Comparison of Fuzzy and Brain Emotional Learning Based Intelligent Control approaches for a Full Bridge DC-DC Converter

(1)

Vol. 2, No. 3, pp. 197-206, July (2019)

A Comparison of Fuzzy and Brain Emotional

Learning-Based Intelligent Control Approaches for

a Full Bridge DC-DC Converter

Reza Keramat

1

, Mohammad Hosein Ershadi

†,2

, and Shahrokh Shojaeian

3

1,2,3

Department of Electrical Engineering, Khomeinishahr Branch, Islamic Azad University, Khomeinishahr/Isfahan, Iran

In this paper, a Brain Emotional Learning-Based Intelligent Controller (BELBIC) and a fuzzy controller were used to control the output voltage of a full bridge DC-DC converter. The converter was presented by its state space averaged model assuming that it operates in a continuous conduction mode (CCM). A comparison was also made between the results. The effectiveness of control approaches was demonstrated by the uncertainty of system parameters and acceptable load variations. The performance of the BEBLIC and fuzzy controller in controlling the output voltage of the full bridge DC-DC converter was satisfactory. Since these controllers are not designed to reduce error to zero, it is not possible to claim that the error rate is precisely zero. Compared to the fuzzy controller, the BELBIC showed negligible overshoots and fluctuations. Both controllers reached stabilization almost at once. It can, therefore, be concluded that the BELBIC outperforms the fuzzy controller Given the uncertainty of the system parameters (including inductance, capacitance, and input voltage and acceptable variations of load).

Article Info

Keywords:

BELBIC, DC-DC converter, Fuzzy controller, Robustness Article History:

Received 2018-07-19 Accepted 2019-01-06

I.

I

NTRODUCTION

Due to the rapid advancement of power electronics in the recent decays, modern control approaches have needed to be developed accordingly. Since the controller of a linear power supply is in the active mode, its efficiency in high powers is less than 50%. The size of linear power supplies is large owing to the large dimensions of transformers and the need for large heat sinks because of high energy dissipation of switches.

In switched mode power supplies, because the frequency is high, the transformer size can be decreased. In these power supplies, the switches operate in the saturated or threshold point and thus their dissipation is modest. Accordingly,

linear power supplies have been replaced by switched-mode power supplies in many applications over time [1]. A full bridge DC-DC converter operates in the switched mode. The performance of this converter and its state space averaged model are discussed in section II.

Based on Brain Emotional Learning-Based Intelligent Controller (BELBIC) there exist a few adjustable parameters to achieve the desired solution. In [2-10], BELBIC was used to control electric drivers or power systems.

In the present study, BELBIC was used to control a full bridge DC-DC converter and the results were compared to the results of a fuzzy controller.

After a brief review of the performance of DC-DC converters and their state space averaged model sectionII, BELBIC and fuzzy controller are described in section III and IV respectively. Section V discusses the results obtained by BELBIC and fuzzy controllers. Moreover, a comparison is also made between the system parameters and load variations of these two controllers under uncertain conditions.

A B S T R A C T

†_{Corresponding Author:}_{ershadi@iaukhsh.ac.ir} Tel: +98-3133660011,

(2)

converter. The circuit up transformer (with D3, and D4), and switches

-V

signal. The switching frequency is equal to the carrier signal frequency. When

the command signal turn

the command signal is zero and turned

the switches

all four switches are off, load current keeps cycling through rectifiers. The duration of this process is

the period of switching and

half cycle of the switching period, the switch pair

and in the second half cycle of the switching period, the switch pair

switches as well as the periodic signal and constant signals

Vcontrol

II.

FULL

Fig. 1 shows the circuit diagram of a full bridge DC converter. The circuit

up transformer (with D3, and D4), and

switches are created by comparing the constant signals

Vcontrol and Vcontrol

signal. The switching frequency is equal to the carrier signal frequency. When

the command signal turn

the command signal is zero and turned off.

Fig. 1. The c

Moreover, when the switches T3 and

all four switches are off, load current keeps cycling through rectifiers. The duration of this process is

the period of switching and

and in the second half cycle of the switching period, the switch pair T1 and

The command pulse signals reaching

Vcontrol and –Vcontrol

Fig. 2. The command signals, periodic signals and constant signals for controlling

FULL

BRIDGE

DC

Fig. 1 shows the circuit diagram of a full bridge DC converter. The circuit is composed of a

up transformer (with a ratio of n:1), a diode bridge (D1, D2, D3, and D4), and a LC filter.

are created by comparing the constant signals

control with a high

signal. The switching frequency is equal to the carrier signal frequency. When Vcontrol is smaller than the periodic signal, the command signal turns T1 and

the command signal is zero and

The circuit schematic of the full bridge DC converter

Moreover, when –Vcontrol is larger than the periodic signal, 3 and T4 are on, otherwise they are off. When all four switches are off, load current keeps cycling through rectifiers. The duration of this process is

the period of switching and d

and in the second half cycle of the switching period, the 1 and T2 are off respectively [11

The command pulse signals reaching

control are shown in Fig. 2.

ommand signals, periodic signals and constant signals for controlling

DC-DC

CONVERTER

Fig. 1 shows the circuit diagram of a full bridge DC is composed of a high frequency step

ratio of n:1), a diode bridge (D1, D2, LC filter. The command

are created by comparing the constant signals with a high-frequency periodic carrier signal. The switching frequency is equal to the carrier signal is smaller than the periodic signal,

1 and T2 switches on. Otherwise, the command signal is zero and T1 and T2 switches are

ircuit schematic of the full bridge DC converter[11]

is larger than the periodic signal, 4 are on, otherwise they are off. When all four switches are off, load current keeps cycling through rectifiers. The duration of this process is 0.5Ts

is the duty cycle. In the first half cycle of the switching period, the switch pair

and in the second half cycle of the switching period, the 2 are off respectively

[11-The command pulse signals reaching T1, T

switches as well as the periodic signal and constant signals are shown in Fig. 2.

ommand signals, periodic signals and constant signals for controlling the switches [11]

CONVERTER

Fig. 1 shows the circuit diagram of a full bridge DC-DC high frequency step-ratio of n:1), a diode bridge (D1, D2,

command signals of the are created by comparing the constant signals

periodic carrier signal. The switching frequency is equal to the carrier signal is smaller than the periodic signal,

2 switches on. Otherwise, 2 switches are

ircuit schematic of the full bridge DC-DC

is larger than the periodic signal, 4 are on, otherwise they are off. When all four switches are off, load current keeps cycling through

-d.Ts .Here Ts is is the duty cycle. In the first half cycle of the switching period, the switch pair T3 and T4 and in the second half cycle of the switching period, the

12].

T2, T3 and T4 switches as well as the periodic signal and constant signals

ommand signals, periodic signals and constant [11]

DC - ratio of n:1), a diode bridge (D1, D2, of the are created by comparing the constant signals

periodic carrier signal. The switching frequency is equal to the carrier signal is smaller than the periodic signal, 2 switches on. Otherwise, 2 switches are

is larger than the periodic signal, 4 are on, otherwise they are off. When all four switches are off, load current keeps cycling through is is the duty cycle. In the first 4 and in the second half cycle of the switching period, the

4 switches as well as the periodic signal and constant signals

III.

It is assumed 1

Conduction Mode (CCM). 2

In each duty half modes:

Duty Mode 1:

During the time

transferred to the load via the transformer and two diodes. The circuit schematic of this mode is depicted in Fig. 3. In this mode, with KVL and KCL shown in

converter state space equations can be obtained

1 2

nV L x x x C x

X& A X B V

where:

1 2

[ ]

X = x x

0 1

In which

the voltage of the capacitor C.

Duty Mode 2:

In this mode, all switches are off and

cycling via rectifiers. The circuit schematic for this mode is shown in Fig. 4. In this mode, with KVL and KCL shown in Eq.(4) and (5), the converter state space equations can be obtained as

0

1 2

x C x

= +

X& =A X+B V

where:

FULL

BRIDGE

STATE

SPACE

It is assumed that:

1. the converter operates in the Continuous

Conduction Mode (CCM).

2. the transformer, diodes, and switches are

In each duty half-cycle, the converter can be shown in two

Duty Mode 1:

During the time T1 and

1 2

2

1 2

nV L x x

d

x

x C x

R

= +

&

1. 1. d

X& =A X+B V

1 2

[ ]T

X x x

1 1 ,

In which x1 denotes the

voltage of the capacitor C.

Fig. 3. The circuit schematic for duty mode 1

Duty Mode 2:

cycling via rectifiers. The circuit schematic for this mode is shown in Fig. 4. In this mode, with KVL and KCL shown in 4) and (5), the converter state space equations can be obtained as Eq.(6).

1 2

2

1 2

L x x

x

x C x

R

= +

&

2. 2. d

X& =A X+B V

BRIDGE

DC-DC

SPACE

AVERAGED

he converter operates in the Continuous Conduction Mode (CCM).

he transformer, diodes, and switches are cycle, the converter can be shown in two

1 and T2 switches are on and

1 2

nV L x x

. . _d

X A X B V

0

the current of the inductor L and x voltage of the capacitor C.

ircuit schematic for duty mode 1

cycling via rectifiers. The circuit schematic for this mode is shown in Fig. 4. In this mode, with KVL and KCL shown in 4) and (5), the converter state space equations can be

d

X A X B V

DC

CONVERTER

AVERAGED

MODEL

he converter operates in the Continuous

he transformer, diodes, and switches are cycle, the converter can be shown in two

switches are on and power is transferred to the load via the transformer and two diodes. The circuit schematic of this mode is depicted in Fig. 3. In this mode, with KVL and KCL shown in Eq.(1) and (2), the converter state space equations can be obtained as Eq.

current of the inductor L and x

ircuit schematic for duty mode 1

In this mode, all switches are off and the load current keeps cycling via rectifiers. The circuit schematic for this mode is shown in Fig. 4. In this mode, with KVL and KCL shown in 4) and (5), the converter state space equations can be

CONVERTER

MODEL

he converter operates in the Continuous

he transformer, diodes, and switches are ideal. cycle, the converter can be shown in two

power is transferred to the load via the transformer and two diodes. The circuit schematic of this mode is depicted in Fig. 3. In (1) and (2), the

Eq.(3).

(1) (2) (3)

current of the inductor L and x2 is

load current keeps cycling via rectifiers. The circuit schematic for this mode is shown in Fig. 4. In this mode, with KVL and KCL shown in 4) and (5), the converter state space equations can be

(3)

A the

model of the converter described in [12] and [13], the final full bridge DC

be obtained as follow

(BELBIC) is a learning com model includes

orbitofrontal, and thalamus shown in Fig. 5 [14].

,

Fig. 4.

Since the second half

A1 and B1 can be used for the period 2dT

the period (1-2d)T

model of the converter described in [12] and [13], the final full bridge DC-DC converter state space averaged model be obtained as follow

. .

. o

X A X B V V C X

= + = & where:

. 2

IV.

BRAIN

The Brain Emotional Learning (BELBIC) is a

learning computational model of the mammals model includes

0 0

Fig. 4. The circuit of duty mode 2

Since the second half-cycle is similar to the can be used for the period 2dT

2d)Ts. According to the state space averaged

model of the converter described in [12] and [13], the final DC converter state space averaged model be obtained as follows:

. . _d

X = A X+B V

. 1 2

BRAIN

EMOTIONAL

COMPUTATIONAL

The Brain Emotional Learning

(BELBIC) is a control algorithm based on the emotional putational model of the mammals

model includes an amygdala,

ircuit of duty mode 2

cycle is similar to the can be used for the period 2dTs and A

. According to the state space averaged model of the converter described in [12] and [13], the final

DC converter state space averaged model

0 1 2 0

EMOTIONAL

LEARNING

COMPUTATIONAL

MODEL

The Brain Emotional Learning-Based Intelligent Controller control algorithm based on the emotional putational model of the mammals

amygdala, a sensory cortex, orbitofrontal, and thalamus shown in Fig. 5 [14].

ircuit of duty mode 2

first half-cycle, and A2 and B2 for

. According to the state space averaged model of the converter described in [12] and [13], the final DC converter state space averaged model can

(7) (8) 1 1

LEARNING

MODEL

Based Intelligent Controller control algorithm based on the emotional putational model of the mammals brains. This sensory cortex, an orbitofrontal, and thalamus shown in Fig. 5 [14].

cycle, for . According to the state space averaged model of the converter described in [12] and [13], the final can

Based Intelligent Controller control algorithm based on the emotional brains. This an

Fig. 5. The b

The amygdala and orbitofrontal cortex have a grid structure in the computational model. There is a node in the amygdala and orbitofrontal context for

There is also a node in the amygdala for thalamus stimulus. The input to these no

inputs [15].

The outputs of the nodes in the amygdala and orbitofrontal cortex are obtained

j j j

A V S O W S

= =

where,

sensory input. Variations of are calculated

the learning coefficients of the amygdala and orbitofrontal cortex and

j j j

j j j j

V S A

W S A O R

∆ = −

∆ = − −

As seen, it is not possible to reduce

to say, learned information is not forgotten through the amygdala. In fact, forgetting or prevention is carried out in the orbitofrontal cortex. The output of the node in the amygdala is a stimulus originating from

obtained by by Eq.(14) [14]:

th j

A S

E A O A

=

= − +

Fig. 6 shows the system control block diagram using BELBIC.

inputs S are as presented

The brain emotional learning computational model

There is also a node in the amygdala for thalamus stimulus. The input to these nodes is equal to the maximum amount of inputs [15].

The outputs of the nodes in the amygdala and orbitofrontal cortex are obtained by Eq.

. .

j j j

A V S O W S

= =

here, V and W are the weights of nodes, and sensory input. Variations of

are calculated by Eq. (1

the learning coefficients of the amygdala and orbitofrontal cortex and R is the sensory signal.

. .max(0 , R )

. .( )

j j j

j j j j

i i

V S A

W S A O R

α

β

∆ = −

∆ =

∑

−

∑

−

As seen, it is not possible to reduce

obtained by Eq. (13). Finally, the model output is calculated (14) [14]:

max( )

th j

j j th

i i

A S

E A O A

=

∑

−

∑

+

Fig. 6 shows the system control block diagram using BELBIC. The functions of emotional signal R and sensory inputs S are as presented

rain emotional learning computational model

There is also a node in the amygdala for thalamus stimulus. des is equal to the maximum amount of

The outputs of the nodes in the amygdala and orbitofrontal by Eq. (9) and (10):

are the weights of nodes, and

sensory input. Variations of V and W in the learning process (10) and (11), where

the learning coefficients of the amygdala and orbitofrontal is the sensory signal.

. .max(0 , R )

. .( )

j j j

i

j j j j

i i

V S A

W S A O R

∆ = −

∆ = − −

∑

As seen, it is not possible to reduce the

(13). Finally, the model output is calculated

j j th

E=

∑

A −

∑

O +A

Fig. 6 shows the system control block diagram using unctions of emotional signal R and sensory inputs S are as presented in Eq. (15) and (16), respectively.

rain emotional learning computational model

The amygdala and orbitofrontal cortex have a grid structure in the computational model. There is a node in the amygdala and orbitofrontal context for sensory stimulus. There is also a node in the amygdala for thalamus stimulus. des is equal to the maximum amount of

The outputs of the nodes in the amygdala and orbitofrontal (9) and (10):

are the weights of nodes, and S

in the learning process ), where α_and

the learning coefficients of the amygdala and orbitofrontal

. .max(0 , R )

. .( )

j j j

W S A O R

∑

the values of Aj. That is to say, learned information is not forgotten through the amygdala. In fact, forgetting or prevention is carried out in the orbitofrontal cortex. The output of the node in the amygdala is a stimulus originating from the thalamus. It is (13). Finally, the model output is calculated

Fig. 6 shows the system control block diagram using unctions of emotional signal R and sensory (15) and (16), respectively.

rain emotional learning computational model [14]

The amygdala and orbitofrontal cortex have a grid structure in the computational model. There is a node in the sensory stimulus. There is also a node in the amygdala for thalamus stimulus. des is equal to the maximum amount of

The outputs of the nodes in the amygdala and orbitofrontal

(9) (10)

Sj is the in the learning process and β are

the learning coefficients of the amygdala and orbitofrontal

(11) (12)

. That is to say, learned information is not forgotten through the amygdala. In fact, forgetting or prevention is carried out in the orbitofrontal cortex. The output of the node in the thalamus. It is (13). Finally, the model output is calculated

(13) (14)

(4)

controller output, respectively. In this paper, the sensory input and emotional signal are defined

similar to a conventional PI controller. K

determined for the settling time adjustment and K responsible for smoothing the system response. Also, K and K

two sources. One of the sources is the experts, who define their knowledge of the system using the natural language. The other consists of measurements and mathematical models derived from physical laws. Hence, w

incorporate these types of information into the design of systems. The question is that how it is possible to formulate human knowledge within a framework similar to mathematical models. In other words, the main question is “How is it

mathematical formula?”

such a conversion possible. Fuzzy systems are based on knowledge or rules. The core of a fuzzy system is a knowledge base following the IF

fuzzy system is used by a controller, the controller is called “fuz

converter takes two inputs: error (

The membership input functions for this controller are in the range of

input components of error and error variations are shown in Fig.7. Seven membership functions are used

input [17]

negative small (NS), zero (Z), positive big (PB), positive medium (PM), and pos

Fig. 6. The system control block diagram using BELBIC

( , , ) ( , , )

R f e y u S g e y u

= =

where e, y, and

controller output, respectively. In this paper, the sensory input and emotional signal are defined

1 2 3 4

5 6

.| | . . .e . | u |

. . .e

R K y K e dt K K

S K e dt K

= + + +

=

∫

+

The control of gains K

determined for the settling time adjustment and K responsible for smoothing the system response. Also, K and K5 gains affect the steady

V.

In practical sys

two sources. One of the sources is the experts, who define their knowledge of the system using the natural language. The other consists of measurements and mathematical models derived from physical laws. Hence, w

incorporate these types of information into the design of systems. The question is that how it is possible to formulate human knowledge within a framework similar to mathematical models. In other words, the main question is “How is it possible to convert human knowledge to a mathematical formula?”

Basically, the main function

fuzzy system is used by a controller, the controller is called “fuzzy controller”.

The fuzzy controller designed for the full bridge DC converter takes two inputs: error (

The membership input functions for this controller are in the range of -5 and 5. Membership functions for each of the t input components of error and error variations are shown in Fig.7. Seven membership functions are used

input [17]-[18]: negative big (NB), negative medium (NM), negative small (NS), zero (Z), positive big (PB), positive medium (PM), and pos

ystem control block diagram using BELBIC

( , , ) ( , , )

R f e y u S g e y u

, and u are the system error, system output and controller output, respectively. In this paper, the sensory input and emotional signal are defined

1 2 3 4

5 6

.| | . . .e . | u |

. . .e

R K y K e dt K K

S K e dt K

= + + +

= +

∫

ontrol of gains K3 and K

determined for the settling time adjustment and K responsible for smoothing the system response. Also, K

gains affect the steady-state error [16].

FUZZY

CONTROLLER

In practical systems, important information originates from two sources. One of the sources is the experts, who define their knowledge of the system using the natural language. The other consists of measurements and mathematical models derived from physical laws. Hence, w

incorporate these types of information into the design of systems. The question is that how it is possible to formulate human knowledge within a framework similar to mathematical models. In other words, the main question is possible to convert human knowledge to a mathematical formula?”

Basically, the main function

fuzzy system is used by a controller, the controller is called zy controller”.

The fuzzy controller designed for the full bridge DC converter takes two inputs: error (

The membership input functions for this controller are in the 5. Membership functions for each of the t input components of error and error variations are shown in Fig.7. Seven membership functions are used

: negative big (NB), negative medium (NM), negative small (NS), zero (Z), positive big (PB), positive medium (PM), and positive small (PS).

ystem control block diagram using BELBIC

system error, system output and controller output, respectively. In this paper, the sensory input and emotional signal are defined as below

1.| | 2. . 3.e 4. | u |

R=K y +K e dt+K +K

and K6 as well as K

determined for the settling time adjustment and K responsible for smoothing the system response. Also, K

state error [16].

CONTROLLER

tems, important information originates from two sources. One of the sources is the experts, who define their knowledge of the system using the natural language. The other consists of measurements and mathematical models derived from physical laws. Hence, what matters is how to incorporate these types of information into the design of systems. The question is that how it is possible to formulate human knowledge within a framework similar to mathematical models. In other words, the main question is possible to convert human knowledge to a

of a fuzzy system is to make such a conversion possible. Fuzzy systems are based on knowledge or rules. The core of a fuzzy system is a knowledge base following the IF-THEN fuzzy rules. If a fuzzy system is used by a controller, the controller is called

The fuzzy controller designed for the full bridge DC converter takes two inputs: error (e) and error variations ( The membership input functions for this controller are in the

5. Membership functions for each of the t input components of error and error variations are shown in Fig.7. Seven membership functions are used

: negative big (NB), negative medium (NM), negative small (NS), zero (Z), positive big (PB), positive

itive small (PS).

ystem control block diagram using BELBIC [14]

(15) (16)

system error, system output and controller output, respectively. In this paper, the sensory

as below: [14]

(17) (18)

.| | . . .e . | u |

as well as K2 and K5 is

similar to a conventional PI controller. K3 and K6 are

determined for the settling time adjustment and K1 is

responsible for smoothing the system response. Also, K2, K4

CONTROLLER

tems, important information originates from two sources. One of the sources is the experts, who define their knowledge of the system using the natural language. The other consists of measurements and mathematical models hat matters is how to incorporate these types of information into the design of systems. The question is that how it is possible to formulate human knowledge within a framework similar to mathematical models. In other words, the main question is possible to convert human knowledge to a

a fuzzy system is to make such a conversion possible. Fuzzy systems are based on knowledge or rules. The core of a fuzzy system is a THEN fuzzy rules. If a fuzzy system is used by a controller, the controller is called

The fuzzy controller designed for the full bridge DC-DC ) and error variations (∆e). The membership input functions for this controller are in the 5. Membership functions for each of the two input components of error and error variations are shown in Fig.7. Seven membership functions are used here for the : negative big (NB), negative medium (NM), negative small (NS), zero (Z), positive big (PB), positive

) )

system error, system output and controller output, respectively. In this paper, the sensory

) ) is are is 4

tems, important information originates from two sources. One of the sources is the experts, who define their knowledge of the system using the natural language. The other consists of measurements and mathematical models hat matters is how to incorporate these types of information into the design of systems. The question is that how it is possible to formulate human knowledge within a framework similar to mathematical models. In other words, the main question is possible to convert human knowledge to a

a fuzzy system is to make such a conversion possible. Fuzzy systems are based on knowledge or rules. The core of a fuzzy system is a THEN fuzzy rules. If a fuzzy system is used by a controller, the controller is called

DC ). The membership input functions for this controller are in the wo input components of error and error variations are shown in for the : negative big (NB), negative medium (NM), negative small (NS), zero (Z), positive big (PB), positive

Fig. 7.

The controller output is assumed to be equal to the duty cycle, which varies between 0 and 0.5. Fig. 8

controller output functions.

LI, MI, SI, NC, SD, MD and LD stand for medium i

medium

maximum duty cycle (i.e. 0.25). presented in Table 1.

The parameters of

in this paper are presented in the Appendix. It is assumed that the converter operates in the CCM mode. Simulations were carried out using Matlab/Simulink

A. Effectiveness of BELBIC for the full bridge

DC-The reference signal increases linear from to 250V, and then it stops growing.

controller obtained by trial and error for the system with this input, are shown in Table 2. As seen in Fig. 9, at t=

output voltage is stabilized with a slight overshoot. BELBIC is a model

the system remains stabilized in spite of variations values of

(Figs. 10

In Fig. 10, the reference voltage changes from 250V to 350V. In Fig. 11, the values of

(330µf, 7

Fig. 7. The membership functions for e and

Fig. 8. The output membership functions

LI, MI, SI, NC, SD, MD and LD stand for increase, small

edium decrease and

maximum duty cycle (i.e. 0.25). presented in Table 1.

THE e

NB NM ∆e

NB LD LD NM LD LD NS LD LD Z LD MD PS LD MD PM LD MD PB MD SD

VI.

SIMULATION

parameters of the full bridge DC

in this paper are presented in the Appendix. It is assumed that the converter operates in the CCM mode. Simulations were carried out using Matlab/Simulink

Effectiveness of BELBIC for the full bridge -DC converter

The reference signal increases linear from to 250V, and then it stops growing.

output voltage is stabilized with a slight overshoot.

BELBIC is a model-independent controller. Consequently, the system remains stabilized in spite of variations

s of L, C, load, input voltage and reference voltage (Figs. 10 - 13).

In Fig. 10, the reference voltage changes from 250V to 350V. In Fig. 11, the values of

, 7mH) to (700µf

embership functions for e and

utput membership functions

LI, MI, SI, NC, SD, MD and LD stand for mall increase, no changes,

ecrease and large decrease in the half of the maximum duty cycle (i.e. 0.25). The f

TABLE I THE FUZZY RULES [17]

NM NS Z PS

LD LD MD MD LD MD SD MD LD MD SD SD MD SD NC Si MD SI SI MI MD MI SI MI SD MI MI LI

SIMULATION

RESULTS

the full bridge DC

in this paper are presented in the Appendix. It is assumed that the converter operates in the CCM mode. Simulations were carried out using Matlab/Simulink®.

Effectiveness of BELBIC for the full bridge DC converter

The reference signal increases linear from to 250V, and then it stops growing. The c

output voltage is stabilized with a slight overshoot.

independent controller. Consequently, the system remains stabilized in spite of variations

, load, input voltage and reference voltage

In Fig. 10, the reference voltage changes from 250V to 350V. In Fig. 11, the values of L and

µf, 4mH), (200µ

embership functions for e and ∆e

The controller output is assumed to be equal to the duty cycle, which varies between 0 and 0.5. Fig. 8 displays

utput membership functions

LI, MI, SI, NC, SD, MD and LD stand for large increase, hanges, small decrease, ecrease in the half of the The fuzzy rules are

[17]

PS PM PB

MD SI LI MD MI LI SD MI LI Si MI LI MI LI LI MI LI LI LI LI LI

RESULTS

the full bridge DC-DC converter in this paper are presented in the Appendix. It is assumed that the converter operates in the CCM mode. Simulations were

Effectiveness of BELBIC for the full bridge

The reference signal increases linear from a voltage of zero The coefficients of the controller obtained by trial and error for the system with this input, are shown in Table 2. As seen in Fig. 9, at t=0.01s the output voltage is stabilized with a slight overshoot.

independent controller. Consequently, the system remains stabilized in spite of variations

, load, input voltage and reference voltage

In Fig. 10, the reference voltage changes from 250V to and C are changed from

µf, 5mH), and (700 The controller output is assumed to be equal to the duty

displays the

ncrease, ecrease, ecrease in the half of the uzzy rules are

DC converter used in this paper are presented in the Appendix. It is assumed that the converter operates in the CCM mode. Simulations were

voltage of zero oefficients of the controller obtained by trial and error for the system with this 0.01s the

independent controller. Consequently, the system remains stabilized in spite of variations in the , load, input voltage and reference voltage

(5)

10mH)

to 20

in the input voltage from 40V to 50V.

B.

application of

this figure, the BELBIC has a slight overshoot while the fuzzy controller has a considerable overshoot (the initial overshoot is 35 V).

10mH) respectively

Fig. 12 shows the result of an increase in load from 12.5 to 20Ω (at t=0.05s). Fig. 13

Fig. 9.

Fig. 10. The output voltage as a result of changes

Fig. 11. The output voltage as a result of changes

Fig. 12. The output voltage as a result of changes in the load

B. Comparison of BELBIC with fuzzy controller

Fig. 14 shows the system output voltage after application of the

this figure, the BELBIC has a slight overshoot while the fuzzy controller has a considerable overshoot (the initial overshoot is 35 V).

respectively.

Fig. 12 shows the result of an increase in load from 12.5 (at t=0.05s). Fig. 13 display

The output voltage of the full bridge converter

The output voltage as a result of changes in the reference voltage

The output voltage as a result of changes in the values

The output voltage as a result of changes in the load

Comparison of BELBIC with fuzzy controller

Fig. 14 shows the system output voltage after the fuzzy and BELBIC controllers. As seen in this figure, the BELBIC has a slight overshoot while the fuzzy controller has a considerable overshoot (the initial overshoot is 35 V).

Fig. 12 shows the result of an increase in load from 12.5 displays the result of an increase in the input voltage from 40V to 50V.

utput voltage of the full bridge converter

The output voltage as a result of changes in the reference voltage

The output voltage as a result of changes s of L and C

Fig. 14 shows the system output voltage after BELBIC controllers. As seen in this figure, the BELBIC has a slight overshoot while the fuzzy controller has a considerable overshoot (the initial

Fig. 12 shows the result of an increase in load from 12.5Ω

s the result of an increase

utput voltage of the full bridge converter

The output voltage as a result of changes

Fig. 14 shows the system output voltage after the BELBIC controllers. As seen in this figure, the BELBIC has a slight overshoot while the fuzzy controller has a considerable overshoot (the initial

Ω

s the result of an increase

the BELBIC controllers. As seen in this figure, the BELBIC has a slight overshoot while the fuzzy controller has a considerable overshoot (the initial

The time required for stabilization is almost the same in both cases. It is therefore c

outperforms

Figs. 15 and 16 show

values obtained from the two controller types earlier. As seen in these figures, the capacitance of is initially equal to 33

reaches 297

decrease in the capacity of the capacitor using the fuzzy controller leads to higher growth in the range of fluctuations and an overshoot of

BELBIC. Hence, the BELBIC demonstrates better

performance. When the capacitance of

output voltage obtained from the BELBIC shows a higher delay in reaching stabilization compared to the fuzzy controller. Moreover, the range

of the output voltage are higher with the fuzzy controller. Fig. 17

obtained using the aforementioned two controllers. In this figure, the initial capacitance of

and reaches 396 overshoot of

surpass those of the BELBIC controller. Hence, the BELBIC demonstrates a better performance.

Fig. 13.

Fig. 14.

The time required for stabilization is almost the same in both cases. It is therefore c

outperforms the fuzzy controller. Figs. 15 and 16 show

values obtained from the two controller types . As seen in these figures, the capacitance of is initially equal to 33

reaches 297µf and 290µ

decrease in the capacity of the capacitor using the fuzzy controller leads to higher growth in the range of fluctuations

overshoot of the

output voltage obtained from the BELBIC shows a higher delay in reaching stabilization compared to the fuzzy controller. Moreover, the range

of the output voltage are higher with the fuzzy controller. Fig. 17 depicts a comparison

obtained using the aforementioned two controllers. In this figure, the initial capacitance of

and reaches 396µf. Obviously

overshoot of the output voltage using the fuzzy controller surpass those of the BELBIC controller. Hence, the BELBIC demonstrates a better performance.

Fig. 13. The output voltage as a result of changes in the input voltage

Fig. 14. The comparison of the fuzzy controller with the

The time required for stabilization is almost the same in both cases. It is therefore concluded that the BELBIC

the fuzzy controller. Figs. 15 and 16 show a comparison values obtained from the two controller types

. As seen in these figures, the capacitance of

is initially equal to 330µf, reduces by 10% and 12% and and 290µf, respectively. In addition, decrease in the capacity of the capacitor using the fuzzy controller leads to higher growth in the range of fluctuations

the output voltage compar

output voltage obtained from the BELBIC shows a higher delay in reaching stabilization compared to the fuzzy controller. Moreover, the range of fluctuations and overshoot of the output voltage are higher with the fuzzy controller.

comparison of

obtained using the aforementioned two controllers. In this figure, the initial capacitance of C (330

Obviously the range of fluctuations and output voltage using the fuzzy controller surpass those of the BELBIC controller. Hence, the BELBIC demonstrates a better performance.

The output voltage as a result of changes in the input voltage

omparison of the fuzzy controller with the BELBIC

The time required for stabilization is almost the same in oncluded that the BELBIC

comparison of output voltages values obtained from the two controller types described

. As seen in these figures, the capacitance of C

reduces by 10% and 12% and , respectively. In addition, decrease in the capacity of the capacitor using the fuzzy controller leads to higher growth in the range of fluctuations output voltage compared to the

performance. When the capacitance of C reduces by 12%, the

output voltage obtained from the BELBIC shows a higher delay in reaching stabilization compared to the fuzzy of fluctuations and overshoot of the output voltage are higher with the fuzzy controller.

output voltage values obtained using the aforementioned two controllers. In this (330µf) increases by 20% the range of fluctuations and output voltage using the fuzzy controller surpass those of the BELBIC controller. Hence, the BELBIC

The output voltage as a result of changes in the input voltage

omparison of the fuzzy controller with the

The time required for stabilization is almost the same in oncluded that the BELBIC

output voltages described

C, which reduces by 10% and 12% and , respectively. In addition, a 10% decrease in the capacity of the capacitor using the fuzzy controller leads to higher growth in the range of fluctuations ed to the

reduces by 12%, the output voltage obtained from the BELBIC shows a higher delay in reaching stabilization compared to the fuzzy of fluctuations and overshoot of the output voltage are higher with the fuzzy controller.

output voltage values obtained using the aforementioned two controllers. In this ses by 20% the range of fluctuations and output voltage using the fuzzy controller surpass those of the BELBIC controller. Hence, the BELBIC

The output voltage as a result of changes

(6)

Fig. 15. The comparison of the BELBIC controller and fuzzy controller with a 10% reduction in C.

Fig. 16. The comparison

controller with a 12% reduction in capacity.

Fig.17. The comparison of the fuzzy controller and BELBIC with a 20% increase in capacitance of

Fig.18. The comparison of the BELBIC and fuzzy controllers with a 10% redu

omparison of the BELBIC controller and fuzzy controller with a 10% reduction in C.

omparison of the BELBIC controller and fuzzy controller with a 12% reduction in capacity.

omparison of the fuzzy controller and BELBIC with a 20% increase in capacitance of

omparison of the BELBIC and fuzzy controllers with a 10% reduction in inductance

of the BELBIC controller and fuzzy controller with a 12% reduction in capacity.

omparison of the fuzzy controller and BELBIC with a 20% increase in capacitance of

omparison of the BELBIC and fuzzy controllers ction in inductance

of the BELBIC controller and fuzzy controller with a 12% reduction in capacity.

omparison of the fuzzy controller and BELBIC with a 20% increase in capacitance of C

omparison of the BELBIC and fuzzy controllers ction in inductance

Fig.19.

values obtained from the two controller

The results of

6.3mH the results are shown in

decreasing the value of inductance by 12% to 6.3mH are shown in

decrease in inductance, the range of fluctuati

overshoot of the output voltage obtained using the fuzzy controller

BELBIC shows a better performance. When inductance is reduced by 12%, the output voltage obtained by using the BELBIC controller reaches stab

Moreover, the fluctuations and overshoot of voltage obtained using the fuzzy controller than those derived from

and 21 show

using the aforementioned two controllers. In these figures,the

decreased by 20% to reach 15 and 10 ohms, respectively. As seen in the figures,

higher range of fluctuations and the output voltage than

Therefore, the BELBIC functions better.The

fluctuations and overshoot of the output voltage are higher by using the fuzzy controller with a decrease and an increase in the input voltage to 39V and 41V, respectively. Hence, the system yields a more satisfactory output when the BELBIC is us

decreases and increases to 38V and 42V, the BELBIC demonstrates a higher delay in reaching stabilization. In addition, the range of fluctuations and overshoot of output voltage obtained through the fuzzy controller ar

(Figs. 24 and 25).

The comparison of the BELBIC and fuzzy controllers with a 12% reduction in inductance

values obtained from the two controller

The results of decreasing the value of inductance by 10% to 6.3mH the results are shown in

decreasing the value of inductance by 12% to 6.3mH are shown in Fig. 19. As seen in these figures, with a 10% decrease in inductance, the range of fluctuati

overshoot of the output voltage obtained using the fuzzy controller exceed those of

BELBIC shows a better performance. When inductance is reduced by 12%, the output voltage obtained by using the BELBIC controller reaches stab

Moreover, the fluctuations and overshoot of voltage obtained using the fuzzy controller

those derived from show a comparison

the aforementioned two controllers. In these the initial load (12.5

decreased by 20% to reach 15 and 10 ohms, respectively. As seen in the figures,

range of fluctuations and the output voltage than

Therefore, the BELBIC functions better.The

fluctuations and overshoot of the output voltage are higher by using the fuzzy controller with a decrease and an increase in the input voltage to 39V and 41V, respectively. Hence, the system yields a more satisfactory output when the BELBIC is used (Figs. 22 and 23). When the voltage decreases and increases to 38V and 42V, the BELBIC demonstrates a higher delay in reaching stabilization. In addition, the range of fluctuations and overshoot of output voltage obtained through the fuzzy controller ar

(Figs. 24 and 25).

omparison of the BELBIC and fuzzy controllers % reduction in inductance

Figs. 18 and 19 show a comparison values obtained from the two controller

decreasing the value of inductance by 10% to 6.3mH the results are shown in Fig.

decreasing the value of inductance by 12% to 6.3mH are

. As seen in these figures, with a 10% decrease in inductance, the range of fluctuati

overshoot of the output voltage obtained using the fuzzy exceed those of the BELBIC. Hence, the BELBIC shows a better performance. When inductance is reduced by 12%, the output voltage obtained by using the BELBIC controller reaches stabilization in a longer time. Moreover, the fluctuations and overshoot of

voltage obtained using the fuzzy controller

those derived from the BELBIC controller.Figs. 20 comparison of the output voltages obtained the aforementioned two controllers. In these initial load (12.5Ω) is increased by 20% and is decreased by 20% to reach 15 and 10 ohms, respectively. As seen in the figures, when the fuzzy controller is used, a

range of fluctuations and overshoot the output voltage than when the BELBIC Therefore, the BELBIC functions better.The

fluctuations and overshoot of the output voltage are higher by using the fuzzy controller with a decrease and an increase in the input voltage to 39V and 41V, respectively. Hence, the system yields a more satisfactory output when ed (Figs. 22 and 23). When the voltage decreases and increases to 38V and 42V, the BELBIC demonstrates a higher delay in reaching stabilization. In addition, the range of fluctuations and overshoot of output voltage obtained through the fuzzy controller ar

omparison of the BELBIC and fuzzy controllers % reduction in inductance

comparison of output voltages values obtained from the two controllers described earlier

decreasing the value of inductance by 10% to Fig. 18. The results of decreasing the value of inductance by 12% to 6.3mH are

. As seen in these figures, with a 10% decrease in inductance, the range of fluctuations and overshoot of the output voltage obtained using the fuzzy the BELBIC. Hence, the BELBIC shows a better performance. When inductance is reduced by 12%, the output voltage obtained by using the ilization in a longer time. Moreover, the fluctuations and overshoot of the voltage obtained using the fuzzy controller, are higher

the BELBIC controller.Figs. 20 the output voltages obtained the aforementioned two controllers. In these

Ω) is increased by 20% and is decreased by 20% to reach 15 and 10 ohms, respectively. when the fuzzy controller is used, a overshoot are gained for

the BELBIC is used Therefore, the BELBIC functions better.The range of fluctuations and overshoot of the output voltage are higher by using the fuzzy controller with a decrease and an increase in the input voltage to 39V and 41V, respectively. Hence, the system yields a more satisfactory output when ed (Figs. 22 and 23). When the voltage decreases and increases to 38V and 42V, the BELBIC demonstrates a higher delay in reaching stabilization. In addition, the range of fluctuations and overshoot of output voltage obtained through the fuzzy controller are higher omparison of the BELBIC and fuzzy controllers

output voltages described earlier.

decreasing the value of inductance by 10% to The results of decreasing the value of inductance by 12% to 6.3mH are

(7)

Fig. 20. The c

Fig. 21. The c

Fig. 22. The comparison of

decrease in the input voltage (V

The comparison of the fuzzy controller and BELBIC with a 20% increase

The comparison of the fuzzy controller and BELBIC with a 20% decrease in load

omparison of the fuzzy controlle decrease in the input voltage (V

omparison of the fuzzy controller and BELBIC increase in load

omparison of the fuzzy controller and BELBIC decrease in load

fuzzy controller and BELBIC with a decrease in the input voltage (Vd=39V)

omparison of the fuzzy controller and BELBIC

r and BELBIC with a =39V)

Fig. 23. The c

Fig. 24.

Fig. 25. The c

The performance of the BEBLIC and fuzzy controller in controlling the output voltage of the full bridge DC converter was satisfactory. Since these controllers are not designed to reduce error to zero, it is not possible to claim that the error rate is precisely zero. Compared to the fuzzy controller, the BELBIC shows negligible overshoots and fluctuations. Both

once.

It is, therefore, concluded that the BELBIC

the fuzzy controller. Considering the uncertainty of system parameters (including inductance, capacitance, and input

The comparison of

increase in the input voltage (V

Fig. 24. The comparison of decrease in the

The comparison of

increase in the input voltage (V

VII.

The performance of the BEBLIC and fuzzy controller in controlling the output voltage of the full bridge DC converter was satisfactory. Since these controllers are not designed to reduce error to zero, it is not possible to claim that the error rate is precisely zero. Compared to the fuzzy controller, the BELBIC shows negligible overshoots and fluctuations. Both controllers reach stabilization almost at

omparison of the fuzzy controller and BELBIC with a increase in the input voltage (V

omparison of the fuzzy controller and BELBIC with a decrease in the input voltage (V

omparison of the fuzzy controller and BELBIC with a increase in the input voltage (V

CONCLUSION

The performance of the BEBLIC and fuzzy controller in controlling the output voltage of the full bridge DC converter was satisfactory. Since these controllers are not designed to reduce error to zero, it is not possible to claim that the error rate is precisely zero. Compared to the fuzzy controller, the BELBIC shows negligible overshoots and controllers reach stabilization almost at

fuzzy controller and BELBIC with a increase in the input voltage (Vd=41V)

fuzzy controller and BELBIC with a input voltage (Vd=38V)

fuzzy controller and BELBIC with a increase in the input voltage (Vd=42V)

CONCLUSION

The performance of the BEBLIC and fuzzy controller in controlling the output voltage of the full bridge DC converter was satisfactory. Since these controllers are not designed to reduce error to zero, it is not possible to claim that the error rate is precisely zero. Compared to the fuzzy controller, the BELBIC shows negligible overshoots and controllers reach stabilization almost at

It is, therefore, concluded that the BELBIC outperforms the fuzzy controller. Considering the uncertainty of system parameters (including inductance, capacitance, and input

fuzzy controller and BELBIC with an

fuzzy controller and BELBIC with a

fuzzy controller and BELBIC with an

The performance of the BEBLIC and fuzzy controller in controlling the output voltage of the full bridge DC-DC converter was satisfactory. Since these controllers are not designed to reduce error to zero, it is not possible to claim that the error rate is precisely zero. Compared to the fuzzy controller, the BELBIC shows negligible overshoots and controllers reach stabilization almost at

(8)

voltage and acceptable variations of load), the BELBIC presentss better performance than the fuzzy controller. Although fuzzy control is a robust and effective method for a large number of engineering systems, but its design (and consequently its performance) is almost depend on the experience and tact of the designer. Furthermore, after design and installation, its performance is not improved, and in other word, it is not a learning-based or intelligent controller. It can be stated that a Learning Based Intelligent Control which 'may' in the first step, act not as satisfying as any another modern controller, any straight-forward-designed controller after a few iterations thanks to its learning automata feature. Based on this deduction, we felt no need to emphasize the comparison of these controllers after setting optimization.

APPENDIX

Specifications of the full bridge DC-DC converter

0 Switches and diodes

resistance(Ω) 200

0 Switching frequency (Hz)

12.5 Load resistance(Ω) 7

Filter inductance(mH)

40 Input voltage(V) 330

Filter capacitor(µf)

1:10 Transformer ration 9

Transformer rating (KVA)

BELBIC CONTROLLER COEFFICIENTS 50

K5

0.00045

K1

2

K6

0.21

K2

0.000039

α

0.000002

K3

0.0000008

β

0.0007

K4

REFERENCES

[1] G.C. Chrysis, High Frequency Switching Power Supplies, 2nd Edition, 1989, McGraw-Hill

[2] M. Samadi , A. Afzali-Kusha, C. Lucas, “Power management by brain emotional learning algorithm,” ASIC, 2007, pp77-81.

[3] M. Abd Al-Alim El-Garhy, R.I. Mubarak and M. El-Bably "Improving maximum power point tracking of partially shaded photovoltaic system by using IPSO-BELBIC,"

Journal of Instrumentation, Volume 12, August 2017

[4] E. Daryabeigi, G. Arab Markade, C. Lucase," Interior permanent magnet synchronous motor (IPMSM), with a developed brain emotional learning based intelligent controller (BELBIC) ", in Proc IEEE, IEMDC, May. 2009, pp1633 – 1640.

[5] S. Valizadeh, M. Jamali, C. Lucas. "A Particle-Swarm-Based Approach for Optimum Design of BELBIC Controller in AVR System", International Conference on Control Automation and Systems, pp. 2679-2684, 2008.

[6] N.R. Abjadi, M. Mahdavi, H. Akbari," Emotional controller (BELBIC) for positive output Luo converter " 5th International Conference on Control, Instrumentation, and Automation (ICCIA),2017

[7] E. Daryabeigi, H. Abootorabi Zarchi, Gh.R. Arab Markadeh, M. A. Rahman, C. Lucas, “Implementation of emotional controller (BELBIC) for synchronous reluctance motor drive”

IEMDC, 2011, pp 1088 – 1093.

[8] E. Bijami, R. Abshari, S.M. Saghaiannejad, J. Askari, “Load frequency control of interconnected power system using brain emotional learning based intelligent controller”, ICEE, 2011,

pp 1-6.

[9] E.Bijami, M. Jadidoleslam, , A.Ebrahimi, , M.M. Farsangi, K.Y.Lee, “Power system stabilization using brain emotional learning based intelligent controller,”Power and Energy Society General Meeting, 2011, pp 1-7.

[10] E. Daryabeigi, , A. Emanian, , M.M. NamaziRashidi, S.M. Saghaian-Nejad, , “Torque ripple reduction of switched reluctance motor (SRM) drives, with emotional controller (BELBIC)”, APEC, 2012, pp 1528-1535

[11] A. Ghadimi, A. Movali Daryani, H. Rastegar, " Detailed Modelling and Analysis of a Full Bridge PWM DC-DC Converter" Department of Electrical Engineering Amirkabir University of Technology, Tehran

[12] Mohan, Undeland, Robbins, Power Electronics Converter. Application, and Design, thired edition Wiley, 2003

[13] C. Iannello, S. Luo, and I. Batarseh, "Small-Signal and Transient Analysis of a Full-Bridge, Zero-Current-Switched PWM Converter Using an Average Model", IEEE Transaction on Power Electronics, Vol. 18, NO.3, 2003 [14] A. R. Mehrabian, C. Lucas, "a toolbox for brain emotional

learning based intelligent controller," IEEE International Conference on Engineering of Intelligent Systems, pp. 1-5, 2006.

[15] J. Moren, C. Balkenius, "A Computational Model of Emotional Learning in the Amygdala," From animals to animals 6: Proc. of the 6th International conference on the simulation of adaptive behavior, Cambridge, Mass., 2000.The MIT Press.

[16] H, A. Zarchi, E. Daryabeigi, G. R. A. Markadeh, J. Soltani, "Emotional controller (BELBIC) based DTC for encoderless Synchronous Reluctance Motor drives, " in Proc. Int. Conf. PEDSTC, pp. 478-483 2011.

[17] N.L. Diaz and J.J. Soriano, "Study of Two Control Strategies Based in Fuzzy Logic and Artificial Neural Network Compared with an Optimal Control Strategy Applied to a Buck Converter," Annual Meeting of the North American Fuzzy Information Processing Society, June 2007 [18] M.R. Rafiei, R. Ghazi, R. Asgharian, H.A. Toliyat, and S.M.

(9)

of Engineering department in electrical engineering group at Islamic Azad University, Isfahan, Iran. His research

electric motor drives, power electronics converters design and control.

electrical engineering. He is currently Electrical Engineering

University, , Isfahan, Iran. His research inter control, power system control, stability, and reliability

of Engineering department in electrical engineering group at Islamic Azad University, Isfahan, Iran. His research

electric motor drives, power electronics converters design and control.

electrical engineering. He is currently Electrical Engineering

University, , Isfahan, Iran. His research inter control, power system control, stability, and reliability

Reza Keramat

degrees from

Azad University, Fars, Iran, in 2010 and Khomeini-shar branch

University, Isfahan, Iran, in 2013 respectively. His research interests are DC

and inverters and intelligent controller

Mohammad Hossein Ershadi

Isfahan, Iran, in 1981. He received his B.S., M.Sc and PhD

engineering from Isfahan University of Technology, Isfahan, Iran, in 2004, 2007 and 2012 respectively. He is the faculty member of Engineering department in electrical engineering group at Islamic Azad University, Isfahan, Iran. His research

electric motor drives, power electronics converters design and

Shahrokh Shojaeian

M.Sc. degrees

Technology, Isfahan, Iran, in 1997 and 2000 respectively, and Ph.D. degree from

and Research Branch

University, , Tehran, Iran, in 2012, all in electrical engineering. He is currently

Electrical Engineering, Khomeinishahr Branch University, , Isfahan, Iran. His research inter control, power system control, stability, and reliability

Reza Keramat received the B.Sc. and M.S from Kazerun branch

Azad University, Fars, Iran, in 2010 and shar branch of

University, Isfahan, Iran, in 2013 respectively. research interests are DC

and intelligent controller

Mohammad Hossein Ershadi

Isfahan, Iran, in 1981. He received his B.S., M.Sc and PhD degrees

engineering from Isfahan University of Technology, Isfahan, Iran, in 2004, 2007 and 2012 respectively. He is the faculty member of Engineering department in electrical engineering group at Islamic Azad University, Isfahan, Iran. His research

electric motor drives, power electronics converters design and

Shahrokh Shojaeian received his B.Sc. and s from Isfahan University of Technology, Isfahan, Iran, in 1997 and 2000 respectively, and Ph.D. degree from

and Research Branch of

University, , Tehran, Iran, in 2012, all in electrical engineering. He is currently a Faculty

Khomeinishahr Branch of

University, , Isfahan, Iran. His research interests are nonlinear control, power system control, stability, and reliability

received the B.Sc. and M.Sc. Kazerun branch of Islamic Azad University, Fars, Iran, in 2010 and of Islamic Azad University, Isfahan, Iran, in 2013 respectively.

research interests are DC-DC converters and intelligent controllers.

Mohammad Hossein Ershadi was born in Isfahan, Iran, in 1981. He received his B.S., degrees in electrical engineering from Isfahan University of Technology, Isfahan, Iran, in 2004, 2007 and 2012 respectively. He is the faculty member of Engineering department in electrical engineering group at Islamic Azad University, Isfahan, Iran. His research interests are electric motor drives, power electronics converters design and

received his B.Sc. and from Isfahan University of Technology, Isfahan, Iran, in 1997 and 2000 respectively, and Ph.D. degree from Science of Islamic Azad University, , Tehran, Iran, in 2012, all in Faculty member of of Islamic Azad ests are nonlinear control, power system control, stability, and reliability.

. Islamic Azad University, Fars, Iran, in 2010 and Islamic Azad University, Isfahan, Iran, in 2013 respectively.

s

was born in Isfahan, Iran, in 1981. He received his B.S., in electrical engineering from Isfahan University of Technology, Isfahan, Iran, in 2004, 2007 and 2012 respectively. He is the faculty member of Engineering department in electrical engineering group at interests are electric motor drives, power electronics converters design and

(10)

IECO