Artificial Neural Network-Based Hybrid Controller for Electric Vehicle

Applications

RAKESH BABU BODAPATI*, R. S. SRINIVAS, P. V. RAMANA RAO

Department of Electrical and Electronics Engineering,

Acharya Nagarjuna University,

Guntur, Andhra Pradesh-522510

INDIA

*Corresponding Author

Abstract: - Power management among different energy sources of electric vehicles (EV) is one of the complex

issues during the transition from one to another. A specific control is modeled based on the current and speed

range of the electric motor named as Measurement of Parameter-Based Controller (MPBC), which will play a

key role during transition of energy sources as per the load requirement. Two bidirectional converters are

utilized to control the pulse signals generated by the traditional controllers which are connected at the battery

and Supercapacitors (SCap) ends, which are treated as passive sources of the system. The Controller's artificial

neural network (ANN), fuzzy logic (FLC), and proportional-integral (PI) are utilized to generate the pulse

signal to the switches present in the converters by load. Further, specific controller MPBC is combined with

three controllers ANN/FLC/PI individually and obtained three separate hybrid controllers as per the proposed

control technique. The MPBC+PI/FLC/ANN controller-based MATLAB/Simulink model was designed, and

applied to the electric motor individually at different load conditions. This model considered three different

power delivery states from the PV array and assessed the motor's performance under different load scenarios.

Compare the three hybrid controllers' final results to find out which one is more effective than the others.

Key-Words: - Artificial Neural Network (ANN), Photo Voltaic (PV) energy, Fuzzy Logic System, Super

Capacitor, Proportional Integral (PI) controller, Hybrid controller (HC), Measurement of

parameter-based controller (MPBC), Electric vehicle, Battery.

Received: April 23, 2024. Revised: August 27, 2024. Accepted: September 25, 2024. Published: October 31, 2024.

1 Introduction

An eco-friendly and sustainable substitute for

traditional energy sources like nuclear power and

fossil fuels is solar power generation. Due to the

abundant and endless supply of sunlight, one of the

main benefits of solar electricity is that it is

renewable. Solar energy is a clean and sustainable

energy source since it produces no greenhouse gas

emissions, in contrast to fossil fuels. On the other

hand, traditional energy sources such as coal, oil,

and natural gas are limited resources that need to be

extracted and processed, resulting in greenhouse gas

emissions and damage to the environment.

Significant dangers to the environment and human

health are additionally created by these sources,

including habitat destruction, air and water

pollution, and climate change.

Photovoltaic (PV) panels and concentrated solar

power (CSP) plants are examples of solar power

generation systems that use the sun's energy to

create electricity. Decentralized and distributed

energy solutions can be achieved by integrating

them into building materials, solar farms, and

rooftop deployments. Furthermore, as a result of

improvements in efficiency as well as cost

generated by solar technology, solar power is now

more affordable and accessible than conventional

energy sources, [1]. In general, solar energy has

many advantages over traditional energy sources,

such as lower carbon emissions, long-term cost

savings, energy independence, and environmental

sustainability. It is essential to switch to clean and

renewable energy sources in the future, [2].

FLC-based tracking system. Different types of

camera tracking controllers have been developed

based on the FLC also tested in simulation software.

In this the pixel positions from the image are taken

as input on the other hand provided pan will be

considered as output. Two types of rules-based

tuning technique is applied to the membership

functions to obtain the better-quality picture from

the camera, [3].

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

192

Volume 23, 2024

In [4], an FLC for autonomous vehicle

propulsion. The FLC can control the speed of the

vehicle on its own based on the load. The range of

the heading range is taken and given to the FLC for

proper operation of the vehicle. The noise analysis

of the controller was also performed to attain the

importance of other controllers’ performance.

Stated the effect of the FLC on the sliding mode

control technique applied to the various speed-

controlling applications. The sliding mode control

(SMC) generally requires the all data of the plant

model. On the other hand, the FLC is more robust

than the SMC. In this, the effect of the FLC is

investigated on the SMC and executed with all

simulation results, [5]. Developed an automatic

system with FLC. The automatic transmission (AT)

system for the vehicle propulsion is one of the

useful factors for the vehicle. Generally, a

mechanical gear system is used in conventional

transport vehicles, the driver is required to change

the gear depending on the vehicle condition. An

FLC-based gear system is designed and

implemented for the vehicles for better

performance, [6], [7], [8].

Proposed the NN-based technique for fast

convergence and accuracy [9]. In this, a procedure is

developed to train ANN for effective control. This

type of system gives better results after completion

of the successful training. Especially, feedforwarded

NN is developed to know the performance of the

non-linear components or systems, [10]. A

statistical-based algorithm was also developed, to

examine the performance of the ANN technique. At

each iteration a new network is modelled to remove

the previously occurred errors. The intended

approach is rapidly applicable where a large amount

of trained data can be produced, [11].

An NN-based dynamic simulation of the

suspension for irregular road conditions, [12]. The

NN is trained from the validation data obtained from

the laboratory. The results obtained from NN show

the effectiveness for a wide range of applications.

This has been applied to the vehicle based on the

elastic bushings and dynamics of the tire. And this

NN-based model is very much useful for assessing

vehicle ride, vibration, and noise because of its

better computational efficiency and accuracy, [13].

Introduced an ANN controller for maximum

power point tracking (MPPT), [14]. The ANN

control with an MPPT is applied to the boost

converter to attain the high efficiency and the

reference voltage is obtained for ANN from MPPT.

The adopted technique will change the PV module

voltage as per the MPPT instruction at any

temperature, or insulation. To attain an effective

response from the proposed model, the system is

implemented with a digital signal processor (DSP).

The overall information obtained from the system is

applied to the ANN to generate the duty cycle as per

the load requirement, [15].

[16], Stated control technique to the HEV based

on the ANN to improve the effectiveness of the

AC/DC converters. The switches in the Series HEV

are exposed to open conditions because the internal

problems that occurred during the operation. The

ANN-based controller is used to identify the

switching problem at various levels, and various

patterns are considered. The intended control

technique is implemented in simulation and also

validated through a small experimental setup, [17].

An effective braking system corresponding to

HESS for EVs, HEVs, and PHEV with BLDC

motor is proposed. Different character batteries and

UC are combined for HESS for better utilization of

the battery and also charging of UC from the

regenerative braking system (RGS), at this time the

BLDC acts as a generator. Using a proper

controlling algorithm, the voltage is boosted, sent to

the UC or battery. The crest power is useful for

avoiding the complete discharge of the battery

during the uphill of the vehicle. The ANN-based

technique is used here to attain smooth breaking

distribution, [18].

The EV charge scheduling using ANN is

illustrated. The machine-to-machine (M2M) plays a

vital role if the EVs are integrated with the power

grid for charging of the power source, here energy

management is one of the key factors to control. The

smart devices connected to the system will predict

the power demand and supply needs of the system.

In this case, ANN-based algorithm is used for EVs

charging schedule by taking data from M2M. The

household data generally used to decide the

charging schedule of the vehicle, [19].

An ANN for instant power development of the

EVs. In this, the accuracy of the instant power has

been increased to estimate the power the battery-

powered vehicles are developed. In general, the

battery can be utilized as a secondary power source

or main source in the EV application depending

upon the requirement. The power source

characteristics are identified from the SOC, voltage,

and health of the device. The ANN is used here for

estimating the instant power to the EV for better

efficiency, [20].

2 Artificial Neural Network

The NN (Neural Network) controller, using the

Backpropagation (BPN) algorithm, uses

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

193

Volume 23, 2024

reinforcement-type learning to self-tune its

parameters to allow it to follow a specified

trajectory. When dealing with control difficulties in

the real world, it is usual to draw on extensive past

knowledge about different subsystems of control

systems. This abundance of data improves the

accuracy and dependability of Neural Networks

(NNs) used to simulate these subsystems. With the

help of extensive data, NNs may be trained to

precisely reflect and imitate the behavior of

complex plants, leading to enhanced performance in

control applications.

The goal of a self-learning computational

system integrates the closely related fields of neural

networks, parallel processing, connectionism, and

neural computing. Neural networks are based on the

neuron, a basic building block that makes

complicated learning and decision-making possible.

Neurons are the processing elements of neural

networks. These simple elements are connected by a

variety of topological classes, trained by yet another

class of learning algorithms.

2.1 Backpropagation Principles of Operation

The BPN neural network belongs to the class of

feedforward networks. This implies that information

flows in one direction only - from input to output.

The multilayered feedforward neural network can

learn a mapping of any complexity. The network's

learning of a particular pattern is based on repeated

presentation of the data set. This type of neural

network has a propagate-adjust cycle allowing the

neural network to learn an entire data set. The data

set is presented to the inputs of the neural network

and the information travels through the hidden layer

to the output layer. This action constitutes the

propagation phase. The adjust stage entails the

comparison of these outputs to desired outputs, and

the error information is to modify the weights of the

neural network. This error BPN is the basis for the

training algorithm used to train a multilayered

feedforward network.

2.2 Backpropagation Structure and

Components

The BPN is one of the many existing neural network

topologies. It is composed of several nodes at which

point computation takes place. This node or neuron

is the basic building block upon which the neural

network structure is built It is intended to simulate a

biological neuron.

CONNECTION WEIGHTS

INPUT

NODE NONLINEARITY

OUTPUT

BIAS

Fig. 1: Basic Neuron model

The Backpropagation Neural Network (BPN)

neuron model shown in Figure 1 has several inputs

and one output. Every input contributes to the total

output after going through a connection weight Wn.

The inputs, the associated weights of the inputs, a

bias term, and a squashing function—which adds

non-linearity to the node—all decide the output. By

modifying weights and biases during training, this

structure helps the BPN understand complex data

patterns, improving its capacity for learning and

precise forecasting.

INPUTS OUTPUTS

HIDDENLAYER

WEIGHTS

HIDDENLAYER

NODE NONLINEARITY

FOR HIDDENLAYER

OUTPUTLAYER AND WEIGHTS

OUTPUT LAYER NODE NONLINEARITY

FOR OUTPUT LAYER

Fig. 2: General BPN NN structure

The basic backpropagation structure is

composed of layers of these interconnected neurons

as shown in Figure 2. The network is made up of

three distinct layers. The input layer serves as an

interface with the connecting system. It directly

feeds the hidden layer via a network of connection

weights. Since the hidden layer is built from the

neuron model it takes the sum of al1 the inputs and

a bias. This sum known as neth is then passed

through a nonlinear function. This result flows

through the interconnecting weights to the output

layer where the summing process starts all over

again. The error terms between the desired output

and the network output are calculated and used to

adjust the output weights and then propagated back

to contribute to the process for al1 the hidden layer

weights. The net continues in this propagate-adjust

manner until all patterns are learned.

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

194

Volume 23, 2024

2.3 Backpropagation Governing Equations

The equation governing the operation of the BPN

shown in Figure will now be presented. The input to

the hidden layer is:

1()

pj ji pi j

neth wh x h





  



(1)

The weight connections between the input and

the hidden layer are denoted as Whg and &



hji is a

bias input. The use of the bias input is optional.

These are then processed by a squashing function:

()

pj j pj

Y Fh neth

(2)

The outputs of these hidden nodes become the

inputs to the output layer. Thus,

 

pk kj pj k

neto wo Y o





  



(3)

where the connections between the hidden layer and

the output layer are denoted as

and



is the

bias input. These outputs must also be treated by a

squashing function. This results in an output of:

()

pk k pk

O FO neto

(4)

The next step will be to describe the delta

update rule. The backpropagation algorithm

performs the steepest descent minimization on a

surface in weight space whose height at any point is

equal to the error.

The error between the actual net output and the

target is defined as:

()

k k k

E d O

(5)

For the network to learn this error must be

minimized and used in some way to update the

weights in the output layer as well as those

associated with the hidden layer. The following

error is defined.

 

p k pk

E Tp O



  







(6)

And

 

p pk

pk pk

kj pk kj

E neto

wo neto wo





    

  

(7)

 

pk kj pj k

kj kj

neto wo Y o

wo wo





   





(8)

 

pk pk pj

kj pk

EFo

T O Y

wo neto



    



(9)

The above equation allows for a weight update rule

of the following form:

( 1) ( ) ( ) ( )

kj kj pk pk k k j

wo t wo t T O Fo neto Y



      

(10)

Letting

( ) ( )

pk pk pk k pk

o T O Fo neto



  

(11)

we have for the output layer

( 1) ( )

kj kj pk pj

wo t wo t o Y



    

(12)

and smeller for the hidden layer

 

()

pj hj pj pk kj

h F neth o wh



  



(13)

( 1) ( )

ji ji pj i

wh t wh t h X



    

(14)

where the derivative of the squashing function is

needed for both the hidden layer as well as the

output layer. Here we can see the algorithm’s

dependency upon the error terms computed for the

output layer as well as the hidden layer. The above-

described equations are a basis for the following

study of some modifications on the generalized

delta rule backpropagation networks.

3 A Proposed Control Technique for

Energy Management Strategy

This suggests a novel control technique called SCap

to allow the battery to alternate between its charging

and discharging cycles based on the speed and

current values of the electric motor. The

recommended control technique is implemented

using three different hybrid controllers; the

traditional controllers, ANN, FLC, and PI, are

utilized to generate the pulse signals that are

transmitted to the converter. Furthermore, the

MPBC controller regulates these signals based on

the speed and current levels of the electric motor.

Ultimately, the combination of MPBC+PI,

MPBC+FLC, and MPBC+ANN provides the

method to implement the recommended procedure.

The main circuit of the proposed model consists

of up of the PV array, battery, SCap, two

bidirectional converters, Boost converter, and

electric motor which are depicted in Figure 3. In this

instance, the SCap is utilized to meet the peak

power consumption, and the battery powers the EV

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

195

Volume 23, 2024

if solar power generation is unavailable. On the

other hand, PV arrays are employed to meet the

demand for EVs as well as for battery and SCap

charging, depending on the load placed on the EV.

The three different controller combinations known

as MPBC plus PI/FLC/ANN constitute the

suggested control technique, which is the base for

all of the regulated signals of the two bidirectional

converters. In this case, the PI, FLC, or ANN

controller generates the pulse signal to the converter

at the battery and SCap end based on the actual and

reference voltage levels of the converter. But MPBC

functions according to the current and speed

readings of the EM. This is accomplished by using

mathematical functions, which, based on the current

and speed values of the EM, generate several

different types of signals. MPBC, PI, FLC, and

ANN combine to supply the EV with the proper

power supply based on the applied load, supplying

the converter at the battery and SCap ends with

controlled pulse signals.

Battery

SCap

The controlled pulse signals are provided to switches S2,S3,S4 and S5

based on the speed and current of the electric motor

C3 C4

Breaker MPBC

Contr

oler

Outputs of MFB

Controller Input of MPBC

controller (Current and Speed)

Bidirectional

converter2

Bidirectional

converter1

Array

L1 D

Unidirectional

converter

Fig. 3: Main circuit with a Proposed control

technique

4 MATLAB/Simulation Results and

Discussions

In this case, the controlled signals to the EV are

designed considering three different load conditions.

This will take place in line with the recommended

controlled approach, which blends MPBC, ANN,

FLC, and PI.

Fig. 4: Speed response of the electric motor with

MPBC+ANN Controller

Fig. 5: Speed response of the electric motor with

MPBC+FLC and MPBC+PI Controllers

The speed curve response of EM under varied

loads is displayed in Figure 4 and Figure 5. To

verify the effectiveness of the recommended control

strategy, three different loads are applied to the EV

at three different times. The motor slows to roughly

1500 rpm when a standard load is put to the EM for

0.2 seconds. In 0.4 seconds, the motor drops to 1400

rpm when a rated load is applied. Finally, the

motor's speed drops to even less than 1400 rpm

when an additional 0.6-second rated load is

introduced. MPBC+PI took 0.09 seconds, 0.1

seconds, and 0.30 seconds, MPBC+ANN took 0.05

seconds, 0.07 seconds, and 0.21 seconds, and

MPBC+FLC took 0.075 seconds, 0.08 seconds, and

0.28 seconds to reach stable with the applied loads

in that order. MPBC with ANN improved the other

controller output, according to the results.

4.1 Mode-I Operation (PV Active Case)

Fig. 6: Representation of the PV, Load & Battery,

SCap Power curves, related PV active case

In this mode of operation, PV is able to supply

power to the load along with the battery and SCap

changing (for a short period). Figure 6 shows the

power curves of the load, passive source, and active

source PV.

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

196

Volume 23, 2024

Fig. 7: Representation of the pulse signals of the

five switches of the complete circuit, related PV

active case

Figure 7 shows the pulse signal generation of

the 5 switches present in the circuit, this mode of

operation S1having the continues pulses which

shows that the PV array is supplying power

throughout the time as specified, S2 is used in the

converter at battery end used for boost operation

(discharging) no signal found which shows the no

output provided from the battery, S3 is used for

buck operation of the converter at battery end, as per

the Figure 7 shows the pulse signals up to 0.6 sec

which shows the battery charging period. Switches

S4 and S5 are used in the bidirectional converter

used at SCap, here S4 is for boost operation and S5

is for buck operation, it is clear when ever load

applied to the EM SCap can discharge to meet the

instance power requirement of the load.

Fig. 8: Representation of the PV array voltage,

current and power curves, related PV active case

In this mode of operation total power is supplied

by the PV array which is required by the two

passive sources and the load which is clear in Figure

Fig. 9: Representation of the Battery voltage,

current, %SOC and power curves, related PV active

case

Figure 9 shows the battery parameters-related

representation. In this case, power consumed by the

battery shows with negative region, even current

values also.

Fig. 10: Representation of the SCap voltage, current,

%SOC, and power curves, related PV active case

Figure 10 shows the SCap parameters, the

power becomes negative during the starting of the

motor, further all load-applied cases power curve

shows the positive gesture corresponding to that

voltage, % SOC, and current values are also

changed.

4.2 Mode-II Operation (PV and Battery

Active Case)

Fig. 11: Representation of the PV, Load & Battery,

SCap Power curves, related PV and Battery active

case

In this mode of operation PV can supply 50% of

the power to the load along and remain part can be

sent from the battery which clearly from Figure 11

shows the power curves of the load, passive source,

and active source PV.

Fig. 12: Representation of the pulse signals of the

five switches of the complete circuit, related PV,

and Battery active case

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

197

Volume 23, 2024

Figure 12 shows the pulse signal generation of

the 5 switches present in the circuit, this mode of

operation S1having continuous pulses which show

that the PV array is supplying power throughout the

time as specified, S2 is used in the converter at

battery end used for boost operation (discharging)

signals found from 0.4 sec onwards, S3 is used for

buck operation of the converter at battery end, as per

the Figure 12 shows the pulse signals up to 0.4 sec

which shows the battery charging period. Switches

S4 and S5 are used in the bidirectional converter

used at SCap, here S4 is for boost operation and S5

is for buck operation, and it is clear whenever load

applied to the EM SCap can discharge to meet the

instance power requirement of the load.

Fig. 13: Representation of the PV array voltage,

current and power curves, related PV, and Battery

active case

In this mode of operation total power is supplied

by the PV array which is required by the two

passive sources and the load which is clear in Figure

13 and 50% of the load requirement is done

thorough PV array.

Fig. 14: Representation of the Battery voltage,

current, %SOC and power curves, related PV, and

Battery active case

The representation connected to battery

parameters is shown in Figure 14. In this instance,

both the battery's power consumption and its current

values are negative.

Fig. 15: Representation of the SCap voltage, current,

%SOC and power curves, related PV, and Battery

active case

Figure 15 displays the SCap parameters. The

power decreases when the motor starts, and in all

load-applied scenarios, the power curve displays a

positive gesture that corresponds to changes in

voltage, percentage SOC, and current values.

4.3 Mode-III Operation (Battery Active

Case)

Fig. 16: Representation of the PV, Load & Battery,

SCap Power curves, related Battery active case

In this mode of operation, no power flows are

there from PV, and all required power is sent from

the battery which is clear from Figure 16 shows the

power curves of the load, passive source.

Fig. 17: Representation of the pulse signals of the

five switches of the complete circuit, related Battery

active case

Figure 17 shows the pulse signal generation of

the 5 switches present in the circuit related to no

power supplied from the PV array case. Here no

pulse signal generation there for switch S1 since no

sunlight condition. Further, all power needed by the

load is provided by the battery with the help of the

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

198

Volume 23, 2024

SCap during peak power requirement which enables

the signal switches signals of the S2 and S4 for

boost operation at the battery, SCap end.

Fig. 18: Representation of the PV array voltage,

current and power curves, related Battery active

case

Figure 18 shows the no parameters available

case from the PV array since no sunlight condition

and shows all values as zero.

Fig. 19: Representation of the Battery voltage,

current, %SOC, and power curves, related Battery

active case

The representation connected to battery

parameters is shown in Figure 19. In this instance,

both the battery's power and current numbers

display a positive region.

Fig. 20: Representation of the SCap voltage, current,

%SOC, and power curves, related Battery active

case

Figure 20 displays the SCap characteristics.

When the motor is first started, the power turns

negative. When a load is supplied, the power curve

also displays a positive gesture that corresponds to

changes in voltage, percentage SOC, and current

values.

Table 1. Based on the speed value, a performance

comparison is made between MPBC plus PI and

MPBC plus FLC, MPBC plus ANN.

Controller

Name

Time Taken to reach Steady

state in sec

Mode1

Mode2

Mode3

MPBC+PI

0.09

0.1

0.30

MPBC+FLC

0.075

0.08

0.28

MPBC+ANN

0.05

0.07

0.21

Based on the speed curve settling value, Table 1

shows the comparison analysis of MPBC with PI,

FLC, and ANN. Moreover, MPB plus ANN

excelled with MPBC plus PI/FLC in terms of

performance.

5 Conclusion

This study looks at a control system that adjusts the

battery and SCap about the motor's speed. The

MPBC is the result of combining three distinct

mathematical functions that are independently

realized based on the motor's speed and current

values. To achieve the main objective of the

research, the suggested MPBC controller is

integrated with a conventional PI controller, FLC,

and ANN to form a hybrid controller. While the PI

controller, FLC, and ANN produced the switching

signals required by the converter, the MFBC

controlled the pulse signals that correlated with the

motor speed. In the final analysis, the requirements

of electric vehicles are fulfilled by employing the

recommended control strategy, which results in a

smooth transition between battery and SCap.

Another hybrid controller, termed MPBC with

ANN, is designed to perform the same purpose as

MFBC plus FLC. A comparative analysis is done

before applying the afterload and proceeding. The

results for both approaches are tabulated and

presented in the final section. In the end, the MPBC

plus ANN outperformed the MPBC+PI and

MPBC+FLC hybrid controllers in terms of

performance.

References:

[1] S. M. Shariff, M. S. Alam, F. Ahmad, Y.

Rafat, M. S. J. Asghar and S. Khan, (2020).

"System Design and Realization of a Solar-

Powered Electric Vehicle Charging Station,"

in IEEE Systems Journal, vol. 14, no. 2, pp.

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

199

Volume 23, 2024

2748-2758, doi:

10.1109/JSYST.2019.2931880.

[2] M. H. Mobarak, R. N. Kleiman and J.

Bauman (2021). "Solar-Charged Electric

Vehicles: A Comprehensive Analysis of Grid,

Driver, and Environmental Benefits," in IEEE

Transactions on Transportation

Electrification, vol. 7, no. 2, pp. 579-603,

doi: 10.1109/TTE.2020.2996363.

[3] D. Phan, A. Bab-Hadiashar, M. Fayyazi, R.

Hoseinnezhad, R. N. Jazar and H. Khayyam

(2021), "Interval Type 2 Fuzzy Logic Control

for Energy Management of Hybrid Electric

Autonomous Vehicles," in IEEE Transactions

on Intelligent Vehicles, vol. 6, no. 2, pp. 210-

220, doi: 10.1109/TIV.2020.3011954.

[4] B. Sah, P. Kumar and S. K. Bose (2021), "A

Fuzzy Logic and Artificial Neural Network-

Based Intelligent Controller for a Vehicle-to-

Grid System," in IEEE Systems Journal, vol.

15, no. 3, pp. 3301-3311,doi:

10.1109/JSYST.2020.3006338.

[5] M. Suhail, I. Akhtar, S. Kirmani and M.

Jameel (2021), "Development of Progressive

Fuzzy Logic and ANFIS Control for Energy

Management of Plug-In Hybrid Electric

Vehicle," in IEEE Access, vol. 9, pp. 62219-

62231, doi: 10.1109/ACCESS.2021.3073862.

[6] J. Li, Q. Zhou, H. Williams and H. Xu (2020),

"Back-to-Back Competitive Learning

Mechanism for Fuzzy Logic Based

Supervisory Control System of Hybrid

Electric Vehicles," in IEEE Transactions on

Industrial Electronics, vol. 67, no. 10, pp.

8900-8909, doi: 10.1109/TIE.2019.2946571.

[7] G. C. Lazaroiu and M. Roscia (2022), "Fuzzy

Logic Strategy for Priority Control of Electric

Vehicle Charging," in IEEE Transactions on

Intelligent Transportation Systems, vol. 23,

no. 10, pp. 19236-19245, doi:

10.1109/TITS.2022.3161398.

[8] Hayat El Aissaoui, Abdelghani El Ougli,

Belkassem Tidhaf (2022). MPPT Using PSO

Technique Comparing to Fuzzy Logic and

P&O algorithms for Wind Energy Conversion

System. WSEAS Transactions on Systems and

Control. vol. 17, pp.305-313,

https://doi.org/10.37394/23203.2022.17.35.

[9] Katuri, R., Gorantla, S. (2020). Optimal

performance of lithium-ion battery and ultra-

capacitor with a novel control technique used

in E-Vehicles. Journal of New Materials for

Electrochemical Systems, Vol. 23, No. 2, pp.

139-150, doi: 10.14447/jnmes.v23i2.a11.

[10] S. Arandhakar, N. Jayaram, Y. R. Shankar,

Gaurav, P. S. V. Kishore and S. Halder

(2022), "Emerging Intelligent Bidirectional

Charging Strategy Based on Recurrent Neural

Network Accosting EMI and Temperature

Effects for Electric Vehicle," in IEEE Access,

vol. 10, pp. 121741-121761, doi:

10.1109/ACCESS.2022.3223443.

[11] A. A. Hussein (2019), "Adaptive Artificial

Neural Network-Based Models for

Instantaneous Power Estimation Enhancement

in Electric Vehicles’ Li-Ion Batteries," in

IEEE Transactions on Industry Applications,

vol. 55, no. 1, pp. 840-849, doi:

10.1109/TIA.2018.2866102.

[12] Bin Xu;Xiaosong Hu;Xiaolin Tang;Xianke

Lin;Huayi Li;Dhruvang Rathod;Zoran Filipi

(2020)., "Ensemble Reinforcement Learning-

Based Supervisory Control of Hybrid Electric

Vehicle for Fuel Economy Improvement," in

IEEE Transactions on Transportation

Electrification, vol. 6, no. 2, pp. 717-727. doi:

10.1109/TTE.2020.2991079.

[13] Katuri, R., & Gorantla, S. (2020). Realization

of prototype hardware model with a novel

control technique used in electric vehicle

application. Electrical Engineering, 102(4),

2539-2551, doi: 10.1007/s00202-020-01052-

[14] Shen, J., & Khaligh, A. (2015). A supervisory

energy management control strategy in a

battery/ultracapacitor hybrid energy storage

system. IEEE Transactions on transportation

electrification, 1(3), 223-231, doi:

10.1109/TTE.2015.2464690.

[15] Cao, J., & Emadi, A. (2011). A new

battery/ultracapacitor hybrid energy storage

system for electric, hybrid, and plug-in hybrid

electric vehicles. IEEE Transactions on power

electronics, 27(1), 122-132, doi:

10.1109/TPEL.2011.2151206.

[16] Chiheb Ben Regaya, Fethi Farhani, Hichem

Hamdi, Abderrahmen Zaafouri, Abdelkader

Chaari (2020). Robust ANFIS Vector Control

of Induction Motor Drive for High-

Performance Speed Control Supplied by a

Photovoltaic Generator. WSEAS Transactions

on Systems and Control. vol. 15, pp.356-365,

https://doi.org/10.37394/23203.2020.15.37.

[17] Khaldi Hamza, Mounir Hamid, Boulakhbar

Mouaad (2022). A Review of Future Fuel Cell

Electric Vehicles and Challenges Related to

Morocco. WSEAS Transactions on Power

Systems. vol. 17, pp.339-353,

https://doi.org/10.37394/232016.2022.17.34.

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

200

Volume 23, 2024

[18] X. Li and S. Wang (2021), "Energy

management and operational control methods

for grid battery energy storage systems," in

CSEE Journal of Power and Energy Systems,

vol. 7, no. 5, pp. 1026-1040, doi:

10.17775/CSEEJPES.2019.00160.

[19] Ning Yan; Yao Zhong; Xiangjun Li; Yulong

Wang; Longtao Su; Wan Jiang; Jianxiang

Zhou (2021), "Energy Management Method

of Electricity Heat Hydrogen Multi-Coupling

System for Retired Power Battery Echelon

Utilization in Microgrids," in IEEE

Transactions on Applied Superconductivity,

vol. 31, no. 8, pp. 1-5, doi:

10.1109/TASC.2021.3110471.

[20] Mostafa M. E. F. Shams, Hosam Abd El- Aziz

Amr, Rania Rushdy Moussa (2021).

Comparative Analysis on the Effects of

Natural & Artificial Indoors Lighting on the

Learning and Interactive Process. WSEAS

Transactions on Power Systems. vol. 16,

pp.121-127,

https://doi.org/10.37394/232016.2021.16.12.

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

The authors equally contributed in the present

research, at all stages from the formulation of the

problem to the final findings and solution.

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

No funding was received for conducting this study.

Conflict of Interest

The authors have no conflicts of interest to declare.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS

DOI: 10.37394/23201.2024.23.20

Rakesh Babu Bodapati,

R. S. Srinivas, P. V. Ramana Rao

E-ISSN: 2224-266X

201

Volume 23, 2024