IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY, VOL. 22, NO.

3, JUNE 2012 5701904

Compensation for the Power Fluctuation of the

Large Scale Wind Farm Using Hybrid Energy
Storage Applications
Hansang Lee, Byoung Yoon Shin, Sangchul Han, Seyong Jung, Byungjun Park, and
Gilsoo Jang, Senior Member, IEEE

Abstract—This paper proposes an application of supercon- sources in terms of efficiency and economical aspects [1]–[4].
ducting flywheel energy storages (SFESs) to compensate the The scale of wind farms have been increasing and large-scale
power fluctuation of the large scale wind farm. Based on the wind farms have already been constructed and under operation
global interest against global warming, the power capacity of
the renewable generation, especially wind generation, has been and additional sites is planned.
increased steeply. However, since wind generations depend on However, wind power has intermittent output characteristics,
the natural wind speed completely, the power output cannot be which makes it difficult to maintain stable outputs, thereby
controlled. The power fluctuation caused by the non-controllable requiring many issues to be addressed when the large-scale
output characteristic may create voltage problem for local system wind farms are linked to the system [5]. Since the fundamental
and frequency problem for whole power system. To solve those
problems, the hybrid application of the large-capacity battery problem caused by the connection of wind power is the in-
energy storage system (BESS) and the high-speed supercon- termittence in the power output, technology developed for
ducting flywheel energy storage system (SFES) are considered in stabilizing the wind power output will be beneficial for the
Heangwon wind farm in Cheju Island in Korea. Through the case dissemination of wind power.
studies based on the site-measured output data, the optimal power Measures to suppress the effects on the grid owing to the
and energy capacity of the BESSs and SFES are figured out.
output change are being studied from various aspects, out of
Index Terms—BESS, hybrid compensation, power fluctuation, which utilizing the energy storage technology is available. The
SFES, wind generation.
flywheel, the object of the study in this paper, is an energy
storage device that stores mechanical energy utilizing the
I. INTRODUCTION rotational inertia of the rotor, where superconducting bearings
are being applied to eliminate the thermal losses due to the
friction in the pivot bearings [6]. The main advantage of these

D UE to the increasing interest in the smart grid which fo-

cuses on greenhouse reductions, many efforts have been
made to establish the smart grid on a national scale, including
flywheel storage devices is that the response rate and efficiency
is relatively high. In terms of electrical characteristics, the
SFES(Superconductor Flywheel Energy Storage), depending
from the domestic industries and academia. In South Korea, on the design of the rotor, has an advantage of implementing a
where the greenhouse gas emissions are 1.6 times the average high storage capacity since the rotating mass can be increased
level of OECD countries, the necessity for the reduction of for the same mass, but also has the disadvantage that the
a significant amount of greenhouse gases has been required. instantaneous output is not very high since it utilizes devices
Increasing the utilization of renewable energy is being imple- with permanent magnet in the rotor to eliminate the loss due to
mented for the reduction of greenhouse gases where the wind the magnetic coupling in the device [7]–[9].
power is being considered as the appropriate renewable energy This paper proposes a measure to compensate the output of
Cheju Island’s Heangwon wind farm by utilizing a super- con-
ducting flywheel, which has a quick response and low power
Manuscript received September 13, 2011; accepted December 05, 2011. Date
of publication December 21, 2011; date of current version May 24, 2012. This
capacity, and a battery energy storage system, which has a com-
work was supported in part by a National Research Foundation of Korea Grant paratively lower response capability and high power capacity
funded by the Korean Government (20110018632) and in part by Korea Institute when compared to the flywheel. The paper is constituted as fol-
of Energy Technology evaluation and planning.
H. Lee was with the School of Electrical Engineering, Korea University,
lows: Chapter II gives a description of the Heangwon wind farm,
Seoul 136–713, Korea. He is now with the School of Railway and Elec- Chapter III describes the SFES and BESS models, Chapter IV
trical Engineering, Kyungil University, Gyeongsan 712–701, Korea (e-mail: addresses the case studies and finally Chapter V includes the
hsang80@korea.ac.kr).
B. Y. Shin and G. Jang are with the School of Electrical Engineering,
analysis, conclusions and recommendations.
Korea University, Seoul 136-713, Korea (e-mail: shinby@korea.ac.kr;
gjang@korea.ac.kr). II. HEANGWON WIND FARM IN CHEJU ISLAND
S. Han, S. Jung, and B. Park are with the Korea Electric Power Re-
search Institute, Daejon 305-760, Korea (e-mail: schan@kepri.re.kr; Heangwon wind farm is located in Cheju Island with a ca-
shammon@kepri.re.kr; hampstead@kepri.re.kr).
pacity of 9.795 MW and a total of 15 units as shown in Table I.
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. Owing to Cheju Island’s favorable wind resources, a relatively
Digital Object Identifier 10.1109/TASC.2011.2180881 larger proportion of wind power is being installed. However,

1051-8223/$26.00 © 2011 IEEE

5701904 IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY, VOL. 22, NO. 3, JUNE 2012

TABLE I
GENERATOR CONFIGURATION OF HEANGWON WIND FARM

Fig. 4. Compensating operation of BESS.

Fig. 1. Daily output variability of Heangwon wind farm.

Fig. 5. Compensating operation of SFES.

BESS can compensate large fluctuations whereas the SFES can

respond to rapid fluctuations. This paper proposes a hybrid com-
pensation method that combines the advantages of each storage
device to effectively compensate the intermittent wind power
output.
Fig. 2. Battery model and controller set.
A. BESS
In spite of the low response capability, BESS is an energy
storage devices can ensure a high reliability of energy supply
and achieve a high storage capacity with a relatively low instal-
lation cost.
As shown in Fig. 2, the battery is connected to the system
through an AC-DC inverter and a DC-DC converter for flexible
charging/discharging characteristics. The AC-DC inverter plays
the role of the active rectifier, whereas the DC-DC operates as
Fig. 3. Grid connection diagram of 900 kW PMSM/G. the charger.

B. SFES
since the wind is not constant and doesn’t maintain high wind
speeds, there are very high fluctuations in the power generation The superconducting flywheel utilizes the permanent magnet
as shown in Fig. 1. From the perspective of the operator, these synchronous motor/generator. As mentioned earlier, due to the
changes in the output of the wind farms makes it difficult to usage of superconducting bearings, the PMSM/G model, which
maintain stable operation conditions which can cause problems has a friction torque of 0, should be utilized instead of the gen-
with local voltage and frequency, especially in areas with a low eral synchronous machine which generates heat due to the mag-
electric inertia, thereby requiring the compensation of the power netic coupling of the rotor and stator. Fig. 3 shows the grid con-
output utilizing energy storage devices. nection diagram of the 900 kW PMSM/G model.

III. STORAGE SYSTEM MODEL IV. CASE STUDIES

BESS (Battery Energy Storage System) and SFES are being A hybrid connection compensation method, where the char-
utilized in this paper in order to compensate the wind farm’s acteristics of BESS that large amount of fluctuations can be
output variability. Generally, since BESS has slower response compensated for a relatively lower cost despite the compara-
characteristics and a larger capacity when compared to SFES, tively slower response speed and the high response speed of
LEE et al.: POWER FLUCTUATION OF LARGE SCALE WIND FARM USING HYBRID ENERGY STORAGE APPS 5701904

Fig. 6. Cumulative energy in the BESS.

Fig. 7. Cumulative energy in the SFES.

TABLE II
SPECIFICATIONS OF BESS AND SFES FOR FULL COMPENSATION

TABLE III
SIMULATION SCENARIOS FOR PARTIAL COMPENSATION

Assumption 1: same specifications of BESS with the full compensation case

(3.302 MW & 5.172 MWh)
Assumption 2: power capacity of SFES is limited in 0.9 MW.

SFES are being combined for a more effective compensation of

the intermittent characteristics of wind power, is being proposed
and case studies were performed to show the effectiveness of the
following method.
The simulations were carried out to show the changes in the
output due to the implementation of BESS and SFES, based on
actual output data from the 9.8 MW Cheju Island Heangwon Fig. 8. Remained output fluctuation for each storage capacity of SFES.
wind farm.
of the system frequency and the voltage of the adjacent areas.
A. Full Compensation To solve these problems, to supply the system a fixed output of
The intermittent generation of the Heangwon wind farm with 1 MW by utilizing the full compensation scenario of BESS and
a fluctuation range of 4.5 MW is being shown in Fig. 1. This SFES, the instantaneous and storage capacity of each storage
change in the wind power output can result in the fluctuation device was calculated.
5701904 IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY, VOL. 22, NO. 3, JUNE 2012

TABLE IV the increase of the SFES energy capacity. Also the difference
SIMULATION RESULTS FOR PARTIAL COMPENSATION SCENARIOS of the maximum and minimum power output of wind farm is
decreased significantly. This means that when there is sufficient
storage capacity for SFES, the minute output fluctuations can
be rectified thereby facilitating the connections of wind farms
to the grid.

V. CONCLUSION
In this paper, the wind farm output stabilization measures
were simulated by utilizing the energy storage characteristics
of SFES and BESS. In order to stabilize the wind farm output,
a hybrid storage device scheme has been proposed where the
energy storage device capacity and response rate has been con-
As shown in Figs. 4 and 5, BESS compensates the large fluc- sidered to combine the high capacity battery and flywheel de-
tuations and SFES compensates for the small and fast fluctua- vice. Simulation results show that instead of only using BESS,
tion for the reference output (setpoint) of the Heangwon wind adequately mixing SFES with BESS is a more efficient and ef-
farm. BESS uses its large storage capacity to compensate the fective method for stabilizing the output of wind farms. How-
wind power output but doesn’t have the fast response capability ever, above a certain level, increasing the SFES capacity didn’t
to follow rapidly changing output thereby requiring additional show much improvement in the compensation. Therefore, fu-
compensation of SFES. The storage capacity of BESS and SFES ture studies will be required from an economic point of view
is being calculated by the difference of the maximum and min- for the calculation of the SFES capacity and the optimal combi-
imum values of the accumulated energy from the full compen- nation of the energy storage devices.
sation scenario as shown in Figs. 6 and 7. Table II represents the
specifications of BESS and SFES for the full compensation of
the Heangwon wind farm. REFERENCES
[1] P. J. Luicks, E. D. Delarue, and W. D. D’haeseleer, “Effect of the gen-
B. Partial Compensation eration mix on wind power introduction,” IET Renew. Power Gener.,
vol. 3, no. 3, pp. 267–278, 2009.
A case study is conducted to determine the impact of the com- [2] F. Bouffard and F. D. Galiana, “Stochastic security for operations plan-
nings with significant wind power generation,” IEEE Trans. Power
mitted amount of SFES by setting SFES and BESS with the fol- Syst., vol. 23, no. 2, pp. 306–316, 2008.
lowing assumptions shown in Table III. [3] J. E. Bridges, “Wind power energy storage for in situ shale oil recovery
Fig. 8(a) shows the output of the Heangwon wind farm with with minimal emissions,” IEEE Trans. Energy Convers., vol. 22,
no. 1, pp. 103–109, 2007.
only BESSs’ compensation. It can be observed that the output [4] G. Byeon, I. K. Park, and G. Jang, “Modeling and control of a
is maintained only for the small variations around the 1 MW Doubly-Fed Induction Generator (DFIG) wind power generation
setpoint for the actual output of Heangwon wind farm and that system for real-time simulations,” Journal of Electrical Engineering
& Technology, vol. 5, no. 1, pp. 61–69, 2010.
the small and fast fluctuations are noticeably removed with the [5] Y.-Y. Hong and J.-L. Pen, “Optimal VAR planning considering inter-
increase in the SFES capacity. mittent wind power using Markov model and quantum evolutionary
Figs. 8(b)–8(f) shows the Heangwon wind farm output due algorithm,” IEEE Trans. Power Deliv., vol. 25, no. 4, pp. 2987–2996,
2010.
to the operation of various storage capacities of SFES. Since [6] J. Lee, S. Jeong, Y. H. Han, and B. J. Park, “Concept of cold energy
the energy capacity of SFES is not enough high, it can be fully storage for superconducting flywheel energy storage system,” IEEE
charged or discharged for a short time. This means that the Trans. Appl. Supercond., vol. 21, no. 3, pt. 2, pp. 2221–2224, 2011.
[7] J.-P. Lee, N.-H. Jeong, Y.-H. Han, S.-C. Han, S. Jung, B.-J. Park, and
timely charging or discharging cannot be achieved. The more T.-H. Sung, “Assessment of the energy loss for SFES with rotational
the SFES energy capacity is increased the more minute changes core type PMSM/G,” IEEE Trans. Appl. Supercond., vol. 19, no. 3, pt.
are being compensated thereby maintaining the output at its set- 2, pp. 2087–2090, 2009.
[8] J.-P. Lee, S.-C. Han, Y.-H. Han, and T.-H. Sung, “Loss characteristics
point. Table IV shows the simulation results of each scenario. of SFES with amorphous core for PMSM,” IEEE Trans. Appl. Super-
The fluctuation reduction rate, which is the ratio of the cond., vol. 21, no. 3, pt. 2, pp. 1489–1492, 2011.
number of not-compensated samples for the entire measured [9] J.-P. Lee, B.-J. Park, Y.-H. Han, S. Jung, and T.-H. Sung, “Energy loss
by drag force of superconductor flywheel energy storage system with
samples, is obtained for each case and listed in Table IV. There permanent magnet rotor,” IEEE Trans. Magn., vol. 44, no. 11, pt. 2, pp.
is a significant reduction of output fluctuation according to 4397–4400, 2008.