDIX

The nominal system parameters are:

Tn=T T2=0.3s
Tg,=Tg2=0.08s
Tp,=Tp2=24.0096s
T,=0.545s
TCharge=100s
KpB=-0.2
A'2=-1
R,=R2=2.4
KII=K'2=0.1
B,=B2=0.425
Kp,=KP2=lID,=1/0.00833=120.048
Ramp Limitation of generators: 0.001 p. u. /s
In Case 1, KB=5;
In Case 2, KB=150;
In Case 3, KB=15.

.I
rr

j
I'

10

12

Time (Days)

Fig. 15 Capacity of Energy Storage System in two area system

The result of capacity requirement of battery energy storage

system is shown in Fig. 15. The battery storage system locates
in the same area as the wind power generation system. So the
energy storage system could compensate much more power
fluctuation than other cases. As a result, the required capacity
is larger than the result shown in Fig. 12.
V.

CONCLUSIONS

In order to reduce the frequency fluctuation, which is

caused by wind power generations in power systems, a control
method is proposed in this paper for battery energy storage
system serving for frequency control. Three cases with
different system configurations, including single area system,
two area system with battery energy storage and wind power
generation located in different areas, and two areas system
with battery energy storage and wind power generation
located in the same area, have been studied with the proposed
control method.
The results support the idea that battery energy storage is
suitable for providing frequency regulation services to
compensate the fluctuation caused by wind power generations.
Battery energy storages could significantly reduce frequency
fluctuations in single-area and two-area power systems. In
some cases, the frequency deviation could be reduced to less
than 10% of the original one. As shown in the simulation
results, the requirement to the battery energy storage system is
possible to be realized using present battery manufacturing
technologies.
To reduce the interactive charging/discharging fluctuations
between conventional generators and battery storage systems,
if the battery energy storage and wind power generation are
located in the same area, the feedback gain KB should be
reduced.
A 24-day power output profile of wind power generation
has been applied to verify the effects of battery energy storage
system. The time interval for the data is 10 seconds. Although
the applied data is only for 24 days, by implementing an auto
balancing charging/discharging control loop to the battery
energy storage system, the proposed control method of battery
energy storage could be applied in a real-time control system
for a longer time period.

VII.
[I]

[2]
[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[II]

[12]

[13]

[14]

REFERENCES

W. J. B. Bunker and X. Ming, "Dependence of wind energy on electric

utility in the U.S," in Sustainable Systems and Technology
(ISSST)Symposium.,2011,pp. 1-5.
D. Zhou, "Application of CDM in China wind power generation," in
Proc. 2010 IEEE Electricity Distribution (CICED) Conj,pp. 1-5.
B. R. Singh and O. Singh, "21st Century challenges of clean energy and
global warming-can energy storage systems meet these issues?," in
Proc. 2010 IEEEThermalissues in Emerging Technologies Theory and
Applications (ThETA) Conj, pp. 323-329.
H. Hocheng, C. S. Tan,C. A. Lin, C. H. Lai, M. C. Yip,C. C. Hu, and
C. N. Liao, "Advanced energy research of College of Engineering at
National Tsing Hua University," in Proc. 2010 IEEEEnergy and
Sustainable Development: Issues and Strategies (ESD) Con!, pp. 1-5.
E. Ela, B. Kirby, E. Lannoye, M. Milligan, D. Flynn, B. Zavadil, and
M. O'Malley, "Evolution of operating reserve determination in wind
power integration studies," inProc. 2010 lEEEPower and Energy
Society General Meeting Conj, pp. 1-8.
A. Timbus, M. Larsson, and C. Yuen, "Active Management of
Distributed Energy Resources Using Standardized Communications
and Modem Information Technologies," IEEE Trans. Industrial
Electronics,vol. 56,pp. 4029-4037, 2009.
c.-c. Le-Ren, H. Chih-Min, and Y. Yao-Ching, "Dynamic Reserve
Allocation for System Contingency by DFIG Wind Farms," IEEE
Trans. Power Systems, vol. 23,pp. 729-736, 2008.
D. Kottick, M. Blau, and D. Edelstein, "Battery energy storage for
frequency regulation in an island power system," IEEE Trans. Energy
ConverSion, vol. 8,pp. 455-459, 1993.
H. Sekyung, H. Soohee, and K. Sezaki, "Development of an Optimal
Vehicle-to-Grid Aggregator for Frequency Regulation," IEEE Trans.
Smart Grid, vol. 1,pp. 65-72,2010.
S. Teleke, M. E. Baran, S. Bhattacharya, and A. Q. Huang, "Optimal
Control of Battery Energy Storage for Wind Farm Dispatching," IEEE
Trans. Energy Conversion,vol. 25,pp. 787-794,2010.
S. Teleke, M. E. Baran, A. Q. Huang, S. Bhattacharya, and L.
Anderson, "Control Strategies for Battery Energy Storage for Wind
Farm Dispatching," IEEE Trans. Energy ConverSion, vol. 24, pp. 725732,2009.
L. Chun-Feng, L. Chun-Chang, and W. Chi-Jui, "Effect of battery
energy storage system on load frequency control considering governor
deadband and generation rate constraint," IEEE Trans. Energy
Conversion,vol. 10,pp. 555-561,1995.
W.-c. Chan and Y.-Y. Hsu, "Automatic generation control of
interconnected power systems using variable-structure controllers," lEE
Proc. Generation, Transmission and Distribution, vol. 128, pp. 269279,1981.
O. I. Elgerd and C. E. Fosha, "Optimum Megawatt-Frequency Control
of Multiarea Electric Energy Systems," IEEE Trans. Power Apparatus
and Systems, vol. PAS-89,pp. 556-563, 1970.

6
[15]

I. C. Report, "Dynamic Models for Steam and Hydro Turbines in

Power System Studies," iEEE Trans. Power Apparatus and Systems,
vol. PAS-92,pp. 1904-1915, 1973.