See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/304021959

Fuzzy logic controller based STATCOM for voltage profile improvement in a

micro-grid

Conference Paper · April 2016

DOI: 10.1109/SYSCON.2016.7490645

CITATIONS READS
12 156

2 authors:

Kanungo Mohanty Swagat Pati

National Institute of Technology Rourkela ITER - India
179 PUBLICATIONS 2,277 CITATIONS 79 PUBLICATIONS 711 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Kanungo Mohanty on 24 January 2023.

The user has requested enhancement of the downloaded file.

 Fuzzy Logic Controller based STATCOM for
Voltage Profile Improvement in a Micro-Grid

Kanungo Barada Mohanty Swagat Pati

Dept. of Electrical Engineering Dept. of Electrical Engineering, I.T.E.R
National Institute of Technology Rourkela SOA University
Rourkela, India Bhubaneswar, India
[email protected] [email protected]

Abstract—Microgrids constitute a very essential part in today’s During reactive power compensation normally STATCOM
power system for fulfilling the increasing power demand but gives faster response than SVC [3]. STATCOM supplies
when working under isolated conditions microgrids become very additional leading/lagging Var to the system to make the
vulnerable due to the fact that normally microgrids have small voltage constant at a given bus.
generation capacities and most of the generation sources are
renewable/constant power sources, which reduces the transient A lot of research has been already done in this area.
as well as voltage stability limits of a microgrid when working Performance of STATCOM depends on accuracy and
under isolated conditions. In this work a single synchronous robustness of the controller [4], [5]. Various types of
generator based isolated microgrid system is studied. The controllers are used for the power control process in
microgrid system feeds power to a concentrated time varying STATCOM. Basically conventional PI controllers are used.
load. Fluctuation in load causes variation in the load bus voltage But now in recent days Fuzzy Logic Controller (FLC) [6],
which reduces the stability of the system. A Static Synchronous Sliding Mode Controller (SMC) [7] are also used for this
Compensator (STATCOM) is used to improve the voltage profile purpose. Stability margin of DFIG based wind farm is
of the load bus. Two different controllers: PI and fuzzy logic are improved in [8] using a PID damping controller and a hybrid
used for the control purpose of the STATCOM. The performance PID plus fuzzy logic controller of the STATCOM. Both
of the STATCOM with both the controllers is evaluated with frequency domain approach and time domain approach are
different load conditions: R-L load, nonlinear load and dynamic
used for modeling the system. Fuzzy logic approach is also
load. The results are analyzed and compared. The simulation is
used in [9] to determine optimal placement of the
done in Matlab/Simulink.
DSTATCOM and photo-voltaic array in a distribution system.
Keywords- STATCOM, Decoupled Control, Fuzzy Logic Fuzzy logic is also used in [10] for modeling the uncertainty
Controller, PI controller, Micro-grid, Reactive power compensation of loads, distributed generations, and purchased power from
the grid. By considering an uncertain environment, the effect
on decreasing the risk of technical issues like active loss and
I. INTRODUCTION voltage profile in distribution network is evaluated.
Over the last two decades demand of electricity has
increased a lot along with the increased population. To meet In this work a single alternator based isolated micro-grid
this increased electricity demand the ultimate solution is system is considered. The generator feeds power to a
formation of micro-grid. Micro-grid consists of small concentrated time varying load. A STATCOM is utilized to
generating stations combined with some local load. For improve the voltage profile at the load bus. The performance
generation of electricity renewable energy sources are also of the STATCOM is evaluated with PI and Fuzzy logic
used, whose power is variable in nature. Along with this the controllers and the performance comparison of the both the
micro-grid has one demerit that voltage profile is highly controllers is done with different load conditions. Three
distorted due to small generating capacities along with different loads have been considered in this work: R-L load,
fluctuating loads [1]. These fluctuations in load affect the nonlinear load and dynamic load. The micro-grid and loads
quality of power as well as the voltage profile. So maintaining are modeled and simulated using Matlab/Simulink. The results
a flat voltage profile with fluctuating power demand is a big are presented in this paper.
challenge for power engineers. Implementation of Flexible AC
Transmission Systems (FACTS) devices is the most feasible II. SYSTEM DESCRIPTION
solution for the above mentioned problem. Among all the
Fig. 1 shows the schematic diagram of the micro-grid
FACTS devices such as Static Var Compensator (SVC), Static
system. A single generator connected to a linear RL load
Synchronous Compensator (STATCOM), Static Synchronous
through a transmission line. This system is verified under
Series Compensator (SSSC) etc. [2], STATCOM is often used
several load condition and load changes. Due to these load
to improve the voltage profile in micro grids.
changes system’s voltage profile is also affected, when load

978-1-4673-9519-9/16/$31.00 ©2016 IEEE

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA. Downloaded on January 24,2023 at 16:25:51 UTC from IEEE Xplore. Restrictions apply.
 increases a voltage dip appears, when load decreases a voltage B. Modeling of AVR
swell appears. But as the load changes, maintaining the profile
is a difficult challenge. For that reason a STATCOM is Due to continuous fluctuation of load, voltage profile at the
connected at the load bus, which improves the voltage profile load bus as well as the generator bus are distorted. To maintain
at the load bus. a constant voltage at the generator bus automatic voltage
regulator (AVR) is used [12]. The model of the AVR is given
in Fig.2.

Fig. 1 Block diagram of micro grid system

Fig. 2 AVR block diagram

III. SYSTEM MODELLING

C. Modelling of speed governor and diesel engine
The generating station considered in this paper consists of
three major parts. In regulating the speed, Speed governor plays a very
crucial role and by this the frequency is also kept constant at
1. Alternator the reference value. According to the load and speed,
2. Automatic voltage regulator (AVR) frequency also varies. So, by using the speed governor, the
speed and hence the frequency of the micro grid can be made
3. Speed governor & Diesel engine constant [13]. The block diagram is given below in Fig.3.

A. Modelling of Alternator
In this work a salient pole type synchronous machine is
considered. The modelling of the alternator [11] is done in
synchronously rotating d e - q e reference frame. The modelling
is done according to the following equations

dI ds dI fr dI fr
V fr = R fr I fr − Lmd + (Llfr + Lmd ) + Lmd Fig. 3 Block diagram of speed governor
dt dt dt (1)
dI dI fr dI
0 = Rdr I dr − Lmd ds + Lmd + Ldr dr (2) IV. CONTROL OF STATCOM
dt dt dt
dI qs diqr To maintain a flat voltage profile at the load bus
0 = Rqr I qr − Lmq + Lqr STATCOM is used. Control structure of the STATCOM is
dt dt (3)
shown in Fig. 4. Here STATCOM is used at the load bus. A
dψ d hysteresis current controller based control strategy is used for
Vd = − Rs I d − ωψ q +
dt (4) the decoupled control of active and reactive power of
dψ q STATCOM. In this control strategy two control loops are used
Vq = − Rs I q + ωψ d + for control of active and reactive power independently.
dt (5)
First the voltage at the load bus and current are sensed and
Flux linkage expressions in (4) and (5) are as follows. the peak of load bus voltage is calculated. Then the in phase
component are calculated by dividing the three phase voltages
ψ d = − L ds I ds + L md (I fr + I dr ) by the peak value. Then by the transformation given in (9)-
(6)
(11), the quadrature voltages are calculated.
ψ q = − Lqs I qs + Lmq I qr
(7)
Expression for electro-magnetic torque developed is given wa = −ub / 3 + uc / 3
(9)
below. wb = 3ua / 2 + (ub − uc ) / 2 3 (10)
3P
Te = (ψ ds I qs −ψ qs I ds ) wc = − 3ua / 2 + (ub − uc ) / 2 3
22 (8) (11)

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA. Downloaded on January 24,2023 at 16:25:51 UTC from IEEE Xplore. Restrictions apply.
 Fig. 4. Control structure of STATCOM

The in phase components of the STATCOM currents are are calculated and are normalized. The normalized value of
given in (12). ‘eVt’ and ‘ceVt’ are given to the FLC as inputs. The control
output u' is again multiplied with output gain factor ‘Kop’ to
iua = iu max u a
get proportional –derivative action and integral of the output u'
iub = iu max u b (12) is multiplied with another gain Koi to get proportional-integral
iuc = iu max u c action. Both the values are then added up to get a combined
The quadrature components of the STATCOM currents are proportional-integral-derivative action from FLC.
given in (13). The structure of the fuzzy logic controller is shown in Fig.
iwa = iw max wa 5(a). The input and output membership functions are shown in
Fig. 5(b). The input and output fuzzy sets use both triangular
iwb = iw max wb (13)
and trapezoidal membership function, which are found to be
iwc = iw max wc near optimum. The rule base for the FLC is shown in Table-I.
Then the reference values of the STATCOM currents are
calculated as given in (14).

i*a = iua + iwa

ib* = iub + iwb (14)
i*c = iuc + iwc

The control strategy is given below. (a)

A. Structure of PI controller
The PI controllers used in this work has the following
structure.

u = k p e + k i ³ edt
(15)
(b)
Where ‘kp’ and ‘ki’ are the proportional and integral gain, ‘u’
is the control output and ‘e’ is the error input to the controller. Fig.5 Structure and membership functions of Fuzzy logic Controller (a)
Structure of FLC (b) Input and output membership functions of FLC

B. Structure of Fuzzy logic Controller

The Fuzzy logic controller is used in the reactive power
control loop of STATCOM to control the peak voltage. The
error and rate of change of error, ‘eVt’ and ‘ceVt’ respectively

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA. Downloaded on January 24,2023 at 16:25:51 UTC from IEEE Xplore. Restrictions apply.
 4
TABLE I. RULE BASE FOR FLC x 10
with PI
ce 8
nh n z p ph with FLC
e

Reactive power of Load

nh nh nh nh n z 7
n nh nh n z p
6
z nh n z p ph
p n z p ph ph 5
ph z p ph ph ph
4
V. SIMULATION RESULTS AND DISCUSSION
5 10 15 20
In this work the performance of the STATCOM is Time in seconds
analyzed under different load conditions. First the micro-grid Fig. 7. Reactive power of linear load
is subjected to a linear RL load where the load is varied in
steps and performance of the STATCOM is evaluated, using The performance of the STATCOM is evaluated with non
both PI and Fuzzy logic controller and the results are linear load. Fig. 11 shows the load bus voltage when the
compared. A step increase in load is done at 10 s and at 15 s system is subjected to nonlinear load variations. Fig. 12 shows
the load regains its previous value as shown in Fig. 6 and Fig. the reactive power compensated by the STATCOM. As the
7. Both the active and reactive power of the load increases. load bus voltage with PI controller is less than that with fuzzy
Due to the load change the variation in the load bus voltage is logic controller, the reactive power compensated with fuzzy
shown in Fig. 8. It can be clearly noted from the Fig. 8 that the logic controller is more than that with PI controller.
deviation due to load change is more with PI controller where
as there is no such deviation present with fuzzy logic 324
controller. Other than that there is some steady state error
present with PI controller which is zero in case of fuzzy logic
323
controller. Fig. 9 shows the dc link voltage which is perfectly
Terminal voltage of

maintained at 1000 V. The flickering in the dc link voltage

Load bus

increases with the increase in load as seen from the Fig. 9. The 322
reactive power compensated by the STATCOM is shown in PI
Fig. 10.The reactive power compensated with PI controller is 321 FUZZY
more than that with fuzzy logic controller due to the fact that
the voltage maintained by the STATCOM with PI controller is 320
more than that with fuzzy logic controller.
319
5 5 6 7 8 9 10 11 12
x 10
1.65 Time in seconds
1.6 Fig. 8. Terminal voltage of Load bus with linear load
1.55
Active power of Load

1010
With PI
1.5 1008
With FLC
1006
1.45
Dc link voltage of STATCOM

1004
1.4
1002
1.35 With PI 1000
With FLC
1.3 998

996
1.25
5 10 15 20 994
Time in seconds
992
Fig. 6. Active power of linear load 990
5 10 15 20
Time in seconds

Fig. 9. DC link voltage of STATCOM

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA. Downloaded on January 24,2023 at 16:25:51 UTC from IEEE Xplore. Restrictions apply.
 4 4
x 10 x 10
0
-1.5
-1

-2

Reactive Power of
-2
Reactive power of

STATCOM
STATCOM

-3

-2.5
-4
With PI With PI
with FLC
-5 With FLC
-3
-6

-7
5 10 15 20
-3.5
Time in seconds

Fig. 10. Reactive power of STATCOM with linear load

-4
4 6 8 10 12
The performance of STATCOM when subjected to Time in seconds
dynamic load is discussed in this section. The dynamic load
considered in this work is a squirrel cage induction motor. The Fig. 12. Reactive power STATCOM with nonlinear load
motor is started at time 7 s and at 10 s the motor is subjected
to a load torque of 70 N.m. Fig. 13 shows the load bus voltage 326

Terminal voltage of STATCOM

during dynamic load variations. The steady state error
overshoot, undershoot and settling time are more with PI 324
controllers as compared to fuzzy logic controller. With all the
evaluation results and analysis it is clear that fuzzy logic 322
controller shows robust performance as compared to PI
controllers. The performance of the controllers is compared 320
and given in Table-II, III and IV.
318
336
335.5
316
With PI
335 314 With FLC
Terminal voltage of

334.5
STATCOM

334
7 8 9 10
333.5
Time in seconds
333
Fig. 13. Terminal voltage of STATCOM with Dynamic load
332.5
With PI
332 With FLC
5
5 6 7 8 9 10 11 12 x 10
Time in seconds
0
Fig. 11. Terminal voltage of STATCOM with nonlinear load

-0.5
Reactive power of
STATCOM

-1

-1.5

With PI
-2
With FLC

-2.5
7 8 9 10
Time in seconds
Fig. 14. Reactive power of STATCOM with Dynamic load

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA. Downloaded on January 24,2023 at 16:25:51 UTC from IEEE Xplore. Restrictions apply.
 TABLE II. PERFORMANCE DETAILS WITH LINEAR LOAD

Controller Vt in volts Steady state error Deviation in volts Overshoot Under shoot In
in volts In volts volts
Before load change After load change
PI 320 3.74 2.35 1.39 0.28 0.25
FLC 320 0 0 0 0 0.13

TABLE III. PERFORMANCE DETAILS WITH NONLINEAR LOAD

Controller Steady state error Deviation in volts Overshoot Under shoot In

in volts In volts volts
Before load change After load change
PI 1.8 2 0.2 0.12 0.1
FLC 0.1 0.4 0.3 0.15 0.05

TABLE IV. PERFORMANCE DETAILS WITH DYNAMIC LOAD

Controller During starting After loading

Steady state overshoot Undershoot Settling time Steady state overshoot Undershoot
error error
PI 3.5 3.19 10.62 1.1 2.4 0.26 0.17
FLC 0 0.26 5.9 0.13 0 0.15 0.23

[4] B. Singh, S. S. Murthy, S. Gupta, “STATCOM-based voltage regulator

VI. CONCLUSION for self excited induction generator feeding nonlinear loads,” IEEE
Trans. Industrial Electronics, vol. 53, pp. 1437-1452, Oct. 2006.
The voltage profile of the single alternator based radial [5] B. Singh, S. S. Murthy, S. Gupta, “Analysis and design of STATCOM
microgrid system is improved using a STATCOM. The based voltage regulator for self excited induction generators” IEEE
controllers used for the control of STATCOM i.e. PI and Trans. Energy Conversion, vol. 19, pp. 783-790, Dec. 2004.
fuzzy logic controller are designed in this work. The [6] R. K. Mudi, N. R. Pal, “A robust self tuning scheme for PI and PD type
performances of both the controller are evaluated under three fuzzy controllers,” IEEE Trans. Fuzzy Systems, vol. 7, pp. 2-16, Feb.
different load conditions such as linear RL load, nonlinear 1999.
load and dynamic load. It is found that the steady state error is [7] A. Susperregui, M. I. Martinez, I. Zubia, G. Tapia, “Design and tuning
of fixed-switching-frequency second-order sliding-mode controller for
more with PI controller where as fuzzy logic controller gives doubly fed induction generator power control,” IET Electric Power
accurate results without any steady state error. Again the Applications, vol. 6, pp. 696–706, Sept. 2012.
overshoots, undershoots and settling time are also less with [8] Li Wang, D.-N. Truong, “Stability enhancement of DFIG-based
fuzzy logic controllers. The voltage deviation due to load offshore wind farm fed to a multi-machine system using a STATCOM,”
change with fuzzy logic controllers is much less than IEEE Trans. Power Systems, vol. 28, pp. 2882-2889, Aug. 2013.
conventional PI controllers. These all performances prove that [9] H. T. Bagheri, M. H. Ali, M. Rizwan, “Simultaneous reconfiguration,
fuzzy logic controller gives robust performance as compared optimal placement of DSTATCOM, and photovoltaic array in a
to conventional PI controller in all types of load conditions. distribution system based on fuzzy-ACO approach,” IEEE Trans.
Sustainable Energy,” vol. 6, pp. 210-218, Jan. 2015.
[10] S. Abapour, K. Zare, B. M. Ivatloo, “Evaluation of technical risks in
REFERENCES distribution network along with distributed generation based on active
[1] Y. Uno, G. Fujita, M. Matubara, T. Tsukui, R. Yokoyama, T. management,” IET Generation, Transmission and Distribution, vol. 8,
Toyoshima, “Evaluation of micro-grid supply and demand stability for pp. 609-618, 2014.
different interconnections,.” in First IEEE Int. Power and Energy Conf. [11] I. Jadric, D. Borojevic, M. Jadric, “Modeling and control of a
Procc., 28-29 Nov. 2006, Putrajaya, Malaysia, pp. 611-617. synchronous generator with an active DC load,” IEEE Trans. Power
[2] N. G. Hingorani, L. Gyugyi, Understanding FACTS-Concepts and Electronics, vol. 15, pp. 303-311, March 2000.
Technology of Flexible AC transmission Systems, Wiley-IEEE Press: [12] W. Liu, C. Zhang, B. Yuan, “AVR theory, techniques and application,”
New York, 2000, pp. 179-260. in 5th Int. Conf. on Signal Processing Procc., 21-25 Aug. 2000, Beijing,
[3] M. Molinas, J. A. Suul, T. Undeland, “Low voltage ride through of wind China, vol. 2, pp. 1163-1166.
farms with cage generators: STATCOM versus SVC,” IEEE Trans. [13] K. E. Yeager, J. R. Willis, “Modeling of emergency diesel generators in
Power Electronics, vol. 23, pp 1104-1117, May 2008. an 800 megawatt nuclear power plant,” IEEE Trans. Energy Conversion,
vol. 8, pp. 433-441, Sept. 1993.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA. Downloaded on January 24,2023 at 16:25:51 UTC from IEEE Xplore. Restrictions apply.
View publication stats