The 4th International Power Engineering and Optimization Conf.

(PEOCO2010), Shah Alam, Selangor, MALAYSIA: 23-24 June 2010

Comparative Study of Fuzzy Logic Controller and

Proportional Integral Derivative Controller on
DC-DC Buck Converter
1

M.J Yusoff, 2N.F Nik Ismail, Member IEEE, 3 Ismail Musirin, Member IEEE, 4N. Hashim, Member
IEEE, 5 D. Johari

regulator provides non-isolated, switched-mode dc-dc

conversion with accepts a DC input and uses pulse-width
modulation (PWM) of switching frequency to control the
output of an internal power MOSFET. An external diode,
together with external inductor and output capacitor,
produce the regulated dc output [3].

Abstract The main objective of this document is to compare

the performance between Fuzzy Logic controller (FLC) and
Proportional Integral Derivative controller (PIDC) in
improving the performance of DC/DC Buck Converters. The
evaluation of the output has been carried out and compared by
software simulation using MATLAB Simulink. Fuzzy logic
controller has been implemented to the system by developing
fuzzy logic control algorithm. The signals will be processed
based on the fuzzy logic rules-based and PID algorithm.

It is class of switching-mode power supply (SMPS)

containing at least two semiconductor switches (a diode and
a transistor) and at least one energy storage element. Filters
made of inductor and capacitor combinations are often

Keyword Fuzzy Logic Controller (FLC), Proportional Integral

Derivative Controller (PIDC), DC/DC Buck Converter, Simulink
MATLAB.

I.

INTRODUCTION

Power input

DC-to-DC converters have been largely dominated

controlled by analog integrated circuit technology and linear
system design techniques. Recently, with the development
of advanced high-speed digital circuits, digital control will
regularly replace the current used of analog controller in
high frequency switching converters [1]. Even though the
analog controllers are gradually trade, there are some
controllers still being used generally in industrial
application.

vs

Controller

vo

Measurements
Reference

Figure 1.1: Block diagram of Power Electronic System

added to a converters output to improve performance.

Its also called the fly-back converter because the energy
storage transfer, from source to a load, takes place only
during the off period of the switch [4].
The DC/DC converters can operate in two distinct
modes, which are either in Continuous Conduction Mode
(CCM) or Discontinuous Conduction Mode (DCM). The
term continuous and discontinuous is referred to inductor
current. Continuous means the inductor does not reach the
zero value at the end of OFF period. Whereas, the current
goes to zero when works in discontinuous mode. However,
the DC-DC converters that operates in CCM is only
considered for this purposed.

Dc-Dc Buck Converter

The name Buck Converter most probably evolves from

the fact that the input voltage is bucked/chopped or
attenuated, in amplitude and a lower amplitude voltage
appears at the output. In other names, a step-down voltage
______________________________________________

1) Circuit Operation
Figure 1.2 shows the DC-DC Buck Converter circuit
topology. The circuit operation can be divided into two
modes. First mode (mode 1) begins when controlled switch
(e.g MOSFET) is switch on by pulse width modulation
(PWM), the input current, which rises, flows through filter
inductor, L filter capacitor, C and load resistor, R [5].
During mode 1, the diode reversed biased and resulted from

M. Jamhuri can be reached at muhammadjamhuriyusoff@yahoo.com

Nik Fasdi Nik Ismail is with Universiti Teknologi MARA, Shah Alam,
Malaysia. He can be reached at nikfasdi@yahoo.com
3
Ismail Musirin can be reached at t i_musirin@yahoo.co.uk (UiTM)
4
Norazlan Hashim is with Universiti Teknologi MARA, Shah Alam,
Malaysia. He can be reached at azlan4477@yahoo.com.
4
Dalina Johari is with Universiti Teknologi MARA, Shah Alam, Malaysia.
He can be reached at dalinaj@yahoo.com
2

978-1-4244-7128-7/10/$26.00 2010 IEEE

Load
io

Control signals

Task of power electronics is to process and control the

flow of electrical energy by supplying voltages and currents
that suited for optimally loads user. Figure 1.1 shows a
power electronic system block diagram. The output of
power processor is as desired by load. Generally, the
feedback controller compares the output of the power
processor unit with reference value, and the error between
this two is minimized by the controller [2].
A.

is

Power
Processor

142

The 4th International Power Engineering and Optimization Conf. (PEOCO2010), Shah Alam, Selangor, MALAYSIA: 23-24 June 2010

Proportional term (sometimes called gain) makes a

change to the output that is proportional to the current error
value. The contribution from integral term (sometimes
called reset) is proportional to both the magnitude of error
and duration of error. The rate of change of the process error
is calculated by determining the slope of the error over time
and multiplying this rate of change by derivative gain Kd
(sometimes called rate). The PID control scheme is named
after its three correcting terms, whose sum constitutes the
manipulated variable (MV) as shown in equation 2.1.
Hence:

flowing current, the input provides energy to the load as

well as to the inductor [6]. The result is in positive inductor
voltage, VL = Vd - Vo. It causes linear increase in inductor
current.
Mode 2 begins when the controlled switch is switch off.
The freewheeling diode (uncontrolled switch), D conducts
due to energy stored in the inductor, L and the inductor
current continues to flow through inductor, L capacitor, C,
load, and diode, D [5]. During this interval, VL = -Vo for
time duration (1-D)T until the switch is turned on again [6].
The circuit topologies for the converter during both
modes are shows in Figure 1.3. Equation 1.1 shows the
relationship between inductor, L and the linearity in inductor
current.

MV (t )

Pout  I out  Dout

(2.1)

where Pout, Iout, and Dout are the contributions to the

output from the PID controller from each of the three terms.
By tuning the three constant in PID controller algorithm, the
controller can provide control action designed for specific
process requirement [8]. Figure 2.1 shows a block diagram
of PID controller.

P K e(t)
p

Figure 1.2: DC-DC Buck Converter

Setpoint

Error

IKie(W)dW

Process

Out

de

D Kddt(t)
Figure 2.1: A block diagram of PID controller
Figure 1.3(a): Mode 1

Figure 1.3(b): Mode 2

B.

Figure 1.3: The Converters circuit topology during switch is closed

(mode 1) and during switch is opened (mode 2)

v L (t )

iL.

diL
dt
T
1
v L (t ).dt
L 0

Unlike Boolean logic which the state value of any

variable is either 0 or 1, fuzzy logic allows states between
them. More specifically, thats call membership value. The
grade of membership value of fuzzy variable can be
described in linguistic term [9]. Fuzzy major features are the
use of linguistic variables rather than numerical variables.
Linguistic variables are defined as the variables whose
values are define in a usual language (e.g. small and big),
may be represented by fuzzy set.

(1.1)

Equation 1.1: The relationship between inductor and inductor current

II.
A.

Fuzzy Logic (FL) Controller.

The general structure of FLC controller is shows in

Figure 2.2. That comprises four principal components [10]:

METHODOLOGY

Proportional Integral Derivatives (PID) Controller.

1) Fuzzifier
A fuzzyfication interface which converts input data
into suitable linguistic values.

PID control is a traditional linear control method used

commonly in industrial applications. PID letter is stand for
the calculation (algorithm) that involves three separate
parameter, the proportional, the integral and derivatives. A
PID and a PI controller are designed for buck converter for
operation during a start up transient and steady state,
respectively. The derivative term in a PID controller is liable
to noise and measurement error of the system, which could
result in oscillation of the duty cycle during steady state [7].

2) Rule Base and Data Base

Both are known as knowledge base which consists
of data base with necessary linguistic definition and
control rule set.
3) Decision Making
A decisionmaking logic which is simulating a
human decision process, infers the fuzzy control
action from the knowledge of the control rules and
the linguistic variable definition.

143

The 4th International Power Engineering and Optimization Conf. (PEOCO2010), Shah Alam, Selangor, MALAYSIA: 23-24 June 2010

2.3, which are error and change of error as input variables,

and the duty cycle as output variable. Each variables control
has been divided into five partitions. These partition, which
are called as membership function has named into five fuzzy
subsets: PB (Positive Big), PS (Positive Small), ZO (Zero),
NS (Negative Small), And NB (Negative Big). The partition
of fuzzy subsets and the shape of the membership function
are shown in Figure 2.4 (error variable), Figure 2.5 (change
of error variable), and Figure 2.6 (output variable).
The triangular shapes of the membership function of this
arrangement presume that for any particular input there is
only one dominant fuzzy subset. From the combination of
error and change of error, a maximum of four rules are
applied. The rules for the fuzzy controls buck converter are
tabulated in Table 2.1. From the tabulated table, the fuzzy
rule base is formulated into 25 rules.

4) Defuzzifier
A defuzzyfication interface which yields a
nonfuzzy control action from an inferred fuzzy
control action.
The inputs of the fuzzy logic controller are the error e
and the change of error ce, which are defined in equation 2.2
as:

e Vo  Vref
ce

ek  ek 1

(2.2)

where VO is the present output voltage, Vref is the

reference output voltage, and subscript k denotes values
taken at the beginning of the kth switching cycle.
The output of the fuzzy controller is the duty cycle and is
defined in equation 2.3 as:

dk

d k 1  K .Gd k

(2.3)

where dk is the inferred change of duty cycle by the

fuzzy controller at the kth sampling time, and is the gain
factor of the fuzzy controller. Adjusting can change the
effective gain of the controller [10].
Fuzzy Controller
Rule Base

e
ce

Figure 2.4: Membership function of input variable error

Decision
Making

Fuzzifier

d
dt

Defuzzifier

Data Base

Vo

DC/DC
Converter

dk

Vref
Figure 2.2: Basic configuration of FLC
Figure 2.5: Membership function of input variable change of error

Figure 2.3: The editor of fuzzy inference system

The fuzzy logic control for DC-DC buck converter is

made from 2 inputs and 1output variable, as shown in figure

Figure 2.6: Membership function of output variable

144

The 4th International Power Engineering and Optimization Conf. (PEOCO2010), Shah Alam, Selangor, MALAYSIA: 23-24 June 2010
TABLE 2.1: The rule base with 25 rules
TABLE 3.2: Fuzzy logic controller for Buck Converters Parameters
ERROR

CHANGE
ERROR

NB

NS

ZO

PS

PB

NB

NS

ZO

PS

PB

PS

NS

ZO

PS

PB

PS

ZO

ZO

PS

PB

PS

ZO

NS

PS

PB

PS

ZO

NS

NB

PB

PS

ZO

NS

NB

NS

Vin (V)

Vref (V)

Vpeak (V)

OS (%)

Tr (ms)

Ts (ms)

24

14

14.005

0.00664

9.02

9.58

24

13

13.01

0.00769

8.43

9.73

24

12

12.011

0.00167

7.75

9.19

24

11

11.003

0.00636

7.87

8.75

24

10

10.021

0.01996

7.83

8.94

TABLE 3.3: PID controller for Buck Converters Parameters

III.

RESULTS AND DISCUSSION

The simulation has been constructed in Matlab-Simulink.

Table 3.1 shows the tabulated results from open loop control
buck converter simulation test. The results are obtained by
adjusting the duty cycles with respect to different voltage
reference manually. The waveform obtain from the test is
shown in Figure 3.1. The reference of 14V voltage is taken
to represent the other open loops output waveform.

Vref (V)

Vpeak (V)

OS (%)

Tr (ms)

Ts (ms)

24

14

23.505

79.94

1.112

86.1

24

13

21.739

79.87

1.119

85.1

24

12

19.93

79.78

1.117

78.2

24

11

18.124

79.69

1.115

77.1

24

10

16.362

79.6

1.113

73.2

Vref (V)

Vpeak (V)

OS (%)

Tr (ms)

Ts(ms)

24

14

14.081

0.128

13.8

15.9

24

13

13.084

0.069

13.5

15.3

24

12

12.089

0.066

12.4

14.6

24

11

11.08

0.208

11.8

14.9

24

10

10.089

0.02

11.5

14.8

Table 3.4(a) and 3.4(b) show the comparison results of

simulation with 3 different controllers of Buck converter.

TABLE 3.1: Open Loop Buck Converter Parameters

Vin (V)

Vin (V)

From the results obtained, it can be observed that with

FLC, almost 0% of percentage of over shoot and shortest
settling time is obtained compared to PID controller. Where
else, PID controller has minimum 0.1% overshoot and
longer of settling time in average compared to FLC,
Therefore, by comparing these results it can be conclude
that fuzzy logic control gives a better performance in
regulating the output of Buck converter. Thus, Buck
converter with FLC is more efficient by offering better
dynamic performance compared to PIDC.
TABLE 3.4(a): The comparison of the output voltage (Vo) in open loop,
fuzzy control and PID control for Buck converter

Figure 3.1: Output waveform of open loops buck converter for 14Vref

From the results, the value of percentage overshoot

(%OS) and the settling time (Ts) are very high. Taken a 14V
as voltage reference, the duty cycle with the duty cycle
58.3%, a 79.94% of voltage over shoot, and 86ms of settling
time, with 0.932V of deviation voltage are obtained.
Therefore in order to eliminate these problem, the closed
loop controllers are needed for the converter.

Vo,PID
(V)
14.063

Vo,O/L
(V)
13.063

Vin (V)

Vref (V)

D (%)

Vo, FLC(V)

24

14

58.33

14.004

24

13

54.17

13.009

13.075

12.086

24

12

50

12.011

12.076

11.086

24

11

45.83

11.002

11.057

10.086

24

10

41.67

10.019

10.087

9.11

TABLE 3.4 (b): The comparison of voltage deviation (Vd) in open loop,
fuzzy control and PID control for Buck converter

Table 3.2 and Table 3.3 show the results from simulation
test on FLC and PID control, respectively. The comparison
of waveforms obtained from both control topologies are
shown in Figure 3.2 to Figure 3.3 with reference voltage
(Vref), 14V to 13V, respectively.

Vin (V)

Vref (V)

Vd,fuzzy (V)

Vd,PID (V)

24

14

0.0044

0.063

Vd,O/L (V)
0.937

24

13

0.009

0.075

0.914

24

12

0.011

0.076

0.914

24

11

0.0023

0.057

0.914

24

10

0.019

0.087

0.89

The plotted graph for comparison of parameter peak

voltage (Vpeak) vs reference voltage (Vref) parameter
percentage overshoot (OS%) vs voltage reference (Vref)
parameter rise time (Tr) vs voltage reference (Vref) and
parameter settling time (Ts) vs voltage reference (Vref)
145

The 4th International Power Engineering and Optimization Conf. (PEOCO2010), Shah Alam, Selangor, MALAYSIA: 23-24 June 2010

between the two controller are shown in Figure 3.7 to Figure

3.12, respectively.

Figure 3.9: Rise time, Trise (ms) vs reference voltage, Vref of plotted graph
for FLC and PIDC

Figure 3.2 (a): FLC

Figure 3.2 (b): PIDC
Figure 3.2: Vin = 24V, Vref = 14V

Figure 3.10: Settling time, Tsettling (ms) vs reference voltage, Vref of

plotted graph for FLC and PIDC

Figure 3.3 (a): FLC

Figure 3.3 (b): PIDC
Figure 3.3: Vin = 24V, Vref = 13V

Figure 3.11: Output voltage, Vo vs reference voltage, Vref of plotted graph

for open loop, FLC and PIDC of buck converter.

Figure 3.7: Peak voltage, Vpeak vs reference voltage, Vref plotted graph
for FLC and PID
Figure 3.12: Deviation voltage, Vd vs reference voltage, Vref of plotted
graph for open loop, FLC and PIDC of buck converter.

From the waveforms and the plotted graphs obtained in

Figure 3.2 to Figure 3.12, its proved that fuzzy logic control
is better in every aspect of parameter output, including the
voltages deviation and output voltages. Its because fuzzy
logic use the linguistic term which is describing the grade of
membership value of fuzzy variable. More specifically, the
membership value can be defined as the states of any
variable between the 0 and 1 of Boolean logic. Unlike the
PID controller which the state value of any variable is either
0 or 1, fuzzy logic gives more accurate output variables in
order to control the error of buck converter output.

Figure 3.8: Percentage overshoot, OS (%) vs reference voltage, Vref of

plotted graph for FLC and PIDC

146

The 4th International Power Engineering and Optimization Conf. (PEOCO2010), Shah Alam, Selangor, MALAYSIA: 23-24 June 2010

A. Test on Performance of Fuzzy Logic Controller and

PID controller with Step Referenc.
Further test has been measured to observe how well the
controllers contribute to a better performance for DC-DC
buck converter. A step reference is used instead of constant
reference in simulation test and act as a transient signal.
With setting of the voltage drop -0.5V on every test, during
0.02s to 0.025s, the performance of the controller have been
tested measured and compared.

Figure 3.13: The output voltage for fuzzy logic control on buck converter
with step reference applied.

The results obtained are tabulated in Table 3.5 and Table

3.6 for Fuzzy Logic control and PID control, respectively.
Peak voltage (Vp), output voltage (Vo), percentage
overshoot (%OS), rise time (Tr), and settling time (Ts) are
taken after the step reference. While peak voltage drop
(Vpdrop) is taken during the step reference.
Figure 3.13 and Figure 3.14 are the sample waveform
obtained from the step reference test, which are taking a
14V voltage reference and a 13.5V voltage step reference
test as a sample.

Figure 3.14: The output voltage for PID control on buck converter with step
reference applied.

From the waveforms and the results obtained, its show

that both controllers are almost achieving back to the desired
outputs after the step reference voltage is applied. Only
e-3V on output voltages (Vo) has been varied compared to
before a step reference is applied. Even though the output
voltages of PID control is closed in achieved back to the
value before the step happens, there is still has
disadvantages compared to fuzzy logic control. A 14V
reference voltage and 13.5V voltage step reference as a test
sample, PID control gives 0.063V of voltage deviation (Vd),
0.078% of percentage overshoot (%OS), 9ms of rise time
(Tr) and 13ms of settling time (Ts) while fuzzy logic control
has 0.002 of voltage deviation (Vd), 0.049 of percentage
overshoot (%OS), 1.9ms of rise time (Tr) and 3.82ms of
settling time (Ts). In all aspects, including voltage deviation,
percentage overshoot, rise time, settling time, and peak
voltage drop during step, fuzzy logic control still gives the
better performance and more precise output parameters even
the step reference (transient) is applied.

TABLE 3.5: The tabulated results for fuzzy logic control with step
reference voltage applied.
TABLE 3.5(a)
Vp,2nd (V)
Vo,2nd (V)

Vin (V)

Vref,step

24

14-13.5

14.009

14.002

0.049

24

13-12.5

13.013

13.005

0.062

24

12-11.5

12.012

12.002

0.083

24

11-10.5

11.007

11.001

0.055

24

10-9.5

10.027

10.012

0.149

Vin (V)

Vref,step

24

14-13.5

1.9

3.82

24

13-12.5

2.3

12.525

24

12-11.5

2.7

11.521

24

11-10.5

3.8

10.518

24

10-9.5

3.3

3.9

9.517

TABLE 3.5(b)
Tr,2nd (ms) Ts,2nd (ms)

OS,2nd (%)

Vpdrop,step (V)
13.583

TABLE 3.6: The tabulated results for PID control with step reference
voltage applied.
TABLE 3.6(a)
Vp,2nd (V)
Vo,2nd (V)

Vin (V)

Vref,step

24

14-13.5

14.074

14.063

0.078

24

13-12.5

13.071

13.062

0.069

24

12-11.5

12.07

12.053

0.141

24

11-10.5

11.069

11.059

0.091

24

10-9.5

10.088

10.075

0.129

Vin (V)

Vref,step

24

14-13.5

13

13.6

24

13-12.5

6.3

9.8

12.614

24

12-11.5

9.5

12.2

11.613

24

11-10.5

10.8

12.5

10.596

24

10-9.5

6.5

11.1

9.638

TABLE 3.6(b)
Tr,2nd (ms) Ts,2nd (ms)

IV.

CONCLUSION

In conclusion, fuzzy logic controller is much better in

overall performance and term including rise time, peak time,
settling time and robustness as compared to PID controller.
Furthermore, the fuzzy logic control produced less voltage
deviation from variation of voltage reference. Small
overshoot, more damping and sensitive to parameters
variation makes fuzzy logic control works in better dynamic
performance. Not only buck converters, FLC also can be
applied to many converter topologies. FLC can also develop
in machinery for semiconductor that uses Fuzzy Logic as
controller instead of Programmable Logic Controller (PLC).
since FLC has advantages in fast response controlled with
higher accuracy. With all of these advantages, FLC has a
potential to improve robustness of DC/DC converters.

OS,2nd (%)

Vpdrop,step (V)

147

The 4th International Power Engineering and Optimization Conf. (PEOCO2010), Shah Alam, Selangor, MALAYSIA: 23-24 June 2010

Further work can be performed with the existing design

to improve the performance of the converter. Hardware
implementation on the controlled converter with using
Digital Signal Processing (DSP) to prove that hardware
application gives same performance as software simulation.

Dr. Ismail Musirin obtained Diploma of Electrical

Power Engineering in 1987, Bachelor of Electrical
Engineering (Hons) in 1990; both from Universiti
Teknologi Malaysia, MSc in Pulsed Power Technology
in 1992 from University of Strathclyde, United
Kingdom and PhD in Electrical Engineering from
Universiti Teknologi MARA, Malaysia in 2005. He has
published 2 books and more than 120 technical papers in
the international and national conferences, and international journals. He is
a reviewer of IEEE transactions and IET journals. He is also appointed as
the permanent reviewer for the World Scientific and Engineering Academy
and Society (WSEAS) centered in Greece. His research interest includes
power system stability, optimization techniques, distributed generator and
artificial intelligence. He is also a member of IEEE, IEEE Power
Engineering Society and Artificial Immune System Society (ARTIST).

V.ACKNOWLEDGEMENT
The author would like to express a gratitude especially to
Mr. Nik Fasdi Nik Ismail and his team members, for the
support, belief, patience, fairness and for the feedback. The
author would like to thank him for the opportunities and
knowledge that he has given to the author over the year.

Norazlan Hashim received his B.Eng degree in

Electrical Engineering and M.Eng degree in
Electrical Energy & Power System from the
University of Malaya in 2001 and 2007
respectively. He is currently a lecturer in the faculty
of electrical engineering, Universiti Teknologi
MARA (UiTM). His research interests include
power system control & stability, power system
modeling and power electronics.

VI. REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]

[10]

Guang Feng, Wanfeng Zhang, Yan-Fei Liu. An adaptive current

mode fuzzy logic controller for DC DC converters, IEEE
2003, page 983-989
Mohan, N., Underland, T.M., and Robbins W.P (2003). Power
Electronics Converters, Applications, and Design, 3rd edn.
John Wiley & Sons.
Anirban Dasgupta and Antip Ghosh, Study and Design of Buck
Converter,
http://www.scribd.com/doc/19037931/BuckConverter#
R. Gurunathan, A.K.S. Bhat (1996). A Soft-Switched Boost
Converter for High FrequencyOperation. IEEE Trans. 1999.
463-468.
Muhammad h. Rashid, Power Electronics: Circuits, Devices
and Applications, 3rd edition, chapter 5, page 187.
Ned Mohan, Tore M. Undeland, William P. Robbins. Power
Electronics: Converter, Applications and Design. Chapter 7,
Page 164
Liping Guo, John Y. Hung and R.M Nelms, Comparative
Evaluation of Linear PID and Fuzzy Control for Boost
Converter, IEEE Trans. 2005, page 556.
http://en.wikipedia.org/wiki/PID_controller
Wing-Chi So, Chi K. Tse, Yim-Shu Lee, Development of a
Fuzzy Logic Controller for Dc-DC Converters: Design,
Computer Simulation, And Experimental Evaluation, IEEE
Trans. 1996 page 24-25.
Paolo Mattavelli, Leopoldo Rossetto, Giorgio Spiazzi and Paolo
Tenti (1997). GeneralPurpose Fuzzy Controller for DCDC Converters. IEEE Trans.
Vol. 12. Issue 1.79-86.

Dalina Johari graduated from the University of

Liverpool, UK with honours degree in Electrical
Engineering in 1999. She then obtained her MSc
in Electrical Engineering from Universiti
Teknologi MARA, Malaysia in 2008. Prior to
continuing her Masters degree, she worked as an
engineer in the telecommunication industry for
almost 6 years. Her working experience has
shaped her interest in lightning, particularly
lightning protection system. Her research interests
also include artificial intelligence such as neural network and optimization
techniques such as evolutionary programming

VII. BIOGRAPHIES
Muhammad Jamhuri Yusoff obntained
Bachelor of Electrical Engineering (Hons)
from Universiti Teknologi MARA in 2010.

Nik Fasdi Nik Ismail obtained B.Eng. (Hons)

from Tokyo Denki University in 2002; M.Sc
from Universiti Malaya in 2008. He currently a
lecture at the Universiti Teknologi MARA,
Malaysia. His main research interests include
power
electronics,
power
optimization
techniques, distributed generator and artificial
intelligence in power system area.

148