Available via license: CC BY 4.0
Content may be subject to copyright.
Optimal control and management of a large-scale battery
energy storage system to mitigate fluctuation and intermittence
of renewable generations
Xiangjun LI
1
, Liangzhong YAO
1
, Dong HUI
1
Abstract Battery energy storage system (BESS) is one of
the effective technologies to deal with power fluctuation
and intermittence resulting from grid integration of large
renewable generations. In this paper, the system configu-
ration of a China’s national renewable generation demon-
stration project combining a large-scale BESS with wind
farm and photovoltaic (PV) power station, all coupled to a
power transmission system, is introduced, and the key
technologies including optimal control and management as
well as operational status of this BESS are presented.
Additionally, the technical benefits of such a large-scale
BESS in dealing with power fluctuation and intermittence
issues resulting from grid connection of large-scale
renewable generation, and for improvement of operation
characteristics of transmission grid, are discussed with
relevant case studies.
Keywords Battery energy storage systems, Renewable
generations, Power fluctuation, Battery energy
management system, Power control
1 Introduction
Renewable energy power generation has become an
important part for China’s power supply. By June 2016, the
grid’s wind-power capacity had been 124 GW and the
photovoltaic (PV) capacity had been 61 GW. The rapid
development and implementation of renewable power
generation pose great challenges to the operation, control,
and security of the Chinese power grid. Large-scale battery
energy storage system (BESS) can effectively compensate
the power fluctuations resulting from the grid connections
of wind and PV generations which are random and inter-
mittent in nature, and improve the grid friendliness for
wind and PV generation grid integration.
Large-scale BESS can participate in the operation as
either the power supply or the load when needed. Unlike
traditional power generation systems, BESS can act as a
rapid-response active and reactive power injection or
absorption device [1–8]. The BESS can be used to smooth
the power fluctuations of PV or wind power stations
[9–12]. Based on the existing researches and implementa-
tions of large-scale BESS worldwide, countries such as the
United States, Germany and Japan, have carried out more
than 200 demonstration projects. For example, redox flow
and sodium sulfur battery is one of the cutting edge tech-
nologies for renewable energy power generation applica-
tions in Japan [13–15]. There are also more applications of
lithium-ion BESS in the United States, such as in the fields
of renewable energy generations, distributed generations,
micro grids, etc. The American Xtreme Power, Duke
Energy, Altairnano, and AES Energy storage companies,
for example, have conducted researches on energy storage
technologies [16–18]. At present, existing applications of
large-scale lithium, sodium-sulfur or redox flow battery
have reached to tens of megawatts (MW) in power rating.
CrossCheck date: 21 September 2016
Received: 3 July 2016 / Accepted: 22 September 2016
The Author(s) 2016. This article is published with open access at
Springerlink.com
&Xiangjun LI
lixiangjun@epri.sgcc.com.cn
Liangzhong YAO
yaoliangzhong@epri.sgcc.com.cn
Dong HUI
huidong@epri.sgcc.com.cn
1
State Key Laboratory of Control and Operation of Renewable
Energy and Storage Systems, China Electric Power Research
Institute, Beijing 100192, China
123
J. Mod. Power Syst. Clean Energy
DOI 10.1007/s40565-016-0247-y
However, they are generally used only for wind energy
storage or solar energy storage respectively. Although the
MW power level of BESS is generally high, the MWh
capacity level is relatively low. For example, the BESS of
Japan Hokkaido wind farm incorporates a vanadium redox
flow BESS with the power capacity 4 MW/6 MWh and is
mainly for smoothing the wind power output fluctuations
[18–20]. The Japan Aomori Six Village energy storage
power station utilizes a sodium sulfur BESS with the power
capacity 34 MW, mainly for smoothing the wind power
fluctuations [18,21]. The Texas wind farm storage power
station uses an advanced lead-acid battery (36 MW/
9 MWh), principally for frequency regulation, energy
transfer and peak load shaving [18,22]. The West Virginia
Elkins wind farm energy storage power station incorporates
a lithium-ion battery (32 MW/8 MWh) which is for fre-
quency regulation and output climbing control [18,23].
In China, there are a number of large-scale BESS demon-
stration projects currently underway. For example, in Zhang-
bei, a large-scale BESS, which includes a 14 MW/63 MWh
lithium-ion BESS and a 2 MW/8 MWh vanadium redox flow
BESS, has been put into operation (flow BESS is still in the site
commissioning stage). It is part of a national wind, PV, storage
and transmission demonstration project. The purpose of this
project is to smooth the wind and PV power fluctuations and
trace the scheduled power outputs to grid. Further, Guodian
Longyuan Woniushi wind farm energy storage power station,
using total vanadium flow batteries (5 MW/10 MWh), is
adopted mainly to resolve wind-curtailment and brownout
issues arising at the Woniushi wind farm. The BESS of
Southern Power Grid Shenzhen Baoqing adopted the lithium-
ion battery (planned capacity is 10 MW and completed
capacity is 4 MW/16 MWh) to achieve peak load shaving,
frequency regulation, and voltage regulation, etc.
In this paper, the system configuration of a national
renewable generation demonstration project, combining a
large-scale BESS with wind farm and PV power station all
coupled to a power transmission system, is introduced, and
the key technologies and operational status of this BESS
are presented. Additionally, the technical benefits of such a
large-scale BESS in dealing with power fluctuation and
intermittence issues resulting from grid connection of
large-scale renewable generation, and for improvement of
operation characteristics of transmission grid, are discussed
with relevant case studies.
2 Large battery energy storage station
in Zhangbei
The Zhangbei energy storage power station is the largest
multi-type electrochemical energy storage station in China
so far. The topology of the 16 MW/71 MWh BESS in the
first stage of the Zhangbei national demonstration project is
shown in Fig. 1. As can be seen, the wind/PV/BESS hybrid
power generation system consists of a 100 MW wind farm,
a 40 MW PV power station, a 14 MW/63 MWh lithium-
ion BESS, a 2 MW/8 MWh redox flow BESS, and a power
grid. The wind farm, PV power station and BESS are
connected to the power grid through transformers. Specif-
ically, the 14 MW/63 MWh lithium-ion BESS includes
nine units (C001 to C009) in parallel (as shown in Table 1),
each connected to a 35 kV AC bus by means of a 380 V/
35 kV transformer unit. The topology of each lithium-ion
BESS connected to the power grid is shown in Fig. 2.As
indicated, each BESS contains multiple lithium-ion battery
energy storage units in parallel, each unit consisting of a
500 kW power converter system (PCS) and multiple
lithium-ion battery packs. Currently the large lithium-ion
electrochemical energy storage station contains 46 sets of
PCS and around 275000 lithium battery single cells. The
lithium-ion battery energy storage unit can be controlled by
using the PCS for management of start/stop and charging/
discharging functions, etc.
The redox flow BESS, meanwhile, includes two sets of
1 MW/4 MWh redox flow sub-BESS (C010 to C011) in
parallel, each connected to a 35 kV AC bus by means of a
380 V/35 kV transformer unit.
3 Key technologies for large battery energy
storage systems
The key BESS technologies includes system integration
and access, monitoring and control, energy management
and application, and other aspects.
3.1 Structure and function of supervisory control &
data acquisition system for BESS
The supervisory control and data acquisition (SCADA)
system is the core component of battery energy storage
power station, by which centralized access, real-time con-
trol and operation scheduling are achieved. More specifi-
cally, it is utilized to send information to and accept
information from BESS equipment such as local control
monitoring system and PCS, as well as to conduct real-time
monitoring and carry out control management functions.
The SCADA structure proposed in this paper is illus-
trated in Fig. 3. As can be seen, a hierarchical distributed
control scheme for BESS mainly includes three layers,
namely, the master station layer, the local control and
monitoring system (CMS) layer, and the device layer. The
master station layer includes servers, workstations, and
coordinated controllers. In the local CMS layer, there are
eleven local controllers, and each is used to manage four or
Xiangjun LI et al.
123
six PCS devices. In the device layer, there are PCS, battery
devices, and power distribution cabinets.
Between these layers, information transmission is
implemented via 100 Mbit/1000 Mbit fiber ring networks,
and is realized by means of a dual network communication
system combining a monitoring network with a control
network, as shown in Table 2. The functions of each layer
are shown as follows.
01
2MW
li-ion
BESS
14 MW/63 MWh lithium-ion BESS
35 kV AC bus
380 V AC
Wind farm
Transformer
PV power station
Transformer
Transformer
35 kV AC bus
2 MW/8 MWh redox-flow BESS
02
2MW
li-ion
BESS
03
2MW
li-ion
BESS
04
2MW
li-ion
BESS
05
2MW
li-ion
BESS
09
1MW
li-ion
BESS
08
1MW
li-ion
BESS
07
1MW
li-ion
BESS
06
1MW
li-ion
BESS
010
1MW
flow
BESS
011
1MW
flow
BESS
Fig. 1 Wind/PV/BESS hybrid power generation system
Table 1 System integration and components for BESS
Unit
name
System
integrator
Battery
supplier
Battery
management
system (BMS)
supplier
PCS
supplier
Total
capacity
(MWh)
Unit
capacity
(MWh)
Unit
maximum
power
(MW)
PCS rated
power
(kW)
PCS
amount
(set)
C001,
C002,
C003
CALB CALB PowerWise SIFANG 9 3 2 500 12 (=4 93)
C004 WANXIANG WANXIANG WANXIANG XJ 2 2 2 500 4 (=4 91)
C005,
C006
ATL ATL ATL SOARING 16 8 3 500 12 (=6 92)
C007,
C008,
C009
BYD BYD BYD BYD 36 12 3 500 18 (=6 93)
C010,
C011
Prudent
Energy
Prudent
Energy
Prudent Energy ABB 8 4 1 500 10 (=5 92)
Optimal control and management of a large-scale battery energy storage system to mitigate…
123
1) Master station layer
Master station layer is mainly responsible for coordi-
nated control and energy management, communication
management, data acquisition, data processing and man-
agement. It calculates power commands of transformer unit
stations by coordinated control and energy management
system. Some key real-time running state data and control
commands have been interactive between master station
layer and local CMS layer through control network by
using an Ethernet for plant automation (EPA) protocol
based strong real-time Ethernet.
Detail running data of each transformer unit, including
all single battery voltage, single battery temperature, and
detailed operation information etc, have been transmitted
from supplier local monitoring system to master station
layer through IEC60875-5-104 (IEC-104) protocol based
monitoring network.
The local controllers assign the target power to each
associated PCS, to control the power of each PCS
according to transformer-unit-based sub-BESS. The sub-
BESS is distinguished based on 380 V/35 kV transformer
unit.
Some key real-time running state data of BESS have
been uploaded to the remote dispatch station center based
on IEC-104 protocol.
PCS #1
Total control cabinet
Battery
cabinet #N
Battery
cabinet #1
PCS #L
Total control cabinet
Transformer unit
380 V AC bus
35 kV AC bus
Battery
cabinet #N
Battery
cabinet #1
Fig. 2 Topology for sub-BESS under transformer unit
#1
unit of
ESS
PCS ...
#2
unit of
ESS
#L
unit of
ESS
...
Sub-BESS based on a transformer unit Sub-BESS based on a transformer unit
#1
unit of
ESS
...
#2
unit of
ESS
#L
unit of
ESS
#1 Local controller #S Local controller
Supplier
local
monitoring
system
Coordinated controller
Master station layer
Remote CMS layer
Supplier
local
monitoring
system
...
... ...
...
...
...
...
PCS PCS PCSPCSPCS
Fig. 3 Hardware platform structure of SCADA
Table 2 Communication protocol between the parts
Layers Supplier local monitoring system Local controller Remote CMS layer
Master station layer IEC-104 EPA IEC-104
PCS layer CAN Modbus –
Xiangjun LI et al.
123
2) Local CMS layer
The local CMS consists of three parts: the local con-
trollers, the I/O stations, and the local monitoring system of
supplier. The local controllers are used to receive com-
mands from the coordination controller and, thus, to
achieve regional-level (380 V voltage level) coordination
control; the I/O stations convert the protocol of the con-
verter interface, collects data and sends control commands.
The local controllers monitors the BESS in real-time by
using the VxWorks embedded operating system. Addi-
tionally, at the local CMS layer, the PCS, battery and
distribution-system operational statuses are monitored in
real-time, and the upper-layer control instructions are
promptly send to each control PCS unit. Overall, this
hierarchical control system effectively guarantees the
control precision and stability of the storage system.
3) Device layer
The device layer contains a number of energy storage
systems. For instance, a 500 kW/2 MWh energy storage
system incorporates a 500 kW PCS, a 2 MWh energy
storage battery unit and some BMSs. The PCS is mainly
used to control the charge/discharge power and manage
protection functions. The BMS is mainly used to manage
the operation and control of the 2 MWh energy storage
battery. Its main functions include the analog signal mea-
surement, running battery system alarm, battery system
protection, self-diagnostics, battery-balanced management,
statistical storage, charge-discharge management, hyper-
tension management, thermal management, communica-
tion, insulation testing, and others. The typical topology of
the BMS is shown in Fig. 4.
3.2 Energy management system for large-scale
BESS
Firstly, the work specification of the energy manage-
ment system (EMS) for large-scale BESS has been
designed as shown in Fig. 5. In the charge status, power is
less than zero; and in the discharge status, power is greater
than zero. The PCS is used to achieve the power and
energy balance. The principles of the energy management
supervision, in this paper, have been proposed according to
the following objectives and constraints.
1) Objectives: To meet the real-time power requirement;
To ensure the energy balance and availability.
2) Constraints: The limit of allowable maximum charge/
discharge power; The limit of the charge/discharge
capacity.
In this paper, a two level EMS has been proposed. One
is main-EMS layer for transformer units in the station
control layer. The other is sub-EMS layer for PCS units in
the local CMS layer.
The main-EMS, according to current collection status
information of each transformer unit, calculates the power
commands in real-time based on BESSs’ total power
demand, so as to prevent excess charge or excess discharge
of the batteries. In accordance with the total power allo-
cation strategy for a transformer unit, the sub-EMS allo-
cates the transformer unit demand power to each PCS unit
according to the status of PCS unit. The status parameters
include the allowable maximum charge/discharge power of
the PCS unit, and the state of charge (SOC), etc. And then
real-time demand power is guaranteed to prevent over-
charge/discharge of the battery, ensuring the safety and
reliability of the transformer unit. The flow chart of the
proposed EMS algorithm is shown in Fig. 6and that is
described in detail in the following sections respectively.
Battery pack #1
CAN bus
BMU #1
BMU #2
BMU #3
BMU #4
BMU #N
BMS #1
Battery pack #2
Battery pack #3
Battery pack #4
Battery pack #N
Battery pack #1
CAN bus
BMU #1
BMU #2
BMU #3
BMU #4
BMU #N
BMS #M
Battery pack #2
Battery pack #3
Battery pack #4
Battery pack #N
CAN bus
Fig. 4 Schematic diagram of typical topological structure of battery
management system
#1 unit
of ESS
PCS ...
#L unit
of ESS
...
# 1 Sub-EMS layer
Main-EMS layer
Master station layer
Sub-BESS based on
a transformer unit
Sub-BESS based on
a transformer unit
#S Sub-EMS layer
...
...
PCS PCS PCS PCS PCS
#2 unit
of ESS
#1 unit
of ESS
#L unit
of ESS
#2 unit
of ESS
P1Q1P2Q2PLQL
PTU1QTU1PTUS QTUS
Pall-BESS Qall-BESS
P1Q1P2Q2PLQL
Fig. 5 EMS structure for BESS
Optimal control and management of a large-scale battery energy storage system to mitigate…
123
-
0?
all BESS
P>
ˆ?
disch
TUi TUi
PP>
?iS≤
N
Y
Y
N
N
End
Y
N
Y
()
1
ˆ
ˆ
disch
TUi TUi
SSEBllaiUT S
disch
TUi TUi
i
uP
PP
uP
−
=
=
Σ
1
k
kTUi
L
k
k
SOC
PP
SOC
=
=
Σ
ˆ?
disch
kk
PP>
N
Y
Y
N
()
1
ˆ
ˆ
disch
kk
iUTk L
disch
kk
k
uP
PP
uP
=
=
Σ
1kk=+
Y
N
0?
all-B ESS
P<
?kL≤
1i=
1k=
()
1
TUi TUi
SSEBllaiUT S
TUi TUi
i
uSOC
PP
uSOC
−
=
=
Σ
Calcul ate target power of eac h transformer unit
i
Calcu late ta rget p ower of each PC S unit
k
based on power of each transformer unit
i
Recalc ulate targ et power of each PC S unit
k
bas ed on po wer of eac h tran sfor mer unit
i
Recalculate target power of each
transformer unit
i
()
1
TUi TUi
SSEBllaiUT S
TUi TUi
i
uSOD
PP
uSOD
−
=
=
Σ
Calcu late target power of eac h trans former unit
i
ˆ?
ch
TUi TUi
PP>
N
Y
1i=
()
1
ˆ
ˆ
ch
TUi TUi
SSEBllaiUT S
ch
TUi TUi
i
uP
PP
uP
−
=
=
Σ
Recalc ulate targ et power of each
tran sfo rmer uni t i
where i=1,
…
,S
where k=1,
…
,L
i=1,
…
,S
1
k
kTUi
L
k
k
SOD
PP
SOD
=
=
Σ
Calcul ate ta rget powe r of each PCS un it
k
based o n power of each transformer unit
i
where
k=
1,
…
,
L
i=
1,
…
,
S
ˆ?
ch
kk
PP>
N
Y
Y
N
()
1
ˆ
ˆ
ch
kk
iUTk L
ch
kk
k
uP
PP
uP
=
=
Σ
1kk=+
?kL≤
1k=
Recalculate target power of each PCS unit k
based on power of each transformer unit i
where i=1,…,S
where i= 1,
…
,S
?iS≤
1ii=+
1ii=+
where i=1,
…
,S
where k=1,
…
,L
i=1,
…
,S
where k=1,
…
,L
i =1,
…
,S
Main-EMS layer:
Power a llocation ba sed on
Obtain total active power demand of BESS,
all BESS
P
−
Determi ne active power demand of each transf ormer uni t,
TUi
P
N
YY
0?
TUi
P>
0?
TUi
P<
Sub-EMS layer: Power allocation based on
Determi ne active power demand of each PCS unit k, P
k
N
where i=1,
…
,S
all BESS
P
−
TUi
P
Start
Fig. 6 Flow chart of the proposed EMS algorithm for active power control
Xiangjun LI et al.
123
3.2.1 Energy management strategy of main-EMS
Total active power demand of BESS, PallBESS , comes
from the master station layer as shown in Fig. 5. The target
power of each transformer unit i,PTUi , is calculated
according to the allowable charging and discharging power
and SOC. The purpose of this energy management step is
to regulate the SOC for each transformer unit at an
appropriate level during regulation while ensuring that the
operational constraints such as charge/discharge power and
battery SOC are not violated.
1) In the discharging status (when PallBESS [0): PTUi
is calculated by considering the SOC under each trans-
former unit i,SOCTUi and allowable maximum discharge
power,
^
Pdisch
TUi under each transformer unit i. Specific cal-
culation steps are as follows.
Firstly, the initial target power of each transformer unit
is calculated by using (1).
PTUi ¼uTUi SOCTUi
P
S
i¼1
uTUi SOCTUi
ðÞ
PallBESS ð1Þ
Secondly, if PTUi [0 and PTUi [
^
Pdisch
TUi ,PTUi calculated
by (1) is determined again by using (2) and (3) based on the
current allowable discharge power constraint for each
transformer unit.
PTUi ¼uTUi
^
Pdisch
TUi
P
S
i¼1
uTUi
^
Pdisch
TUi
PallBESS ð2Þ
^
Pdisch
TUi ¼X
L
k¼1
uk
^
Pdisch
kð3Þ
where PallBESS is the total active power demand of BESS;
TU is representing the transformer unit; uTUi is the start-
stop status under transformer unit i;PTUi is the active
power of transformer unit i;
^
Pdisch
TUi is the allowable maxi-
mum discharge power of transformer unit i;ukis the start-
stop status of PCS i;
^
Pdisch
kis the allowable maximum
discharge power of PCS k;Sis the total number of trans-
former unit; Lis the total number of PCS for each trans-
former unit.
2) In the charging status (when PallBESS \0): PTUi is
calculated considering the stage of discharge (SOD) under
each transformer unit i,SODTUi, and allowable maximum
charge power under each transformer unit i,
^
Pch
TUi. Specific
calculating steps are as follows. Firstly, the target power of
each transformer unit is calculated by using (4).
PTUi ¼uTUi SODTUi
P
S
i¼1
uTUi SODTUi
ðÞ
PallBESS ð4Þ
Secondly, if PTUi\0 and PTUi
[
^
Pch
TUi
,PTUi is
determined again by using (5) and (6) as follows based
on the current allowable charge power constraint for each
transformer unit.
PTUi ¼uTUi
^
Pch
TUi
P
S
i¼1
uTUi
^
Pch
TUi
PallBESS ð5Þ
^
Pch
TUi ¼X
L
k¼1
uk
^
Pch
kð6Þ
where
^
Pch
TUiis the allowable maximum charge power of
transformer unit i;
^
Pch
kis the allowable maximum charge
power of PCS k;Sis the total number of transformer unit; L
is the total number of PCS for each transformer unit.
3) In addition, in the discharging and charging status, the
uTUi and SOCTUi is determined as follows. SOCTUi is gen-
erally calculated by (7).
SOCTUi ¼P
L
k¼1
ukSOCk
P
L
k¼1
uk
ð7Þ
SODTUi ¼1SOCTUi ð8Þ
If the SOC deviation between the controlled PCS units
in the transformer unit is larger (this deviation can be
determined based on the actual operation requirements),
take the maximum SOC
k
or minimum SOC
k
as the SOC of
transformer unit iaccording to the charge or discharge
power needed.
3.2.2 Energy management strategy of sub-EMS
The target active power under each transformer unit,
PTUi, comes from the main-EMS layer as shown in Fig. 6.
The initial target power of each PCS is calculated using (9)
and (11) below, and is then modified based on the allow-
able maximum charge/discharge capacity. Specific calcu-
lation steps are as follows.
1) In the discharging status (when PTUi [0): Piis
calculated by the SOC of each PCS i,SOCiand allowable
maximum discharge power,
^
Pdisch
iof each PCS i. That is,
firstly, the initial target power of PCS iis calculated by (9).
Optimal control and management of a large-scale battery energy storage system to mitigate…
123
Pi¼SOCi
P
L
i¼1
SOCi
PTUi ð9Þ
Secondly, if Pi[0 and Pi[
^
Pdisch
i,Picalculated by (9)
is determined again by using (10) based on the current
allowable discharging power constraint for each PCS unit.
Pi¼ui
^
Pdisch
i
P
L
i¼1
ui
^
Pdisch
i
PTUi ð10Þ
where uiis the start-stop status of PCS i;SOCiis the SOC
of PCS i.
2) In the charging status (when PTUi \0): Piis calcu-
lated by the SOD of each PCS i,SODiand allowable
maximum charge power,
^
Pch
iof each PCS i. Firstly, the
initial target power of PCS iis calculated based on the SOD
of each PCS unit using (11) and (12) as follows.
Pi¼SODi
P
L
i¼1
SODi
PTUi ð11Þ
SODi¼1SOCið12Þ
where SODiis SOD of PCS i.
Secondly, if Pi\0 and Pi
jj
[
^
Pch
i
,Piis determined
again by using (13) based on the current allowable charge
power constraint of each PCS unit.
Pi¼ui
^
Pch
i
P
L
i¼1
ui
^
Pch
i
PTUi ð13Þ
3.2.3 Reactive power control strategy for BESS
The dynamic reactive power support function is one of
the important applications of large-scale BESS. Typically,
the storage control unit is in active power control mode.
However, depending on the voltage regulation require-
ments, PCS units can provide dynamic reactive power
support to the connected grid, in the form of reactive power
compensation. A multi-level reactive power control strat-
egy for BESS has been proposed based on maximum
allowable reactive power level of each PCS. That is, the
first-level control layer is the reactive power control
between various energy storage transformer units, the
second-level control layer is the reactive power control of
each storage unit inside the transformer between the PCSs.
The specific control method is explained below.
1) Reactive power control for each transformer unit
Total reactive power requirement of BESS, QallBESS ,
comes from the master station layer as shown in Fig. 6.
The purpose of this coordinated control step for reactive
power is to allocate the reactive power demand appropri-
ately based on the reactive power supply capacity of each
transformer unit. Then the reactive power under each
transformer unit i,QTUi, is calculated by allowable maxi-
mum reactive power under each transformer unit i,
^
QTUi as
shown in (14).
QTUi ¼uTUi
^
QTUi
P
L
i¼1
uTUi
^
QTUi
QallBESS ð14Þ
where
^
QTUi is calculated based on the maximum allowable
apparent power of PCS i.
Siand the current active power of
each PCS, Pi, are as shown in (15).
^
QTUi ¼X
S
i¼1ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
S2
iP2
i
qð15Þ
2) Reactive power control for each PCS unit
The real-time reactive power calculation method for
each PCS proceeds as follows, based on the reactive power
demand of each transformer unit.
The reactive power under each transformer unit, QTUi ;
comes from the main-EMS layer as shown in Fig. 6. The
reactive power of each PCS is calculated based on the
allowable maximum reactive power for PCS i,
^
Qi, are as
shown in (16).
Qi¼ui
^
Qi
P
L
i¼1
ui
^
Qi
QTUk ð16Þ
where
^
Qiis calculated by (17).
^
Qi¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
S2
iP2
i
qð17Þ
4 Operational status of large BESS in China
national demonstration project
4.1 Power generation tracking plan
The output power of wind farm in Zhangbei national
renewable generation demonstration station is limited by
regional dispatching system, but there is no output power
restriction on the PV power generation. When the output
power of wind farm is limited, the wind/PV/BESS hybrid
system can be operated in the ‘‘tracking dispatch schedule
output’’ mode.
Under conditions including strong winds, the wind farm
output power increases in Zhangbei area, which may
threaten the stability and security of the power grid. The
BESS adjusts the power output according to the dispatched
wind power generation schedule, so that the maximum
Xiangjun LI et al.
123
power is absorbed and wind power curtailment is reduced,
by implementing the ‘‘tracking power schedule output’’
mode at the specified time.
Figure 7shows the operation status for tracking power
generation plan of 80 MW. By implementing the ‘‘tracking
power schedule output’’ mode in the EMS, the deviation
between the actual wind/PV power and the plan is effec-
tively reduced, and the wind/PV/BESS hybrid system
output power is restricted within the planned power range,
thus meeting the ‘‘tracking power schedule output’’ mode’s
requirements. Figure 8shows power profiles of each
transformer unit and total BESS. As shown in Fig. 8, olive
curve is power of total BESS. Larger version of each
transformer unit power is shown in Fig. 9.
By implementing EMS, the power of each transformer
unit is effectively determined according to the allowable
charging and discharging power ability and SOC. In the
demonstration project, the allowable range of the battery
SOC is usually set between 20% and 80%. Under this
mode, the depth of discharge of the energy storage system
is generally within 60%.
4.2 Reactive power compensation
Figure 10 shows the test result for tracking reactive
power plan by using BESS. The blue curve is target and the
red curve is actual reactive power.
Based on the reactive power demand instructions sent
by the master station, the total reactive power of the
BESS can effectively follow the dispatched reactive
power and its response speed meets the application
requirements for voltage regulation. This reactive power
compensatory utility has been applied in practice to the
16 MW BESS.
5 Technical benefits of system
The technical benefits of the BESS are reflected in many
aspects.
1) Improvement of friendliness of renewable energy
generation connected to the grid: The power fluctuations of
renewable generation (such as wind and PV generations)
12:00:00
12:28:48
12:57:36
13:26:24
13:55:12
14:24: 00
14:52: 48
15:21:36
15:50: 24
16:19:12
16:48:00
Time
Wind/PV hybrid power Wind/PV/BESS hybrid power
Wind power
PV power
BESS power
-20
0
20
40
60
80
100
Power (MW)
Fig. 7 Power generation tracking plan
-15000
-10000
-5000
0
5000
10000
15000
12:00:00
12:28:48
12:57:36
13:26:24
13:55:12
14:24:00
14:52:48
15:21:36
15:50:24
16:19:12
16:48:00
Power (kW)
Time
C007; C008; C009
C005; C006; C001
C002; C003; Total
Fig. 8 Power profiles of each transformer unit and total BESS
-3000
-2000
-1000
0
1000
2000
3000
C007; C008; C009
C005; C006; C001
C002; C003
Power (kW)
Time
12:00:00
12:28:48
12:57:36
13:26:24
13:55:12
14:24:00
14:52:48
15:21:36
15:50:24
16:19:12
16:48:00
Fig. 9 Larger version of each transformer unit power
-5000
-4000
-3000
-2000
-1000
0
1000
2000
3000
5000
16:16:19 16:30:43 16:45:07 16:59:31 17:13:55 17:28:19 17:42:43
Reactive power (kvar)
Time
Target reactive power; Actual reactive power
4000
Fig. 10 Control effect of tracking reactive power command
Optimal control and management of a large-scale battery energy storage system to mitigate…
123
under different time scales has posed great influences on
the requirement of grid reserve, generation scheduling,
security operations, and other aspects. Therefore, the Chi-
na’s national standards for wind power connected to the
grid demand that the maximum power fluctuations of wind
farms on different time scales should satisfy different
technical indicators. Application of wind/PV/BESS hybrid
power generation can effectively stabilize wind power
fluctuation, help to reduce the impact of wind and PV
power fluctuations on the grid, improve the grid stability,
and create a better environment for grid integrations of
large-scale wind farm and PV power station.
2) Wind power curtailment: In March 2012, China’s
National Energy Administration urged the provinces (ci-
ties) to strictly implement wind power project approval
plans. In principle, new wind power project construction
for a region shall not be arranged if the wind power cur-
tailment is more than 20% of the total wind power to be
generated. Wind power curtailment will not only affect the
level of wind power development and the utilization of
clean energy, but also will reduce the wind farm invest-
ment income. Application of an energy storage system can
coordinate a grid to accommodate wind power maximally.
Furthermore, energy storage device can absorb the
renewable generation in ‘‘off peak’’ load period, and con-
duct the peak shaving in ‘‘peak’’ load period. This will
effectively not only increase the wind power accommo-
dation level in the grid, but also reduce the curtailment of
wind power.
3) Reactive power compensation: Wind and PV power
generation can easily cause bus voltage fluctuation, even
beyond the voltage limit. Such fluctuation, when serious,
will affect the voltage stability of the power grid. The
demonstration project’s BESS can provide reactive power
support according to a reactive power dispatch plan. As a
result, the BESS can assist other reactive power compen-
sation equipment to control the bus voltage at an expected
level and reduce voltage fluctuations and flickers, and
ensuring, thereby, the voltage stability of the grid con-
nected with wind and PV farms.
4) Improvement of dynamic characteristic of power
grid: The response speed of traditional frequency regu-
lation unit is relatively slow, and the ramp rate is low.
Thus, it cannot fully meet the needs of new frequency
stability problems caused by the rapid development of the
grid connected with large renewable generation. By uti-
lization of the characteristics of both fast response and
strong short-term power handling capacity of BSEE, more
benefits can be obtained through flexible control of BSEE
to participate in power grid frequency regulation. This
will help maintain the system frequency within the stan-
dard range and improve power grid security and system
stability.
6 Conclusion
In this paper, the system configuration of China’s
national demonstration project which has mixed various
generations, such as wind, PV, and BESS together with a
power transmission system is introduced, and the key
technologies and operation status of large-scale battery
energy storage system have been presented. Moreover, the
benefits of such a large BESS in dealing with the fluctua-
tion and intermittence of large renewable generation and
improving the operation stability for transmission grid, etc.,
are discussed. Overall content is summarized as follows.
1) The coordinated control and monitoring of large-scale
BESS including around 275000 single cells of lithium-ion
battery can be realized effectively. The control response time
of the BESS can be achieved within second level and this
meets the actual operational requirements for the BESS.
2) The key issues for coordinated control and energy
management of large-scale BESS can be solved by the
proposed EMS combined with local controllers and coor-
dinated control architecture.
3) A variety of applications for large BESS have been
implemented and achieved including tracking power gen-
eration plan and reactive power compensation etc. Through
demonstration it has been verified that the controllability and
schedulability for wind farm and PV power station can be
improved effectively by using proposed control systems.
4) Battery energy storage technology has some technical
benefits, however, the high cost of BESS is currently
indeed a problem which cannot be ignored. In order to
more efficient use the BESS, the further R&D and verifi-
cation associated with the operation and application tech-
nologies of the large-scale BESS need to be continued
based on the demonstration project.
Acknowledgment This work was supported by National Natural
Science Foundation of China (No. 51107126 and No. 512111046), the
Key Projects in National Science and Technology Pillar Program (No.
2011BAA07B07), the Beiing Nova Program (No.
Z141101001814094), and the Science and Technology Foundation of
State Grid Corporation of China (No. DG71-14-032).
Open Access This article is distributed under the terms of the
Creative Commons Attribution 4.0 International License (http://
creativecommons.org/licenses/by/4.0/), which permits unrestricted
use, distribution, and reproduction in any medium, provided you give
appropriate credit to the original author(s) and the source, provide a
link to the Creative Commons license, and indicate if changes were
made.
References
[1] Li XJ, Hui D, Lai XK (2013) Battery energy storage station
(BESS)-based smoothing control of photovoltaic (PV) and wind
power generation fluctuations. IEEE Trans Sustain Energy
4(2):464–473
Xiangjun LI et al.
123
[2] Toge M, Kurita Y, Iwamoto S (2013) Supplementary load fre-
quency control with storage battery operation considering SOC
under large-scale wind power penetration. In: Proceedings of the
2013 IEEE Power and Energy Society general meeting, Van-
couver, Canada, 21–25 Jul 2013, 5 pp
[3] Kawakami N, Iijima Y, Fukuhara M et al (2010) Development
and field experiences of stabilization system using 34 MW NAS
batteries for a 51 MW wind farm. In: Proceedings of the 2010
IEEE international symposium on industrial electronics
(ISIE’10), Bari, Italy, 4–7 Jul 2010, pp 2371–2376
[4] Khatamianfar A, Khalid M, Savkin AV et al (2013) Improving
wind farm dispatch in the Australian electricity market with
battery energy storage using model predictive control. IEEE
Trans Sustain Energy 4(3):745–755
[5] Teleke S, Baran ME, Huang AQ et al (2009) Control strategies
for battery energy storage for wind farm dispatching. IEEE
Trans Energy Convers 24(3):725–732
[6] Baccino F, Grillo S, Marinelli M et al (2011) Power and energy
control strategies for a vanadium redox flow battery and wind
farm combined system. In: Proceedings of the 2nd IEEE PES
international conference and exhibition on innovative smart grid
technologies (ISGT Europe’11), Manchester, UK, 5–7 Dec
2011, 8 pp
[7] Li WY (2015) Framework of probabilistic power system plan-
ning. CSEE J Power Energy Syst 1(1):1–8
[8] Nguyen CL, Lee HH (2016) Effective power dispatch capability
decision method for a wind-battery hybrid power system. IET
Gener Transm Distrib 10(3):661–668
[9] Haaren RV, Morjaria M, Fthenakis V et al (2015) An energy
storage algorithm for ramp rate control of utility scale PV
(photovoltaics) plants. Energy 91:894–902
[10] Wang GS, Ciobotaru M, Agelidis VG (2014) Power smoothing
of large solar PV plant using hybrid energy storage. IEEE Trans
Sustain Energ 5(3):834–842
[11] Yuan Y, Sun CC, Li MT et al (2015) Determination of optimal
supercapacitor-lead-acid battery energy storage capacity for
smoothing wind power using empirical mode decomposition and
neural network. Electr Power Syst Res 127:323–331
[12] Trung TT, Ahn SJ, Choi JH et al (2014) Real-time wavelet-
based coordinated control of hybrid energy storage systems for
denoising and flattening wind power output. Energies
7(10):6620–6644
[13] Abbey C, Strunz K, Joo
´s G (2009) A knowledge-based approach
for control of two-level energy storage for wind energy systems.
IEEE Trans Energy Convers 24(2):539–547
[14] Shigematsu T (2011) Redox flow battery for energy storage. SEI
Tech Rev 73:4–13
[15] Senjyu T, Kikunaga Y, Yona A et al (2008) Coordinate control
of wind turbine and battery in wind power generator system. In:
Proceedings of the 2008 IEEE Power and Energy Society gen-
eral meeting—conversion and delivery of electrical energy in
the 21st century, Pittsburgh, PA, USA, 20–24 Jul 2008, 7 pp
[16] Ferreira SR, Rose DM, Schoenwald DA et al (2012) Protocol for
uniformly measuring and expressing the performance of energy
storage systems. PNNL-22010. Pacific Northwest National
Laboratory, Richland, WA, USA
[17] US utility-scale battery storage market surges forward.
2011NARP 825-110928. IHS Emerging Energy Research LLC,
Cambridge, MA, USA
[18] DOE global energy storage database. http://www.
energystorageexchange.org/projects
[19] Skyllas-Kazacos M, Kazacos G, Poon G et al (2010) Recent
advances with UNSW vanadium-based redox flow batteries. Int
J Energy Res 34(2):182–189
[20] Yoshimoto K, Nanahara T, Koshimizu G (2006) New control
method for regulating state-of-charge of a battery in hybrid wind
power/battery energy storage system. In: Proceedings of the 2006
IEEE PES power systems conference and exposition (PSCE’06),
Atlanta, GA, USA, 29 Oct-1 Nov 2006, pp 1244–1251
[21] Viswanathan V, Kintner-Meyer M, Balducci P et al (2013)
National assessment of energy storage for grid balancing and
arbitrage, phase II volume 2: cost and performance. PNNL-
21388 PHASE II/Vol 2. Pacific Northwest National Laboratory,
Richland, WA, USA
[22] Supra JD (2012) Washington energy update: August-September
2012. White & Case LLP, New York, NY, USA
[23] Kumagai J (2012) A battery as big as the grid. IEEE Spectrum
49(1):45–46
Xiangjun LI received the B.E. degree in electrical engineering from
Shenyang University of Technology, China, in July 2001, and the
M.E. and Ph.D. degrees in electrical and electronic engineering from
Kitami Institute of Technology (KIT), Japan, in March 2004 and
March 2006, respectively. From May 2006 to March 2010, he worked
as post-doctoral research fellow at Korea Institute of Energy Research
(KIER), Daejeon, Korea, and Tsinghua University, Beijing, China,
respectively. In March 2010, he joined Electrical Engineering and
New Material Department, China Electric Power Research Institute
(CEPRI), Beijing, China, where he has been engaged in the topic of
integration/control/SCADA/application technologies for large-scale
multi-type battery energy storage system/station, wind/PV/battery
hybrid distributed generation systems, and micro-grids. His research
interests include renewable energy power generation, electric energy
saving/storage technology, and power system engineering. Dr. Li is a
chartered engineer, and the senior members of IEEE, CSEE, CAS,
and CES, etc. He has been the Editor-in-Chief for a Springer book
series entitled ‘‘Renewable Energy Sources & Energy Storage’’ since
2016. He has served as the Chair of the IEEE CIS Task Force on
‘‘ADP and RL in Power and Energy Internets’’ since 2016. He also
serves as several members, such as the IEC TC120 WG3, the IEEE
CIS Adaptive Dynamic Programming and Reinforcement Learning
Technical Committee, and the IEEE CIS Intelligent Systems Appli-
cations Technical Committee, respectively.
Liangzhong YAO received the M.Sc. and Ph.D. degrees in electrical
power engineering from Tsinghua University, Beijing, China, in 1989
and 1993, respectively. He is currently the Vice President and the
Doctoral Supervisor of the China Electric Power Research Institute
(CEPRI). Prior to CEPRI, he was the Senior Power System Analyst at
ABB UK Ltd from 1999 to 2004, and was the Department Manager
for network solution and renewable energy at Alstom Grid Research
& Technology Centre in the UK from 2004 to 2011. Dr. Yao is a
Fellow of IET, a chartered engineer, and a member of CIGRE. He is
also the visiting professor at the University of Bath in the UK, and the
guest professor at both Shanghai Jiao Tong University and Sichuan
University, China.
Dong HUI received his B.S. and M.S. degrees in semiconductor
physics and devices from Huazhong University of Science and
Technology, China in 1990 and 1995, and Ph.D. degree in applied
superconductivity, from Institute of Electrical Engineering, Chinese
Academy of Sciences, in 1998, respectively. He worked as visiting
scientist at Deutsches Elektronen Synchrotron (DESY), Hamburg,
Germany, from May 1999 to May 2002. After then, he worked as
associate professor, at the Institute of Electrical Engineering, Chinese
Academy of Sciences, from May 2002 to May 2007. He has been a
professor at China Electric Power Research Institute since June 2007
and now the chief engineer of electrical engineering and new material
department of CEPRI. His research interests include large scale
battery energy storage and power electronics.
Optimal control and management of a large-scale battery energy storage system to mitigate…
123
