Self-adaptive

Self-adaptive Emergency
Emergency Control
Control Strategy
Strategy of
ofWind
Wind
Farm-Level
Farm-Level Energy
Energy Storage for Improving
Storage for Improving
Frequency
Frequency Stability
Stability of
ofWind
Wind Power
Power Grid-connected
Grid-connected
System
System
st nd
11st Qinglei
Qinglei Zhang
Zhang 22nd Jiaming
Jiaming Wang
Wang 33rdrdGean
Gean Cui
Cui
Dispatching
Dispatching and
and Control
Control Center
Center School
School ofofElectrical
Electrical &
& Dispatching
Dispatching and
and Control
Control Center
Center
State
State Grid Shaanxi Electric
Grid Shaanxi Electric Power
Power Electronic
Electronic Engineering North
Engineering NorthChina
China State
State Grid Shaanxi Electric
Grid Shaanxi Electric Power
Power
Company
Company Electric
Electric Power
Power University
University Company
Company
Xi’an,
Xi' an, China
China Beijing,
Beijing, China
China Xi’an,
Xi' an, China
China
zhangqinglei@sn.sgcc.com.cn
zhangqinglei@sn. sgcc. com. en 1366265698@qq.com
1 366265698@qq. com cgaagc@126.com
cgaagc@ l 26.com
2023 IEEE 6th International Electrical and Energy Conference (CIEEC) | 979-8-3503-4667-1/23/$31.00 ©2023 IEEE | DOI: 10.1109/CIEEC58067.2023.10167240

th th
441h Qi
Qi Dong
Dong 551h Tong
Tong Wang
Wang
Dispatching
Dispatching and
and Control
Control Center
Center School
School of
ofElectrical
Electrical &
&
State
State Grid
Grid Shaanxi
Shaanxi Electric
Electric Power
Power Electronic
Electronic Engineering
Engineering North
NorthChina
China
Company
Company Electric
Electric Power
Power University
University
Xi’an,
Xi' an, China
China Beijing,
Beijing, China
China
7500970@qq.com
7500970@qq.com hdwangtong@126.com
hdwangtong@ 1 26.com

Abstract—Aiming
Abstract-Aiming at at the
the problem
problem that that doubly-fed
doubly-fed induction
induction system
system frequency
frequency modulation.
modulation. TheThe frequency
frequency stability
stability isis
generator
generator (DFIG)
(DFIG) cannotcannot participate
participate in in system
system frequency
frequency greatly
greatly affected
affected when
when DFIG
DFIG isis connected
connected to
to the
the grid
grid [2-3].
[2-3] .
regulation,
regulation, an an self-adaptive
self-adaptive emergency
emergency controlcontrol strategy
strategy applied
applied
to
to battery
battery energy
energy storage
storage system
system is is proposed
proposed in in this
this paper.
paper. Firstly,
Firstly,
At
At present,
present, battery
battery energy
energy storagestorage technology
technology is i s widely
widely
the
the working
working principle
principle of of the
the battery
battery energy
energy storage
storage system
system used
used inin power
power systems,
systems, and and itit isis mainly
mainly used used for for peak
peak shaving
shaving
participating
participating in in frequency
frequency modulation
modulation is is introduced.
introduced. AndAnd then
then aa and
and valley
valley filling,
filling, power
power emergency
emergency control control and and frequency
frequency
battery
battery control
control strategy
strategy is is proposed
proposed based based on on the
the active-
active­ regulation
regulation in in the
the power
power system
system [4]. [4] . When
When using using battery
battery energy
energy
frequency
frequency droop
droop control.
control. The
The strategy
strategy is is based
based onon adaptive
adaptive single
single storage
storage system
system (BESS)
(BESS) to to improve
improve the the frequency
frequency stability
stability of of
neuron proportion integration differentiation
PID
neuron proportion integration differentiation (PID) controller, controller, wind
wind power
power grid-connected
grid-connected system, system, how how to to select
select the the
and reference
reference active
active powerpower will will directly
directly affect affect the the system
system
and the
the control
control parameters
parameters in in the
the frequency
frequency controller
controller are are
adjusted performance.
performance. The The active
active power
power at at the
the grid-connected
grid-connected bus bus ofof
adjusted online
online adaptively
adaptively according
according to to the
the real-time
real-time frequency.
frequency.
Finally,
Finally, the
the simulation
simulation of of wind-storage
wind-storage grid-connected
grid-connected system
system is is
the
the wind
wind turbine
turbine generator
generator isis usedused as as the
the reference
reference powerpower in in
carried
carried out,
out, and
and thethe maximum
maximum absolute
absolute frequency
frequency difference
difference is is
the
the direct
direct power
power control
control [5],
[5], and
and the the frequency
frequency stability
stability isis not
not
introduced
introduced to to analyze
analyze the
the performance
performance of ofthe
the control
control system.
system. TheThe ideal.
ideal. In
In order
order to to further
further improve
improve the the frequency
frequency stability,
stability, thethe
effectiveness
effectiveness of of the
the control
control strategy
strategy is is verified
verified in in the
the single-
single­ real-time
real-time frequency
frequency can can be be used
used as as thethe input
input of of the
the control
control
machine
machine infinity
infinity andand multi-machine
multi-machine systems,
systems, andand thethe control
control system,
system, and and the the reference
reference active
active power power can can be be obtained
obtained
effect
effect of
of the
the strategy
strategy proposed
proposed in in this
this paper
paper is is better
better than
than the
the through
through the active-frequency droop control link[6-7] . However,
the active-frequency droop control link[6-7]. However,
fixed-coefficient
fixed-coefficient droop
droop control
control through
through comparison.
comparison. when
when thethe droop
droop coefficient
coefficient isis small,
small, the the BESS
BESS cannot
cannot be be fully
fully
used.
used. When
When itit isis large,
large, the
the system
system instability
instability will will bebe caused.
caused.
Keywords—frequency
Keywords-frequency regulation;
regulation; single
single neuron
neuron PID;
PID; Battery
Battery Therefore,
Therefore, the the reference
reference active
active power power can can be be generated
generated
energy
energy storage;
storage; emergency
emergency control;
control; wind-storage
wind-storage grid-connected
grid-connected through
through the adaptive droop coefficient link to participate in
the adaptive droop coefficient link to participate in
system
system system
system frequency
frequency modulation
modulation [8]. [8] . In
In order
order to to further
further improve
improve
the
the performance
performance of ofthe
the control
control system,
system, The The fuzzy
fuzzy controller
controller in in
I.I. IINTRODUCTION
NTRODUCTION the
the intelligent
intelligent controller
controller isis adopted
adopted and and thethe reference
reference powerpower isis
Among generated
generated by by the
the frequency
frequency change
change rate, rate, which
which isis proposed
proposed in in
Among many
many newnew energy
energy power
power generation,
generation, wind
wind power
power isis the
applied
applied early,
early, the
the technology
technology is is relatively
relatively mature,
mature, and
and the
the the paper
paper [9].
[9] . Based
Based on on the
the above
above research,
research, the the application
application of of
installed intelligent
intelligent controller
controller for for BESS
BESS isis discussed
discussed in in this
this paper
paper .. InIn
installed capacity
capacity ranks
ranks first
first [1].
[ 1 ] . Doubly-fed
Doubly-fed induction
induction addition,
generator addition, the analysis of frequency stability of DFIG under
the analysis of frequency stability of DFIG under aa
generator (DFIG)
(DFIG) isis oneone of
of the
the widely
widely used used wind
wind turbines,
turbines, certain
which
which is is connected
connected to to the
the grid
grid through
through AC-DC-converter.
AC-DC-converter. certain operating
operating mode mode whenwhen the the system
system isis stablestable isis mostly
mostly
However, involved
involved in in the
the current
current research,
research, and and the the frequency
frequency stability
stability inin
However, due due to
to the
the randomness
randomness and and volatility
volatility of
of wind
wind power
power case
generation, case ofof short
short circuit
circuit fault
fault under
under full full wind
wind speed
speed hashas notnot been
been
generation, and
and DFIG's
DFIG's active
active output
output is is only
only determined
determined byby its
its discussed.
converter discussed.
converter control
control command,
command, so so DFIG
DFIG cannot cannot participate
participate inin
Supported
SupportedbybyScientific
ScientificProject
Project(No.
(No. SGSN0000DKJS2201510)
SGSNOOOODKJS2201 5 1 0)of
ofState
State
Grid
GridShaanxi
Shaanxi Electric
ElectricPower
PowerCompany.
Company.

BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB
‹,(((
979-8-3503-4667-1/23/$31.00 ©2023 IEEE


2290
Authorized licensed use limited to: University of Edinburgh. Downloaded on March 30,2024 at 11:38:30 UTC from IEEE Xplore. Restrictions apply.
 Therefore,
Therefore, anan adaptive
adaptive control
control strategy
strategy for
for BESS
BESS based
based onon The
The weighting
weighting coefficient
coefficient isis continuously
continuously adjusted
adjusted by by the
the
single
single neuron
neuron PIDPID controller
controller is is proposed
proposed in in this
this paper.
paper. The
The single
single neuron
neuron PID PID toto realize
realize self-adaptive
self-adaptive learning.
learning. TheThe
real-time
real-time frequency
frequency isis used
used asas input
input in
in the
the control
control strategy,
strategy, and
and weighting
weighting coefficient
coefficient cancan be
be adjusted
adjusted by by different
different learning
learning
the
the reference
reference active
active power
power isis generated
generated by by aa single
single neuron
neuron PID
PID rules,
rules, thus
thus forming
forming different
different control
control algorithms.
algorithms. In In this
this paper,
paper,
controller
controller toto improve
improve thethe frequency
frequency stability
stability ofof the
the wind
wind the
the supervised
supervised Hebb Hebb learning
learning rule
rule isis adopted,
adopted, that
that isis the
the
power
power grid-connected
grid-connected system
system under
under thethe operating
operating modemode of
offull
full adaptive
adaptive calculation
calculation process
process ofof the
the weighting
weighting coefficient
coefficient isis
wind
wind speed.
speed. Finally,
Finally, DFIG
DFIG and and BESS
BESS areare incorporated
incorporated into
into related
related toto the
the input,
input, output
output and
and output
output errors
errors of
of the
the controller.
controller.
single-machine
single-machine infinite
infinite system
system and and multi-machine
multi-machine system
system forfor The
The specific
specific calculation
calculation process
process ofof adaptive
adaptive ff ofof weighting
weighting
simulation
simulation analysis to verify the effectiveness of the control
analysis to verify the effectiveness of the control coefficient
coefficient isis shown
shown in in equation
equation (5),
(5), and
and the
the output
output ofof
strategy.
strategy. controller
controller isis shown
shown inin equation
equation (6).
(6).

OJi;
(k
k )=
OJi;
(k
k -11)+ry;U
iU
(k
k )XX;i
(k
k )Zz
(k)
k
II.
II. SSINGLE
INGLE N EURON PID
NEURON PID CCONTROLLER
ONTROLLER
5
(5)
ZZ
(k
k )==U
U
(k
k )-U
U
(k
k -11)
PID
PID controller
controller isis widely
widely used
used in
in power
power system.
system. The
The system
system
error
error can
can be
be responded
responded quickly,
quickly, and
and the
the steady-state
steady -state error
error cancan 33
be
be effectively
effectively adjusted
adjusted to
to zero
zero by
by PID
PID controller.
controller. There
There are are Uu
(k)
k  =Uu
(k
k -11)+K
K 
I
OJii;;
(kk )XX;i
(k)
k
two
two types
types of
of PID
PID controllers:
controllers: positional
positional type
type and
and incremental
incremental 1
ii=l
type,
type, and
and the
the equations
equations are
are respectively
respectively equation
equation (1)( 1 ) and
and
i
k 
6
(6)
equation
equation (2).
(2).
ii
k  
3


i
k 
U
k   K p e
k   Ki  e
k   Kd
e
k   e
k  1 
1 i 1

Where
OJ;i (i=1,2,3)

Where (i= l ,2,3) is
is the
the weighting coefficient,
OJ11 is1s
weighting coefficient,
f..UU
(k) KpP
(ee(k
k  = K
k -11)) +KKi;ee(k

k )-ee(k
k )+ corresponded
corresponded toto K
Krp in
in the
the PID controller,
OJ2
PID controller, 2 is
is corresponded
corresponded

2
(2)
Kdd
(ee
(kk) -22ee
(kk-11)+ee
(kk-22))
K to
to K i,

K;, OJ33 is
is corresponded
corresponded to
to K d, 17i; (i=1,2,3)
Ka, (i=l ,2,3) isis the
the learning
learning
efficiency
efficiency of the weighting coefficient, and K isis the
of the weighting coefficient, and K the
Single
Single neuron
neuron model
model isis one
one of
of the
the most
most basic
basic control
control proportional
proportional coefficient.
coefficient.
elements
elements with
with simple
simple structure,
structure, convenient
convenient application
application and
and
strong
strong adaptive
adaptive ability.
ability. When
When combining
combining PID
PID controller
controller with
with it,
it,
the
the PID
PID parameters
parameters can
can be
be adjusted
adjusted online,
online, the
the response
response speed
speed III.
III. TTHE
HE B ATTERY E
BATTERY NERGY SSTORGE
ENERGY TORGE SSYSTEM
YSTEM
can
can be
be improved
improved and
and the
the system
system robustness
robustness and
and adaptability
adaptability The
The main components of BESS include isolation
main components of BESS include isolation
can
can be
be enhanced[10].
enhanced[ l O] . transformer,
transformer, third-order
third-order LCL L CL filter,
filter, three-phase
three-phase bridge
bridge fully
fully
The controlled
controlled converter,
converter, lithium-ion
lithium-ion battery
battery andand control
control system.
system.
The structure
structure of of single
single neuron
neuron PIDPID is i s shown
shown inin Figure
Figure 1,1 ,
where
where rr (k)
(k) is
is the
the given
given value
value and
and yy (k)
(k) is
i sthe
the actual
actual value.
value. The
The
The
The specific
specific structure
structure is is shown
shown in in Figure
Figure 2.2. The
The battery
battery
output control
control system
system isis mainly
mainly composed
composed of of four
four parts,
parts, namely
namely
output after
after the
the converter
converter is is shown
shown in in equation
equation (3).(3). According
According
to
to equation
equation (4)(4) and
and equation
equation (2),
(2), the
the single
single neuron
neuron PIDPID isis an
an
frequency
frequency controller,
controller, power power controller,
controller, SOC SOC controller
controller and
and
incremental
incremental PID, PID, and and thethe weighting
weighting coefficient
coefficient is is
current
current controller.
controller. The The output
output power
power of of BESS
BESS isis shown
shown in in
corresponded equation
equation (7),
(7), and
and itit isis closely
closely related
related toto the
the AC
AC side
side voltage
voltage of
of
corresponded to to the
the control
control parameters
parameters in in the
the incremental
incremental PID.
PID.
The
The difference
difference is is that
that the
the control
control parameters
parameters of of the
the single
single
the
the converter.
converter. Therefore,
Therefore, the the control
control of of output
output power
power can
can be
be
neuron
neuron PID
PID can
can bebe learned
learned adaptively.
adaptively .
converted
converted into
into the
the control
control of of converter.
converter. However,
However, the
the actual
actual
output
output of
of BESS
BESS will will be be limited
limited by by thethe capacity
capacity ofof the
the
r(k) X1 converter.
converter.

1
converter

X2
2 U U Current Direction
f system
Grid Side

y(k)
battery

X3 v u

3 Z -1
U0 I0
SOC
Power
Calculat-
Fig. ion
Fig. 1.1 . Schematic
Schematicdiagram
diagramof
ofsingle
single neuron
neuronPID
PIDstructure
structnre Coordinate
Transforma-

k  -ee(k
X11 ==ee(k)
X
k -11) tion PWM
f0 P Q Id Iq

3 I qref

k 
Control System

X2 ==ee(k)
X (3) P ref Uref
2 frequency Power Current Coordinate
Controller Controller Controller Transforma-
X
k )-22ee(k
x33==ee(k
k -11)+ee(k

k -22) SOC
Controller
tion
I dref
1
(ee(k
f ref Qref =0
U
(k
t-.U k )==
lUJ
k -11)) +
OJ2ee(k

k )-ee(k 2
k )+

4
3
(ee
(kk)
( 4)

�  -22ee
(kk-11)+ee
(kk-22))
Fig.
Fig. 2.
2. Battery
Batteryenergy
energystorage
storagesystem
system


2291
Authorized licensed use limited to: University of Edinburgh. Downloaded on March 30,2024 at 11:38:30 UTC from IEEE Xplore. Restrictions apply.
 Assuming
Assuming that thatthe
theresistance
resistance value
value between
between BESSBESS andandthethe converter
converter output value
output value , the
the reference
reference power
power change
change value
value
power
power gridgrid isis much
much less less than
than thethe reactance
reactance value,
value, andand the
the and
and the reference power change value at the last momentare
the reference power change value at the last moment are
phase
phaseangle
angledifference
differencebetween
betweenthe thetwo
twovoltages
voltagesisisvery
verysmall,
small, used
used to
to adjust
adjust the
the weighting
weighting coefficient
coefficient online.
online. The
The specific
specific
the
the expression
expression of of active
active powerpower in in equation
equation (7) (7) can
can be be process
process isis shown
shown in in equation
equation (5),
(5), where
where thethe proportion
proportion
simplified
simplifiedto to equation
equation (8).(8). Generally
Generally speaking,
speaking, thetheproduct
product ofof coefficient
coefficient has
has aa great
great impact
impact on
on the
the overall
overall performance
performance of of
voltage
voltagecancanbeberegarded
regardedas as aaconstant
constantapproximately,
approximately, so soititcan
can the
the system,
system, and
and thethe learning
learning efficiency
efficiency ofof the
the weighting
weighting
be
be considered
considered that thatthethe phase
phase angle
angle of
ofactive
activepower
powerand and BESS
BESS
coefficient
coefficientcan
canbebeselected
selectedtotodifferent
differentvalues
valuesaccording
accordingto tothe
the
export
exportvoltage
voltageisislinear.
linear. AndAndbecause
becausefrequency
frequency willwillaffect
affectthe
the requirements.
requirements.
phase
phase angle
angle of of voltage,
voltage, itit cancan bebe approximately
approximately considered
considered
that
that the
the active
active powerpower and and frequency
frequency are are linear,
linear, so so the
the The
Thereference
referencepower
powerand andthetheactual
actualpower
powervaluevalueare areused
used
frequency
frequency can can be be adjusted
adjusted by by controlling
controlling the
the active
active power.
power. by
by the
the power
powercontroller
controllerto to generate
generate thethereference
reference current,
current, and and
The
The real-time
real-time frequency
frequency isis usedused asasinput
inputininthe
thebattery
battery control
control the
the specific
specific structure
structure isis shownshown in in Figure
Figure 4.4. TheThe direct
direct and and
system.
system. TheThe reference
reference powerpower isis generated
generated byby aa single
single neuron
neuron quadrature
quadrature axis
axis reference
reference current
current isis output
output byby thethe active
active and and
PID
PID controller,
controller, and and the
the active
active power
power output
output isis controlled
controlled by by reactive
reactivepower
powercontrollers.
controllers. The Thestructure
structureand andparameters
parametersof ofthe the
the
thepower
power outside
outsideloop loopand andthethe current
currentinside
insideloop.
loop. TheThe active
active two
two controllers
controllers areare thethe same,
same, thusthus only
only the
the structure
structure of of thethe
power
power isis affected
affected by by thethe frequency
frequency value
value inin turn,
tum, thus
thus the
the active
active power
power controller
controllerisis shownshown inin figure.
figure. AAfirst-order
first-orderfilter
filter
frequency
frequencyisisadjusted
adjustedand andthethefrequency
frequencystability
stability isisimproved.
improved. isisadded
addedto tothe
thecontrol
controllinklinkto tosmooth
smooththe thecontroller
controllerinput.
input. TheThe
PI-Lead
PI-Lead controller
controller isis composed
composed of of lead
lead correction
correction link link andand
In
In the equation
7and
the equation ( 7 ) and
8,
( 8 ) , PP and
andQQ are arethethe output
output traditional
traditional PI PI controller,
controller, the the output
outputof ofBESS
BESS is is limited
limited within
within
active
active power and reactive power of BESS, v is the grid-
power and reactive power of BESS, v is the grid­ the
the capacity
capacity range
range of of the
the converter
converter by by the
the limiting
limiting link,
link, an an
connected
connected bus bus voltage,
voltage, the
the phase
phaseangle
angleofofvv isis 0,0, uu isis the
theAC
AC equal
equal number
number of ofpoles
poles and and zeros
zeros areare added
added toto the
the system
system by by
side
side voltage
voltage ofofthe
the converter,
converter, and
andthe
thephase
phaseangle
angleisis assumed
assumed the
the lead
lead correction
correctionlink. link. The
The zero
zero point
point isis closer
closerto to the
the origin
origin
to be (}, ,RX
tobe R, Xarearethe
theimpedance
impedancevalue
valuebetween
betweenBESS BESSand andthe
the
than
than thethe pole,
pole, that
that isis TT1>T2
1>T2 in in Figure
Figure 4,4, and
and the
the medium
medium and and
high
highfrequency
frequencyinterference
interferencecan canbe beeffectively
effectivelyrejected
rej ected[11].
[1 1] .
grid.
grid.

Pp=
R
(uu-vvcos
uu[R cos B )+Xv sin B]
Xvsin
1 1  sT1 1
RR22+X
X 22 Pref 1 s K p1 

7
(7) T 1  sT2 sTi Idref
uu[XX
(uu-vvcos
cos B )-Rv sin B]
Rvsin
Q=
Q Pin
RR22+XX22 First-order
First-orderfilter
filter
PI-Lead
PI-Lead
uv
PP= uvB
8
( 8)
XX Fig.
Fig.4.4. Active
Activepower
powercontroller
controller

The
The frequency
frequency controller
controllerisis improved
improved based
based ononthethe active-
active­ Whether
Whether the the SOC
SOC value
value of of the
the battery
battery isis within
within thethe
frequency
frequency droop
droop controller,
controller, and
and the
the detailed
detailed composition
composition isis normal
normal working
working range
range (20%
(20% -- 100%)
1 00%) cancan bebe judged
judged byby thethe
shown
shown inin Figure
Figure 3.3. The
The real-time
real-time frequency
frequency atat the
the connection
connection SOC
SOC controller
controller to to effectively
effectively avoidavoid overcharging
overcharging and and
point
point of
ofthe
the wind
wind storage
storage system
system isis taken
taken asas the
the input
input value.
value. discharging
discharging of ofthe
the battery
battery and
and ensure
ensure thethe safe
safe and
and economic
economic
The
The given
given frequency
frequency value
value isis 50Hz.
50Hz. TheThe reference
reference power
power operation
operation of of BESS.
BESS. The The input
input ofof the
the SOC
SOC controller
controller isis the
the
change
change value
value dP dP isis generated
generated after
after the
the single
single neuron
neuron PIDPID output
output of of the
the active
active power
power controller.
controller. IfIf the
the SOC
SOC value
value isis
controller.
controller. Then
Then thethe reference
reference active
active power
power isis obtained
obtained by by judged
judged to to be
be within
within the
the normal
normal operating
operatingrange,
range, the
the output
outputof of
adding
adding dPdP with
with the
the rated
rated active
active power
power of of BESS
BESS at at 50Hz.
50Hz. the
the SOC
SOC controller
controllerisisequal
equaltoto the
theinput,
input,otherwise,
otherwise, the
theoutput
output
Through
Through droop
droop control,
control, the
the real
real time
time frequency
frequency difference
difference isis isiszero,
zero,that
thatisisthe
thefrequency
frequencyresponse
responseisisnotnotinvolved
involvedby byBESS.
BES S.
multiplied
multiplied byby the
thefixed
fixeddroop
droop coefficient
coefficientto to directly
directly obtain
obtainthethe The
The direct
direct axis
axis and
and quadrature
quadrature axisaxis reference
reference voltages
voltages areare
reference
reference power
power change
change value.
value. The The single
single neuron
neuron PID PID obtained
obtained by by the
the current
current controller
controller through
through the
the direct
directaxis
axis and
and
controller
controller can
can realize
realize the
the adaptive
adaptive adjustment
adjustment of of the
the control
control quadrature
quadrature axis axis reference
reference currents.
currents. The The output
output voltage
voltage isis
coefficient,
coefficient, and
and introduce
introduce thethe frequency
frequency difference
difference atat the
the last
last generated
generatedintointopulse
pulsetototrigger
triggerthe
theconverter
converterthrough
throughthethePWM
PWM
moment
momentto toimprove
improvethe theBESS
BESSfrequency
frequencymodulation
modulationability.
ability . generator
generator after the output voltage is transformed into
after the output voltage is transformed into
coordinate,
coordinate,so sothat
thatthe
theBESS
BESSoutput
outputpowerpowercan
canbe becontrolled.
controlled.
f0 X1

Converter

1
dP Pref
f IV.
IV. SSIMULATION
IMULATION A NALYSIS
ANALYSIS
50Hz
2 Pe
X2 Z -1

A. Single-machine
A. Single-machinesystem
system
single
singleneuron
neuronPID
PIDcontroller
controller When
When the
the grid
grid side
side fails,
fails, the
the active
active power
power output
outputwill
will be
be
Fig.
Fig.3.3 . Frequency
Frequencycontroller
controller abnormal,
abnormal, resulting
resulting the the frequency
frequency atat the the wind
wind turbine
turbine
According connection
connectionpoint
point to
to deviate
deviate from
fromthethereference
referencevalue
value of
of50Hz.
50Hz.
Accordingto tothe
theactual
actualdemand,
demand, only onlypart
partof
ofthe
thefunctional
functional
structure In
Inthe
theprocess
processof
offrequency
frequencyfluctuation,
fluctuation,thethemaximum
maximumvaluevalueof of
structure isis replaced
replaced byby the
the single
single neuron
neuronPID
PID controller.
controller. The
The
real-time the
theabsolute
absolutevalue
valueof ofthe
thefrequency
frequency difference
differenceisisreferred
referredtotoas
as
real-time frequency
frequency difference
difference isis calculated
calculated and
and stored
storedby
by the
the
converter, the
the maximum
maximum absolute
absolute frequency
frequency difference.
difference. Whether
Whether the the
converter, and and the
the output
output expression
expression isis equation
equation (3).
(3). The
The
supervised relevant
relevant protection
protection equipment
equipment operates
operates will
will be be directly
directly
supervised Hebb Hebb learning
learning rule
rule isis adopted
adopted byby the
the adaptive
adaptive
calculation affected
affectedby
bythe
themaximum
maximumabsoluteabsolutefrequency
frequencydifference.
difference. IfIfthe
the
calculation process
process ff of of the
the weighting
weighting coefficient,
coefficient, thethe


2292
Authorized licensed use limited to: University of Edinburgh. Downloaded on March 30,2024 at 11:38:30 UTC from IEEE Xplore. Restrictions apply.
wind
wind turbine
turbine isis off
offthe
the grid
grid and
and cannot
cannot provide
provide power
power support
support the
the control
control strategy
strategy in
in this
this paper
paper can
can effectively
effectively reduce
reduce the
the
at
at all,
all, the
the operation
operation ofof the
the system
system will
will be
be further
further deteriorated
deteriorated frequency
frequency difference and improve the frequency stability. The
difference and improve the frequency stability. The
and
and the
the frequency
frequency stability
stability will
will be
be reduced.
reduced. The
The possibility
possibility of
of effectiveness
effectiveness ofof the
the strategy
strategy isis verified
verified under
under the
the operating
operating
wind
wind turbine
turbine offoff grid
grid operation
operation cancan be
be reduced
reduced andand the
the mode
mode ofoffull
full wind
wind speed.
speed.
frequency
frequency stability
stability can
can be be improved
improved by by decreasing
decreasing thethe
frequency
frequency difference,
difference, soso this
this value
value can
can be
be used
used to
to characterize
characterize B.
the
B. Multi-machine
Multi-machine system
system
the frequency
frequency stability.
stability.
A
A multi-machine
multi-machine system system isis builtbuilt toto further
further verify
verify thethe
A
A single-machine
single-machine infinite
infinite system
system isis built
built to
to preliminarily
preliminarily effectiveness
verify effectiveness of the control strategy proposed in this paper
of the control strategy proposed in this paper
verify the
the effectiveness
effectiveness of ofthe
the control
control strategy
strategy proposed
proposed in in this
this under
paper under full
full wind
wind speed,
speed, andand the the control
control strategy
strategy isis also
also
paper under
under full
full wind
wind speed
speed conditions.
conditions. In In the
the simulation
simulation compared
system, compared withwith thethe droop
droop control.
control. The The multi-machine
multi-machine systemsystem
system, the
the 110kV
l l OkV three-phase
three-phase voltage
voltage source
source isis used
used to to
realize structure
structure isis shown
shown in in Figure
Figure 6, 6, where
where DFIGDFIG and and BESS
BESS are are
realize the
the infinite
infinite power
power grid
grid simulation,
simulation, and and the
the infinite
infinite
power located
located atat the
the farthest
farthest right
right end,
end, and and the
the other
other generators
generators areare
power gridgrid isis connected
connected to to DFIG
DFIG and and BESS
BESS through
through
transformers,
transformers, lines
lines and
and other
other components.
components. The The composition
composition of of synchronous
synchronous generators.
generators. The The ratedrated voltage
voltage at at the
the DFIG
DFIG
BESS
BESS hashas been
been described
described in
in detail
detail in
in the
the previous
previous section
section andand connection
connection point
point isis 690V,
690V, the
the rated
rated active
active power
power of of aa single
single
will
will not
not be
be repeated
repeated here.
here. The
The PID
PID controller
controller inin the
the simulation
simulation DFIG
DFIG is 1 . 5MW, and the wind speed at the rated power
is 1.5MW, and the wind speed at the rated power isis
is
is programmed
programmed and and realized
realized byby SSfunction.
function. 15m/s.
1 5m/s. A
A total
total of
of44 DFIGs
DFIGs areare set
set inin the
the multi-machine
multi-machine system.
system.
The
The output
output active
active power
power isis set
set to
to 0MW
OMW when when thethe frequency
frequency at at
DFIG
DFIG has has different
different operating
operating characteristics
characteristics under
under the
different the wind
wind turbine
turbine connection
connection point
point is is 50
50 Hz,
Hz, and
and the
the capacity
capacity ofof
different wind wind speeds,
speeds, andand the
the frequency
frequency of of unsymmetrical
unsymmetrical BESS
BESS isis set
set to
to 0.3MW
0. 3MW andand 380Ah.
3 80Ah.
and
and symmetrical
symmetrical faults
faults ofof the
the system
system is is analyzed
analyzed under
under full
full
wind
wind speed.speed. TheThe BCBC two-phase
two-phase short
short circuit
circuit and
and three-phase
three-phase A
grounding
grounding short short circuit
circuit respectively
respectively occur occur in in the
the middle
middle ~ ~
section
section of of the
the transmission
transmission lineline at
at 1s.
1 s. The
The high
high wind
wind speed
speed isis DFIG
set
set at at 15m/s,
1 5m/s, and
and the
the low
low wind
wind speed
speed isis set
set at
at 5m/s.
5m/s. The
The
frequency
frequency of ofthe
the BESS
BESS grid-connected
grid-connected systemsystem with
withthe
the control
control
strategy BESS
strategy proposed
proposed in in this
this paper
paper isis obtained
obtained by by simulation,
simulation, and
and ~ ~
itit isis compared
compared with with the
the frequency
frequency when when the
the BESS
BESS is is not
not Fig.
Fig. 6.
6. Schematic
Schematicdiagram
diagramof the simulation
ofthe simulationsystem
system
connected.
connected. The The frequency
frequency change
change curve
curve isis shown
shown inin Figure
Figure 55 ..
hout - with hout - with Under
Under the
the operating
operating mode mode ofof low
low wind
wind speed
speed ofof 5m/s,
5m/s,
50.2
w ithout w ith w ithout w ith
witBESS
BE SS BE SS 50.2
witBESS
B E SS BE SS
DFIG
DFIG and BESS are connected into the multi-machine system,
and BESS are connected into the multi-machine system,
lvi� "'
BESS BESS
5o.o
N50.0 r----....l., l
and
and the
the three-phase
three-phase grounding
grounding short short circuit
circuit fault
fault occurs
occurs at
at
� ilf'
'R ;.50.1
50 . 1
point
point A
A at
at 2s.
2s. When
When the
the control
control strategy
strategy isis applied
applied to
to BESS,
frequency /Hz
frequency/Hz

BESS,
the
the operation
operation situation
situation isis shown
shown in
in Figure
Figure 7.
l l
7.

50.0 1-----___)
�
22oor---�, n rv-----l
�
50.0
49.9 c,------.';:-'-
.,.- -----,-J
5500.1. 1
weighting coefficient w1

ii" 00

0.6 1 .0 1 .4 0.6
i
49. 9 1l
�
0.6 1.0 1.4 0.6 1.0 1.4
frequency/Hz

tim e/s tim e/s

IL
tirne/s tirne/s
� . 0 ,__ �
8
(a)
(a)Unsymmetric
Unsymmetricfault (b)
(b) Symmetric
Symmetricfault
II vlfv-.-..- ]>1 9999.998s
fault fault
atatlow
low wind
windspeed
speed atatlow
lowwind
wind speed
speed "' 5500.0 __ ____)
-

hout with
w ithout
- witBESS w ith
hout - with
w ithout
- witBESS w ith �
50.1 50.2
B E SS BESS B E SS B E SS BESS B E SS

��
N

ll
50.1

time/s
11.5.5 22.0.0 22.5.5 11.5. 5 22.0.0 22.5. 5
tim e/s
50.2

g�
tim e /s
frequency /Hz
frequency/Hz

tim e/s

.11
�
:I:

� 50.1
50.1
X 1J0O- 7·'
>.
( a ) Frequencyvariation

aFrequency variationcurve ( b ) Weightingcoefficient

bWeighting coefficientww11
X 1105
curve
50 0 � 22.5
&   05
&

50.0 %
Jj50.0 Jj .5
" 22.0
weighting coefficient w2

0
active power/W

50.0

�
.0
0 .6 1 .0 1 .4 0.6 1 .0 1 .4
-�
1€ 11.5.5
0.6 1.0 1.4 0.6 1.0 1.4 &

..§ 11.0.0
tim e/s
tirne/s tim e/s
tirne/s 8
(c) Unsymmetric fault (d) Symmetric fault

1 0.5
(c) Unsymmetric fault (d) Symmetric fault
atathigh
highwind
wind speed
speed atathigh
highwind
windspeed
speed
-3':----
- -----:�:- ------::-'
0 .5

11.5.5 22.0.0 22.5 11.5.5 22.0. 0 22.5.5

Fig. -3
Fig. 5.
5. Frequency
Frequencychange
changecurve
curveof
ofshort-circuit
short-circuitfaults
faults in
insingle-machine
single-machine .5
infinite
infinitesystem
system
tim e/s
tim e/s tim e/s
tim e/s
ItIt can
can bebe seen
seen from
from Figure
Figure 55 that
that when
when unsymmetric
unsymmetric faultfault ( c ) Weightingcoefficient

cWeighting coefficientwwz2 ( d ) Activepower

dActive power
and
and symmetrical fault occur in the middle section of
symmetrical fault occur in the middle section ofthe
the line
line at
at Fig.
Fig. 7.
7. System
Systemfrequency
frequencyand
andbattery
batteryoperating
operatingmode
modeofofshort-circuit
short-circuit
faults
faultsat
atlow
lowwind
windspeed
speed
1s
l s under
under the the operating
operating mode,
mode, the
the frequency
frequency value
value at
at the
the DFIG
DFIG
connection
connection point point fluctuates
fluctuates significantly,
significantly, andand the
the maximum
maximum The
The frequency
frequency change
change curve
curve at
at the
the DFIG
DFIG connection
connection point
point
absolute
absolute frequency
frequency difference
difference when
when BESSBESS is is connected
connected to to the
the isis shown
shown in in Figure
Figure 7a.
7 a. During
During normal
normal operation,
operation, thethe frequency
frequency
grid
grid isis less
less than
than the
the value
value when
when itit is
is not
not connected.
connected. ItIt can
can bebe value
value atat the
the wind
wind turbine
turbine connection
connection point
point isis 50Hz,
50Hz, and
and the
the
analyzed
analyzed that that when
when the
the grid
grid side
side fails,
fails, the
the BESS
BESS applied
applied with
with frequency
frequency fluctuates at 2s due to the system fault, and tends to
fluctuates at 2s due to the system fault, and tends to


2293
Authorized licensed use limited to: University of Edinburgh. Downloaded on March 30,2024 at 11:38:30 UTC from IEEE Xplore. Restrictions apply.
be
be stable
stable after
after 0.2s.
0.2s. The
The adaptive
adaptive condition
condition of ofthe
the weighting
weighting and
and the the maximum
maximum absoluteabsolute frequency
frequency difference
difference isis
coefficient
coefficient of of the
the single
single neuron
neuron PID PID controller
controller inin thethe corresponding
corresponding to to the
the lowest
lowest frequency.
frequency. However,
However, itit can can be
be
frequency
frequency controller
controllerisisshown
shownininFigure
Figure7b7band
andFigure
Figure7c.7c. The
The seen
seen from
from Figure
Figure 8b 8b that
that the
the maximum
maximum frequency
frequency of of grid-
grid­
weighting
weighting coefficient
coefficient isis adjusted
adjusted online
online according
according to to the
the connected
connected system
system withwith the
the control
control strategy
strategy proposed
proposed inin this
this
frequency
frequency fluctuation at 2s, and the single neuron PID
fluctuation at 2s, and the single neuron PID paper
paper is less than the value when the BESS is not
is less than the value when the BESS is not connected,
connected,
controller
controller can
can operate
operate normally.
normally . The The output
output active
active power
power of of and
and the
the minimum
minimum frequency
frequency inin Figure
Figure 8d8d isis greater
greaterthan
than the
the
BESS
BESSis is0MW
OMWatat50Hz50Hzand andititisisshown
shownin inFigure
Figure7d.7d. When
Whenthe the value
value when
when BESS
BESS isis notnot connected,
connected, which
which showsshows that
that the
the
frequency
frequency fluctuates,
fluctuates, thethe power
power support
support cancan bebe quickly
quickly maximum
maximumabsolute
absolutefrequency
frequencydifference
differenceisisreduced
reducedeffectively
effectively
provided
provided by BESS. It can be seen from Figure 77 that
by BESS. It can be seen from Figure that the
the and
and the frequency stability under full wind speedconditions
the frequency stability under full wind speed conditionsisis
battery
battery control
control system
system based
basedon onthethe PID
PID controller
controllerrrcancanwork
work improved
improved by by the
the strategy
strategy proposed
proposed ininthis
this paper.
paper. AtAtthethe same
same
normally
normally in in the
the multi-computer
multi-computer system system and
and participate
participate in
in the
the time,
time,ititcan
canalso
alsobe beseen
seenthat
thatthe
thefrequency
frequencydifference
differencecan canalso
also
response
responseof ofsystem
system frequency
frequencyin inreal
realtime.
time. be
be reduced
reduced by by the
the BESS
BESS with
with droop
droop control
control strategy
strategy when
when
short
short circuit
circuit fault
fault occurs
occurs atat full
full wind
wind speed,
speed, but but the
the
The
Thedroop
droopcontrol
controlstrategy
strategyisissimulated
simulatedand andanalyzed
analyzedininthethe improvement
multi-machine improvement effecteffect isis not
not as
as good
good asas the
the BESS
BESS withwith the
the
multi-machinesystem. system. Under
Underthe theoperating
operatingmode mode of oflow
lowwind
wind control
speed, controlstrategy.
strategy .
speed, the
the three-phase
three-phase grounding
grounding short short circuit
circuit fault
fault occurs
occurs inin
2s,
2s, and
and the
the droop
droop coefficient
coefficient isis set set asas 88 and
and 101 0 respectively,
respectively, In
In Table
Table 1,1 , the
the maximum
maximum absolute
absolute frequency
frequency difference
difference
the
thevariation
variationof offrequency
frequency isisshown
shownininFigureFigure8.8. The Themaximum
maximum symbol
symbol isis defined
defined as as max f . . ItIt can
maxl�fl can bebe seen
seen from
from Table
Table 11
frequency
frequency difference and the maximum absolute frequency
difference and the maximum absolute frequency that
that the
the BESS
BESS frequency
frequency modulation
modulation effect effect isis the
the best
best when
when
difference
difference will will be be effectively
effectively reduced
reduced by by BESS
BESS when when thethe the
thesymmetrical
symmetricalfault faultoccurs
occursatatpoint
pointAAunder underlow lowwind
windspeed.
speed.
droop
droop coefficient
coefficient isis set set to
to 8.8. In
In order
order to to give
give full
full play
play to
to the
the
frequency At
Atthis
thistime,
time, the
the max f of
maxl�fl ofthe
theBESS
BESS grid-connected
grid-connectedsystemsystem
frequencymodulation
modulationability
ability of ofBESS,
BESS,the thedroop
droopcoefficient
coefficientisis
increased
increased to to 10.
10. At
Atthis
this time,
time, thethe maximum
maximum frequency
frequency can canbe be with the control strategy proposed in this paper
with the control strategy proposed in this paper is only 58. 98% is only 58.98%
reduced
reducedin incase
caseof ofshort
shortcircuit
circuitfault,
fault, but
butthethefrequency
frequency cannot
cannot of
ofthe
the max f when
maxl�fl whenthe the BESS
BESS isis not not connected.
connected. Compared
Compared
be
be quickly
quickly stabilized
stabilized to to 5050 HzHz after
after fault.
fault. From
From the the above
above with
withdroop
droopcontrol,
control, thethecontrol
controlstrategy
strategy proposed
proposedin inthis
thispaper
paper
analysis,
analysis, itit can
can bebe seen
seen that
thatthethe selection
selection of ofthe
the coefficient
coefficientof of can
can reduce
reduce thethe maximum
maximum absoluteabsolute frequency
frequency difference
difference by by
the
thedroop
droop link
linkisisvery
very important,
important, and andititisisdifficult
difficultto tomake
makethethe about
about10%1 0%on onaverage.
average. When Whendifferent
differenttypestypesof offaults
faultsoccur
occurinin
optimal
optimal choice
choice due due to
to the
the balance
balance between
between the the control
control effect
effect the
thesystem
systematatfull
fullwind
windspeed,
speed,the
theactive
activepower
powersupport
supportcancanbebe
and
and the
the stable
stable operation.
operation. According
According to to the
the above
above simulation
simulation quickly
quickly provided
provided by by the
the BESS
BESS applied
applied inin this
this strategy,
strategy, the
the
results,
results, the
the fixed
fixed droop
droop coefficient
coefficient isis 9,9, which
which can can give
give full
full
play max f l of
maxl�f ofthe
the system
system can can be
be effectively
effectively reduced
reduced , , and
and the
the
play toto the
the BESS
BESS frequency
frequency modulation
modulation function
function underunder the
the
premise
premiseof ofensuring
ensuringstable
stableoperation.
operation. Then Thenititisiscompared
comparedwith with frequency
frequency stability
stability ofofthethe wind
wind power
power grid-connected
grid-connected system
system
the
thecontrol
controlstrategy
strategyproposed
proposedin inthis
thispaper.
paper. can
canbe beimproved.
improved. And Andthe theeffectiveness
effectivenessof ofthe
thecontrol
controlstrategy
strategy
isis verified.
verified. InIn addition,
addition, compared
compared with with traditional
traditional droop
droop
BESS - kkr
without BESS
-without f ==88 - kkr
f ==10
lO control, the max 
control, the maxl�fl can be reduced more effectively and
f can be reduced more effectively andthethe

�
� 50.1
system
systemfrequency
frequencystability
stabilitycancanbebefurther
furtherimproved.
improved.
frequency/Hz

50. 1

I _
without -kk
f=9 -
new r-without
without -kk - new
new
without new
rg strategy rg
f=9
strategy BESS strategy

50.0
BESS

r-------...J
BESS strategy

�50.1
50.1
�
50.0
50.0
50.0
frequency/Hz

frequency /Hz

1.5 2.0 2.5

l i It
�w·vm�"'"---�
1 .5
time/s

l50.0 I�
time/s

50.0
Fig.
Fig.8.8. Frequency
Frequencycondition
conditionwhen
whenusing
usingdroop
droopcontrol
controlstrategy 49.9
strategy Jl49.9

20 2.1
2.1 2.2
22 2.3
23 2.4
24 2.5
25
The
The fault
faultoccurs
occursatatpoint
pointAAofofthe
themulti-machine
multi-machinesystem
systematat
2s,
2s, and
and the
the fault
fault types
types are
are BC
BC two-phase
two-phase shortshort circuit
circuit and
and 2.0 20 2.1
2.0 21 2.2
22 2.3
23 2.4
2A 2.5
2.5
three-phase time/s
time/s time/s
three-phasegrounding
grounding short
shortcircuit.
circuit. The
Thesimulation
simulationanalysis
analysisisis
time/s
(a)
(a)Unsymmetric
Unsymmetricfault
fault (b)
(b)Symmetric
Symmetricfault
fault
carried
carried out
out under
under thethe operating
operating mode
mode of ofhigh
high wind
wind speed
speed of
of atatlow
lowwind
windspeed
speed atatlow
lowwind
wiodspeed
speed
15m/s
1 5m/sandandlow
lowwind
windspeed
speedofof5m/s.
5m/s. TheThefrequency
frequencyof ofdifferent
different
without -kk - without _kk _new
new
- without r9 new r9
new without
f=9 f=9
types
types ofoffaults
faults under
underdifferent
differentwind
windspeeds
speedsisisshown
shownininFigure
Figure BESS
BESS strategy
strategy BESS
BESS Strategy
Strategy

50.0
9,9, and
andthe
themaximum
maximum absolute
absolutefrequency
frequency difference
differenceunder
undereach
�50.0
each
operating
operatingmodemodeisisshown
shownin inTable
Table1.1 .
frequency/Hz

frequency/Hz

[
50.0

ItIt can
can be
be seen
seen from
from Figure
Figure 99 that
that when
when the the same
same type
type ofof �
fault
fault occurs
occurs atat point
point AA under under different
different wind
wind speeds,
speeds, thethe Jl
49.9
fluctuation
fluctuation of of frequency
frequency isis obviously
obviously different,
different, andand the
the time
time 49.9
49.9

2� 2.1
2.1 2.2
22 2.3
23 2.4
24 2.5
25 20 2.1
21 2.2
22 2.3
23 2.4
24 2.5
25
point
point atat which
which the the maximum
maximum absolute
absolute frequency
frequency difference
difference 2.0 2.0
occurs
occurs isis also
also different.
different. TheThe frequency
frequency change
change of ofthree-phase
three-phase time/s
time/s
time/s
time/s
grounding
grounding shortshort circuit
circuit atat low
low wind
windspeed
speed isis shown
shown in in Figure
Figure (c)
(c)Unsymmetric
Unsymmetricfault
fault (d)
(d)Symmetric
Symmetricfault
fault
8b,
8b, andand the
the maximum
maximum absolute absolute frequency
frequency difference
difference isis atathigh
highwind
windspeed
speed atathigh
highwind
wiodspeed
speed
corresponding Fig.
Fig.9.9. Frequency
Frequencycondition
conditionofofshort-circuit
short-circuitfaults
faultsininmulti-machine
multi-machine
corresponding to to the
the highest
highest frequency.
frequency . While
While the
the frequency
frequency system
system
of
ofthe
the same
same fault
fault atat high
high wind
wind speed
speed isis shown
shown in in Figure
Figure 8d,8d,


2294
Authorized licensed use limited to: University of Edinburgh. Downloaded on March 30,2024 at 11:38:30 UTC from IEEE Xplore. Restrictions apply.
 TABLE I.I.
TABLE THE MAXIMUM
THE MAXIMUM ABSOLUTE
ABSOLUTE FREQUENCY
FREQUENCY DIFFERENCE
DIFFERENCE VALUE
VALUE 3) Compared with with the
the droop
droop control,
control, the
the system
MULTI-MACHINE SYSTEM
OF MULTI-MACHINE
3) Compared system
OF SYSTEM
frequency stability can be better improved
frequency stability can be better improved by the by the control
control
strategy proposed in this paper.
maxl4f1
max f 1Hz
operating
operating strategy proposed in this paper.
composition
composition f /HZ
f!HZ /HZ percentage /%
percentage /%
mode
mode
Without
Without 49.8772 0.1228 100
References
References
Unsymmetric
Unsymmetric BESS
49.8772 0.1228 100
BESS [1]
[1] S. Huang,
S. Huang, T. T. JiJi and
and T.T. Wang,
Wang, "Transfer
"Transfer function
function sensitivity
sensitivity based
based DFIG
DFIG
fault at
fault at low
low shedding scheme
shedding scheme to to mitigate
mitigate subsynchronous oscillation," 2021
subsynchronous oscillation," 2021 IEEE
IEEE
kf=9
kf�9 49.8962
49.8962 0.1038
0. 1038 84.53
84.53
wind speed
wind speed 4th International
4th International Electrical
Electrical and and Energy
Energy Conference
Conference (CIEEC),
(CIEEC), pp. pp. 1-1-
New strategy
New strategy 49.9044
49.9044 0.0956
0.0956 77.85
77.85 6,2021..
6,2021
Without
Without [2]
[2] B. Wang,
B. Wang, Z. Z. Xie
Xie and
and L. L. PENG,
PENG, "Input
"Input admittance
admittance modeling
modeling and and stability
stability
Symmetric 50.1570
50. 1570 0.1570
0.1570 100
100 analysis of of DFIG
DFIG under
under different
different modes,"
modes," 2022 IEEE 5th 5th International
International
Symmetric BESS
BESS analysis 2022 IEEE
fault at
fault at low
low Electrical
Electrical andand Energy
Energy Conference
Conference (CIEEC),
(CIEEC), pp. pp. 1518-1523,2022.
1 5 1 8- 1523,2022.
kf=9
kf�9 50.1133
50. 1 133 0.1133
0.1133 72.17
72. 17
wind speed
wind speed [3]
[3] R. Chai and W. Li, "Research
R. Chai and W. Li, "Research on wind-storage joint on wind-storage joint frequency
frequency
New strategy
New strategy 50.0926
50.0926 0.0926
0.0926 58.98
58.98 modulation control
modulation control strategy
strategy basedbased on on DFIG
DFIG frequency
frequency modulation
modulation
Without capability," 2021
capability," 2021 IEEEIEEE 4th 4th International
International Electrical
Electrical and and Energy
Energy
Without 49.9007 0.0993 100 Conference (CIEEC),
(CIEEC), pp. pp. 1-8, 2021..
Unsymmetric
Unsymmetric 49.9007 0.0993 100 Conference 1-8, 2021
BESS
BESS
fault at
at high
high [4]
[4] Z. Xue
Z. Xue and
and B. B. Dong,
Dong, "Design
"Design and and implementation
implementation of of monitoring
monitoring and and
fault kf=9 49.9189
49.9189 0.0811
0.0811 81.67
81.67
kf�9 management system
management system for for battery
battery energy
energy storage,"
storage," 2022 2022 IEEE
IEEE 5th 5th
wind speed
wind speed
New strategy
strategy 49.9275
49.9275 0.0725
0.0725 73.01
73.01 International Electrical
International Electrical and and Energy
Energy Conference
Conference (CIEEC),
(CIEEC), pp. pp. 489-493,
489-493,
New
2022.
2022.
Without
Without
Symmetric 49.8722
49.8722 0.1278
0.1278 100
100 [5]
[5] Y. Zhang,
Y. Zhang, J. J. Jiao
Jiao andand D. D. Xu,Xu, "Direct power control
"Direct power control of of doubly
doubly fedfed
Symmetric BESS
BESS induction generator
generator using using extended
extended power
power theory
theory under
under unbalanced
unbalanced
induction
fault at high
fault at high network," IEEEIEEE Transactions
Transactions on on Power
Power Electronics,
Electronics, vol.vol. 34, no. 12, pp.
kf=9
kf�9 49.8883
49.8883 0.1117
0.1117 87.40
87.40 network," 34, no. 12, pp.
wind speed
wind speed 12024-12037,
12024-12037, Dec. Dec. 202019.
19.
New strategy
New strategy 49.8969
49.8969 0.10311
0.103 80.67
80.67 [6]
[6] N. Harag,
N. Harag, M. M. Imanaka,
Imanaka, M. M. Kurimoto,
Kurimoto, S. Sugimoto, H.
S. Sugimoto, H. Bevrani
Bevrani andand T.T.
Kato, "Autonomous
Kato, "Autonomous Dual Dual Active
Active Power-frequency
Power-frequency Control Control in in Power
Power
V. CONCLUSION
V. CONCLUSION System with
System with Small-scale
Small-scale Photovoltaic
Photovoltaic Power Power Generation,"
Generation," Journal
Journal of of
Modern Power
Modern Power Systems
Systems and and Clean
Clean Energy,
Energy, vol.
vol. 10, no. 4,
10, no. 4, pp.
pp. 941-953,
941-953,
After DFIG
After DFIG is is connected
connected to to the
the grid,
grid, the
the system
system July.2022.
July.2022.
frequency stability
frequency stability will
will be
be greatly
greatly affected
affected by by the
the DFIG,
DFIG, and
and [7] U. Datta,
Datta, A. A. Kalam
Kalam and and J. J. Shi,
Shi, "Battery
"Battery energy
energy storage
storage system to
[7] U. system to
this impact
this impact willwill be
be reduced
reduced by by BESS.
BESS. InIn order
order to
to improve
improve thethe stabilize transient
stabilize transient voltage
voltage and and frequency
frequency and and enhance
enhance powerpower export
export
frequency stability
frequency stability ofof wind
wind power
power grid-connected
grid-connected system,
system, aa capability," IEEE
capability," IEEE Transactions
Transactions on on Power
Power Systems,
Systems, vol. vol. 34,
34, no.
no. 3, pp.
3, pp.
control strategy
strategy applied
applied toto BESS
BESS based
based onon single
single neuron
neuron PID
PID 1845-1857, May.2019.
1 845-1 857, May.20 19.
control
is proposed
proposed in in this
this paper,
paper, and and the
the simulation
simulation modelmodel for
for [8]
[8] Z. Li,
Z. Li, K.
K. W.W. Chan,
Chan, J. J. Hu
Hu and and J.J. M.
M. Guerrero,
Guerrero, "adaptive
"adaptive droopdroop control
control
is using adaptive
adaptive virtual
virtual impedance
impedance for for microgrids
microgrids with with variable
variable PV PV
using
analysis and verification is built. The following
analysis and verification is built. The following conclusions conclusions outputs and
outputs and load
load demands,"
demands," IEEE IEEE Transactions
Transactions on on Industrial Electronics,
Industrial Electronics,
can be
can be drawn:
drawn: vol. 68,
vol. 68, no.
no. 10, pp. 9630-9640,
10, pp. 9630-9640, Oct. Oct. 202
2021.1.
1) In case
1 ) In case of
of system
system failure,
failure, the
the frequency
frequency emergency
emergency [9]
[9] J. Liu, W. Yao, J. Wen, "A
J. Liu, W. Yao, J. Wen, "A virtual inertia virtual inertia compensation
compensation strategy
strategy forfor
control can
control can be be realized
realized byby thethe BESS
BESS applying
applying thethe control
control wind farms
wind farms basedbased on on energy
energy storage technology," Proceedings
storage technology," Proceedings of the
of the
CSEE, vol.
CSEE, vol. 35,
35, no.
no. 7,
7, pp.
pp. 1596-1605, Apr.2015.
1 596-1605, Apr.20 15.
strategy in
strategy in this
this paper,
paper, thethe maximum
maximum absolute
absolute frequency
frequency
[10]
[10] K.. Ang,
K Ang, G.G. Chong
Chong and and Y.Y. Li,
Li, "PID control system
"PID control analysis, design,
system analysis, design, and
and
difference can
difference can bebe reduced
reduced effectively,
effectively, andand the
the frequency
frequency technology," IEEE IEEE Transactions
Transactions on on Control
Control Systems Technology, vol. vol.
technology," Systems Technology,
stability of
stability of wind
wind power
power grid-connected
grid-connected system system can can bebe 13, no. 4,
13, no. 4, pp.
pp. 559-576,
559-576, July.2005.
July.2005.
improved.
improved. [11]
[11] B. Zhang,
B. Zhang, K. K. Zhou,
Zhou, Y. Y. Wang,
Wang, et et a!,
al, "Performance
"Performance improvement
improvement of of
2) The
2) The control
control strategy
strategy proposed
proposed inin this
this paper
paper isis effective
effective repetitive controlled
repetitive controlled PWM PWM inverters:
inverters: A A phase-lead
phase-lead compensation
compensation
at full wind speed. solution," International
solution," International Journal
Journal of of Circuit
Circuit Theory
Theory andand Applications, 2010,
Applications, 2010,
at full wind speed. 38(5): 453-469.
453-469.
38(5):


2295
Authorized licensed use limited to: University of Edinburgh. Downloaded on March 30,2024 at 11:38:30 UTC from IEEE Xplore. Restrictions apply.