B. E.

ELECTRICAL AND ELECTRONICS ENGINEERING

CHOICE BASED CREDIT SYSTEM (CBCS) AND OUTCOME BASED EDUCATION (OBE)
SEMESTER – VII
AI TECHNIQUES FOR ELECTRIC AND HYBRID ELECTRIC VEHICLES (PROFESSIONAL ELECTIVE)
Course Code 18EE743 CIE Marks 40
Teaching Hours/Week (L: T: P) (3:0:0) SEE Marks 60
Credits 03 Exam Hours 03
Module-2
Brushless Direct Current Motor Drive Using Artificial Intelligence for Optimum Operation of the Electric
Vehicle: Basics of Artificial Intelligence, Advantages of Artificial Intelligence in EV, Brushless DC Motor,
Mathematical Representation Brushless DC Motor, Closed-Loop Model of BLDC Motor Drive, PID Controller,Fuzzy
Control, Auto-Tuning Type Fuzzy PID Controller, Genetic Algorithm, Artificial Neural Network-Based Controller,
BLDC Motor Speed Controller with ANN Based PID Controller, Analysis of Different Speed
Controllers.

B. E. ELECTRICAL AND ELECTRONICS ENGINEERING

CHOICE BASED CREDIT SYSTEM (CBCS) AND OUTCOME BASED EDUCATION (OBE)
SEMESTER – VII
AI TECHNIQUES FOR ELECTRIC AND HYBRID ELECTRIC VEHICLES (PROFESSIONAL ELECTIVE)
Course Code 18EE743 CIE Marks 40
Teaching Hours/Week (L: T: P) (3:0:0) SEE Marks 60
Credits 03 Exam Hours 03
Module-2
Brushless Direct Current Motor Drive Using Artificial Intelligence for Optimum Operation of the Electric
Vehicle: Basics of Artificial Intelligence, Advantages of Artificial Intelligence in EV, Brushless DC Motor,
Mathematical Representation Brushless DC Motor, Closed-Loop Model of BLDC Motor Drive, PID Controller,Fuzzy
Control, Auto-Tuning Type Fuzzy PID Controller, Genetic Algorithm, Artificial Neural Network-Based Controller,
BLDC Motor Speed Controller with ANN Based PID Controller, Analysis of Different Speed
Controllers.

1. What is artificial Intelligence (AI) and how machine learning & deep learning is mapped to
AI? And what are the advantages of AI in Electric vehicles?
Artificial Intelligence: Any technology that enables the machine to mimic human intelligence
or any rule based application that simulates human intelligence.
Machine Learning (ML): A subset of AI that incorporates maths & statistics in such a way
that it allows the application to learn from the data.
Deep Learning: A subset of ML that uses neural network and handles unstructured data to
solve complex problems.
AI is the ability of a control process to suitably understand peripheral values, do proper
training and study of the past statistics, and utilizes this in implementing precise goals and
tasks through flexible alteration.
The purpose of AI is mainly focussed on a fuzzy & Neural network based controller in EV
control.
Advantages
1. The conventional control system is designed for a specific application & may not be
suitable for unfamiliar contribution of data. AI has capability to handle unfamiliar
contribution of data.
2. The controller of AI is more accessible to fiddle with the standard controller and is
flexible to the adaptation of new parameters or new sequences.
3. The fuzzy logic controller, for example, responds more quickly, and the percentage
overshoot becomes very less. It’s easier to employ.
4. The development of AI from basic ML to advanced deep learning models have made
it capable to solve complex problems in EVT.
5. Artificial intelligence in automation development can promote innovations and
overall progress in electrical vehicle control.

2. Explain the construction parts and working principle of BLDC motor? And hence list the
advantages, disadvantages and applications of BLDC motor?
Brushless DC Motor
A BLDC electric motor is an electronically switched synchronous DC motor, which are
supplied by DC via an inverter which produces a sequential electrical current to drive the
motor through control loop.
The BLDC motor controller maintains the required output of the motor by giving an exact
energizing pulse to the stator winding

Fig. Cross sectional image of BLDC motor

The BLDC motor structure contains a stator with armature winding (star/delta) and a
rotor with a permanent magnet (NdFeB-Neodymium Iron Boron) which is similar to
PMSM.
The position sensor such as hall sensors installed in the motor can detect the rotor
position & transform it into an electrical signal, providing correct commutation
information for the logic switch circuit.
The main design principle of a BLDC motor is to replace the mechanical commutator by
using an electrical switch circuit. In order to control the motor rotation speed & direction,
a rotor position sensor (hall sensor) a control circuit, together with a power Inverter
must be included in a BLDC motor system

Compared with other kinds of motors, the BLDC motor is excited by a square wave, so
that the motor has lots of advantages such as
(1) Higher permanent magnet utilization
(2) Larger motor torque
(3) Higher efficiency
(4) Reliability
Advantages: compared to brushed DC motor, BLDC motor has the following advantages
1. High torque to weight ratio
2. High speed
3. Attain zero RPM and finite torque
4. Increased efficiency at no-load, low load & full load conditions
5. Improved performance
6. Condense noise
7. Longer service life
8. Maintenance cost is low
9. Elimination of switching ionizing sparks
10. Least EMI [Electromagnetic Interference]
11. No frictional losses due to absence of brushes
Disadvantages
1. Less robust
2. More complex
3. Costly electronic console
4. High temperature weakens the permanent magnets
Applications
1. CD drives 2. printers 3. hand operated power tools 4. Air-craft 5. Spaceships
6. Gramophone Recordings 7. Automotive includes EV, HEV, private transporters, E-
bicycle and self-balancing vehicles etc. 8. Power steering 9. BLDC fans

3. With the help of block diagram, explain the overall model or Closed loop model of BLDC
motor for EV?
 Figure: The overall model of BLDC motor for EV
The closed loop system of BLDC motor for EV application is shown in figure. It basically
consists of BLDC motor, voltage source inverter, DC current source, rotor position sensor
and control unit to generate gate signals to VSI. The two main types of popular drives are
voltage source and current source inverters based BLDC drives.
The inner loop of the drive consists magnetic sensor (hall sensor), which is used to
provide the information about rotor position of the BLDC drive, and based on that
information, the gate signal generator generates the commutating signals for three phase
VSI.
The electronic commutation provides the rotor and stator of BLDC motor to run at the
same frequency, and that is why it is called one type of synchronous motor.
There are different command signals like torque, voltage, speed, the current, which are
used to generate the control signal for the system.

4. Derive an expression for output to input transfer function of PI and IP controller. And
hence discuss about their prime features.
R(S): Input signal
C(S): Output signal
Kp, Ki: proportional and integral constants
Km: motor constants
TL(s): Load torque
E(s): Error signal
Tm: time constant of the motor
Both the controllers, P-I and I-P, as shown in Figures 1 and 2, respectively, minimizes the steady-state
error of a process, but the P-I controller takes less time. The over-shoot is less for the I-P
controller for a step input.

PI controller:
i. Transfer function considering input R(S), [TL(S)=0]
𝐶(𝑆) 𝐺(𝑆)
=
𝑅(𝑆) 𝑇𝐿(𝑆)=0 1 + 𝐺𝐻(𝑆)

𝑆𝐾𝑝 + 𝐾𝑖 𝐾𝑚 𝐾𝑚(𝑆𝐾𝑝 + 𝐾𝑖)

𝐺(𝑆) = ( ) =
𝑆 1 + 𝑆𝑇𝑚 𝑆(1 + 𝑆𝑇𝑚)

H(S)=1 (Feedback signal)

𝐶(𝑆) 𝐾𝑚(𝑆𝐾𝑝 + 𝐾𝑖)/𝑆(1 + 𝑆𝑇𝑚)
=
𝑅(𝑆) 𝑇𝐿(𝑆)=0 1 + 𝐾𝑚(𝑆𝐾𝑝 + 𝐾𝑖)/𝑆(1 + 𝑆𝑇𝑚)

𝐶(𝑆) 𝐾𝑚(𝑆𝐾𝑃 + 𝐾𝑖)

= 2 → (1)
𝑅(𝑆) 𝑇𝐿(𝑆)=0 𝑆 𝑇𝑚 + 𝑆(1 + 𝐾𝑚𝐾𝑝) + 𝐾𝑚𝐾𝑖

𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑃𝐼 𝑐𝑜𝑛𝑡𝑟𝑜𝑙𝑙𝑒𝑟 @ 𝑇𝐿(𝑆) = 0

ii. Transfer function considering load torque TL(S) as input [R(S)=0]

𝐾𝑚
𝐺(𝑆) = : 𝐹𝑜𝑟𝑤𝑎𝑟𝑑 𝑔𝑎𝑖𝑛 𝑠𝑖𝑔𝑛𝑎𝑙
1 + 𝑆𝑇𝑚
𝑆𝐾𝑝 + 𝐾𝑖
𝐻(𝑆) = : 𝐹𝑒𝑒𝑑𝑏𝑎𝑐𝑘 𝑔𝑎𝑖𝑛 𝑠𝑖𝑔𝑛𝑎𝑙
𝑆
𝐶(𝑆) 𝐺(𝑆)
=
𝑅(𝑆)𝑅(𝑆)=0 1 + 𝐺𝐻(𝑆)

𝐶(𝑆) 𝐾𝑚/(1 + 𝑆𝑇𝑚)

=
𝑅(𝑆)𝑅(𝑆)=0 1 + 𝐾𝑚 ∗ 𝑆𝐾𝑝 + 𝐾𝑖
1 + 𝑆𝑇𝑚 𝑆
𝐶(𝑆) 𝑆(𝐾𝑚)
= 2
𝑅(𝑆)𝑅(𝑆)=0 𝑆 𝑇𝑚 + 𝑆(1 + 𝐾𝑚𝐾𝑝) + 𝐾𝑚𝐾𝑖

IP controller
i. Transfer function considering input R(S), [TL(S)=0]
Solving for inner loop at TL(s)=0

𝐶(𝑆) 𝐾𝑚 𝐾𝑖
= 2 → (3) [IP controller]
𝑅(𝑆) 𝑇𝐿(𝑠)=0 𝑆 𝑇𝑚 + 𝑆(1 + 𝐾𝑚 𝐾𝑃 ) + 𝐾𝑚 𝐾𝑖

5. Derive an expression for mathematical equations that constitute the differential

equation mathematical model of BLDC motor.
Mathematical representation of BLDC motor
The mathematical model of BLDC motor is fundamental for the corresponding
performance analysis & control system design. The BLDC motor generally consists of 3
parts
1. Motor structure
2. Power driving circuit
3. Position sensor
The three phase synchronous machine & the BLDC motor has the same mathematical
model supported by voltage source inverter [VSI] as shown in figure

Fig. 1 VSI fed BLDC motor

Fig. 2 Equivalent circuit of BLDC motor

The dynamic equation of phase voltages of three phase star connected stator are mentioned as
follows
 Where
Ls: armature self-inductance
Ms: armature mutual inductance
Rs: armature resistance
Van, Vbn, Vcn: phase voltage at the terminals
Ia, Ib,Ic: motor input currents
Ea, Eb,Ec: back emf of the motor
As the delivery of flux is trapezoidal, that is non-sinusoidal, it is wise to develop a phase variable
model for PMBLDCM.
The development of this phase variable model requires certain assumptions which are as follows
1. Iron & stray losses are not taken into account [stray losses: friction & windage losses]
2. Ignore armature reaction and the distribution of air-gap magnetic field is thought to be a
trapezoidal wave with a flat-top width of 120 electrical degrees [arm. Reaction: effect of
armature flux on main flux]
3. Power switches and flywheel diodes of the inverter circuit have ideal switch features
[on state resistance rds=0, delay time td=0]
The stator voltage equation of equations (1), (2) and (3) are written in matrix form

𝑉𝑎𝑛 𝑟𝑠 0 0 𝐼𝑎 𝐿𝑆 𝑀𝑆 𝑀𝑆 𝐼𝐴 𝐸𝑎
𝑑
[𝑉𝑏𝑛 ]=[0 𝑟𝑠 0 ] [𝐼𝑏 ] + [𝑀𝑆 𝐿𝑆 𝑀𝑆 ] [𝐼𝐵 ] + [𝐸𝑏 ] -> (4)
𝑑𝑡
𝑉𝑐𝑛 0 0 𝑟𝑠 𝐼𝑐 𝑀𝑆 𝑀𝑆 𝐿𝑆 𝐼𝐶 𝐸𝑐

For balanced condition, the summation of phase currents of stator is considered to be zero,
that is Ia+Ib+Ic=0, which provides simplified inductance matrix as follows
 𝑽𝒂𝒏 𝒓𝒔 𝟎 𝟎 𝑰𝒂 𝑳𝑺 − 𝑴𝑺 𝟎 𝟎 𝑰𝑨 𝑬𝒂
𝒅
[𝑽𝒃𝒏 ]=[ 𝟎 𝒓𝒔 𝟎 ] [𝑰 𝒃 ] + [ 𝟎 𝑳𝑺 − 𝑴𝑺 𝟎 ] [𝑰𝑩 ] + [𝑬𝒃 ] -> (8)
𝒅𝒕
𝑽𝒄𝒏 𝟎 𝟎 𝒓𝒔 𝑰 𝒄 𝟎 𝟎 𝑳𝑺 − 𝑴𝑺 𝑰𝑪 𝑬𝒄

Induce emf 𝐸𝑝 = (𝐵𝑙𝑣)𝑁 = 𝑁(𝐵𝑙𝑟𝜔) [𝑣 = 𝜔𝑟)

𝐸𝑝 = 𝑁(𝜑𝑎 )𝜔 = 𝑝 𝜔 → (9) [𝑝 = 𝑁𝜑𝑎 ]

Where
N: No. of conductors arranged in series per phase
𝑣: velocity in m/s
𝑙: length of the conductor in ‘m’
r: radius of rotor bore in ‘m’
𝜔: angular velocity in rad/sec
B: magnetic flux density in T

The electromagnetic torque of the machine is

𝑻𝒆 = [𝒆𝒂𝒔 𝒊𝒂 + 𝒆𝒃𝒔 𝒊𝒃 + 𝒆𝒄𝒔 𝒊𝒄 ]/𝝎 → (10)
Where
𝑒𝑎𝑠 = 𝑓𝑎𝑠 (𝜃𝑟 )𝑝 𝜔

𝑒𝑏𝑠 = 𝑓𝑏𝑠 (𝜃𝑟 )𝑝 𝜔

𝑒𝑐𝑠 = 𝑓𝑐𝑠 (𝜃𝑟 )𝑝 𝜔

The function 𝑓𝑎𝑠 (𝜃𝑟 ), 𝑓𝑏𝑠 (𝜃𝑟 ), 𝑓𝑐𝑠 (𝜃𝑟 )ℎ𝑎𝑣𝑒 𝑎𝑛 𝑖𝑑𝑒𝑛𝑡𝑖𝑐𝑎𝑙 𝑠ℎ𝑎𝑝𝑒 𝑎𝑠 𝑜𝑓𝑒𝑎𝑠 ,𝑒𝑏𝑠 ,
𝑒𝑐𝑠 𝑤𝑖𝑡ℎ 𝑡ℎ𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 𝑝𝑒𝑎𝑘 𝑎𝑠 + 1 and negative peak as -1.
The electromagnetic torque after incorporating the above function is
𝑇𝑒 = 𝑝 [𝑓𝑎𝑠 (𝜃𝑟 )𝑖𝑎 + 𝑓𝑏𝑠 (𝜃𝑟 )𝑖𝑏 + 𝑓𝑐𝑠 (𝜃𝑟 )𝑖𝑐 ] 𝑁𝑚 → (11)

The reason for naming this machine as a DC motor is that the phase voltage equation of the
BLDC motor & armature voltage equation of the DC machine is identical. The motor equation
is as follows
𝒅𝝎
𝑱 + 𝑩𝝎 = (𝑻𝒆 − 𝑻𝑳 ) → (𝟏𝟐)
𝒅𝒕
Where
J: Inertia in kg𝑚2
B: Friction coefficient in Nm(𝑟𝑎𝑑/𝑠)−1
𝑇𝑒 : Electromagnetic torque in Nm
𝑇𝐿 : Load torque in Nm
The equations (8), (10) and (12) constitute the differential equation mathematical model of
BLDC motor

6. What is fuzzy Logic controller? And hence compare fuzzy logic with crisp value systems
with illustrations?

The term fuzzy logic (FL) was introduced by Lotfi Zadeh of the university of California
at Berkeley in the year 1960. FL is an approach for computing based on “degree of
freedom” rather than the usual true or false (1/0) Boolean logic which the modern
computer design is based upon.
In more simple words, A FL state can be 0 or 1 or in between values such as 0.15, 0.3,
0,75 etc.
Crisp vaue: is precise in nature. It deals with values that have a strict boundaries ie.
True or false (1/0). In Boolean system, true value 1 represents absolute truth value
and 0 represents absolute false value.
In fuzzy logic, there are intermediate values present which are partially true &
partially false.
7. With the help of membership functions, rule base table and block diagram, explain
fuzzy logic controller applied to BLDC motor drive?

Figure 1 (a) Triangle, (b) Trapezoid (c) bell membership functions

1. Figure 1 demonstrates various types of fuzzy membership functions. For fuzzy logic
control system models, the most important things are the selection of membership
functions for inputs and outputs and based on these the design of if–then law, i.e., the
fuzzy rule base is adopted.
2. An membership function MF is a curve that defines how the value of fuzzy variable in
a certain region are mapped to membership value or degree of freedom () between
0 and 1.
[MF: membership function=> triangular, Gaussian, Trapezoidal, bell etc.
Fuzzy variable(u): speed, voltage (input or output variable)
𝜇: membership value or degree of membership ]
3. Instead of mathematical equations, Fuzzy logic presents functional linguistic laws.
However, the linguistic term of fuzzy logic provides a feasible method to define
system’s operational characteristics.
 4. Fuzzyfication: The process of generating a membership value () for a fuzzy variable
using a membership function
5. Defuzzification: The process of transforming a fuzzy output of a fuzzy system into a
crisp output.
6. Rule base: IF/THEN rule relating fuzzy variables input (condition) to output fuzzy
variable (action)
7. The overall fuzzy logic system is shown in Figure 2. BLDC drive operation is
controlled in two ways; torque is the first, and speed is the second. These two
parameters are controlled simultaneously by the fuzzy logic controllers. The first loop
contains current control, and the second loop contains BLDC drive speed.
8. Fuzzyfier, interface, and de-fuzzyfier become the three traits (norms) of abstract
controllers. The number of membership functions determines the superiority of
control by using a fuzzy controller, and the control quality depends on the number of
membership functions.

Fig. 2 General Fuzzy controller

Table-1: Rule base fuzzy logic controller

To describe the functions, seven linguistic variables are used for both input and output variables.
Table-1 displays the membership function logic rule base. From this table, it can be found that the
combinations of two inputs, p1 (error) and p2 (change in error), provides different outputs like NB
[Negative big], NM [Negative medium], NS [Negative small], Z [Zero], PB [Positive big], PM [Positive
medium], PS [Positive small]. These outputs decide the different control actions of the system by
defuzzyfication.

8. With the help of block diagram representation, explain the Auto-tuning type Fuzzy PID
control system.
While Ziegler and Nichols have proposed an efficient technique to adjust a PID controller’s
coefficients and improve performance by optimizing the PID parameters using different
optimization techniques but cannot guarantee that it will always be active. PID controller
self-adjustment is required, and this fuzzy-PID controller meets the need.
As shown in Figure 1, the controller consists of two parts: the traditional PID controller and
the self-tuning Fuzzy Logic Control (FLC) part. Now the PID controller’s control operation
after self-tuning can be defined as a self-tuning Fuzzy PID controller. Fuzzy self-tuning PID
controller modifies the online PID parameters using fuzzy rules.

Fig 1. Block representation of the Auto-tuning type Fuzzy PID control system

Auto-Tuning Type Fuzzy PID Controller (continued)

Here the error and the error shift frequency [change in error] are denoted by e and ec,
respectively. The terms Ke and kec are quantitative variables, ku is the ranking factor from
all dimensions of the system, i.e., the stability, response rate, over-shoot etc.
The function of PID controller gains are as follows.
(1) The proportional gain Kp was designed to speed up the system reaction time and
increase the system’s regulatory accuracy.
(2) The integral constant KI reduces the steady state error of the system and also reduces
the rise time.
(3) And the derivative controller Kd is designed to improve the dynamic characteristics,
mainly reduces the change in error in any direction in response to the process before
the error occurs.

Introduction to Neural Network

An artificial neural network is an attempt to simulate the network of neurons that make
up a human brain so that computer will be able to learn things and make decisions in a
human like manner.
Neurons are cells within nervous system that transmits information to other nerve cells, muscles,
or gland cells. Most neurons have a cell body, an axon, dendrites.
Dendrites extend from the neuron cell body and receive messages from other neurons. Synapses
are the contact points where one neuron communicates with another. Axon take away
information from cell body. The contact point between two neuron cell is called synapse.

9. Explain artificial neural network controller and hence discuss the ANN-based PID controller
 An artificial neural network act as a controller by monitoring and altering the working
condition of a dynamic system by using different types of signal.
 The neural network can be implemented as a nonlinear controller if the neural nets are
successfully used. There are several methodologies to train the ANN controller.
 A neural network consists of different layers such as Input layer, hidden layer and output
layer.
 The input layer nodes represent linguistic input variables.
 The hidden layer has two layers such as membership layer and rule base layer.
 The input layer content is altered using weights associated with each connector link and then
fed to membership layer.
 The content of membership layer is connected to different nodes of rule base layer in certain
pattern with weight associated with each link.
 The output of rule base layer is connected to output layer with different weights in each link.
 Figure 1 Neural Network Layers

Figure 2 ANN-based PID controller for BLDC motor of EV system

The overview of BLDC motor drives with a conventional and the ANN-PID speed controller is
shown in Figure 2.
The traditional feedback controller has many applications in industrial as well as commercial
applications. This controller works on the optimal setting of the plant. If any of the parameters of
the plant are removed, or a new parameter is added, the controller action is disturbed because it
is a fixed gain feedback controller.
Therefore, the controller needs to be calculated repeatedly to get a new ideal setting. The system
which is operated under variable time delay, large nonlinearity, and disturbance, the classical PID
controller is not able to provide ultimate control action for that system.
For that purpose, PID controller combining with an artificial neural network (ANN) can be used.
This approach is called a Neuro-PID intelligent controller. ANN controller is involved with any
conventional controller like PI, IP, Fuzzy, to obtain the desired result in process or control
industries.
PID Controller-Based on Neuro-Action
ANN adaptive mechanism is used to measure the disturbance from the output and tunes the
various parameter of the PID controller according to it. With the help of tuning the setting, it can
diminish the noise and thus progresses the operation of the controller. Feed-forward adaptive
control and the back propagation algorithm are mostly considered in different types of
applications among many neural network learning approaches.

10. With the help of flow chart, explain PID controller based on Artificial Neural Network
ANN-Based on PID Controller

Figure 1 PID control algorithm based on Artificial Neural Network

The implementation of the flowchart of the ANN-PID controller is shown in Figure 1. Among the
various control techniques, PID control is a significant method as it is not affected by noise and
constant to change parameters. The purpose of back-propagation [BP] is used in different types
of Neuro-controllers to train them to achieve as much as possible desired plant output.
 The flow chart clearly shows that, it is required to set the ANN structure with respect to
number of nodes ‘n’, weighted coefficient of each layer and learning rate & inertia factor.
 Number of iterations depends upon the number of nodes for up gradations.
 Obtain the input and output signals and calculate the error.
 Calculate the PID controller parameters using ANN
 Perform PID controller output calculations. Based on error, set ANN learning by changing the
weights, threshold values and content of nodes and go for next iterations

11. Compare the Performance of PI, PID, Fuzzy, Genetic algorithm and ANN speed controllers fed
to BLDC motor with respect to settling time, overshoot and rise time

Analysis of Different Speed Controllers

The performance analysis of the BLDC motor based on various parameters like rising and
settling time, peak overshoot with a different type of speed controllers on electric vehicle
is specified in Table 1 below. This comparative analysis is validated, which is shown in
Figure 1, that includes P, PI, and PID of the closed-loop BLDC Drive for Electric Vehicle.
The idea can be established that the addition of each of the gain factors with the P
controller decreases the overshoot of the system, which is best at the time of PID, but it
also makes the system slow.
Table 1 Performance comparison of different speed controllers

In PI controller, ts and overshoot is more

PID controller decreases ts and overshot but tr increases
FLC decreases ts and overshoot compared to PID but tr increases as compared to PID
In GA and ANN, ts, overshoot and tr are reduced very effectively.
 12. Compare the step response of P, PI and PID speed controllers with respect to settling time,
overshoot and rise time

Fig 1 Comparative step response of the closed loop BLDC drive for Electric Vehicle with P, PI and
PID speed controller

This comparative analysis is validated, which is shown in Figure 1, that includes P, PI, and
PID of the closed-loop BLDC Drive for Electric Vehicle. The plot shows that the addition of
the gain factor with P controller decreases the overshoot of the system which is best at the
time of PID, but it also makes the system slow as the ts decreases & also tr increases from P
to PID controller.

12. Compare the step response of Fuzzy, Classical, Genetic algorithm and ANN speed
controllers with respect to settling time, overshoot and rise time

Fig. 1 Comparative step response of the closed loop BLDC drive for electric vehicle with Fuzzy, classical controller, GA
and ANN
Figure 1 clearly shows that the best-optimized output is achieved by implementing an artificial
neural network on this closed-loop control of BLDC motor electrical vehicle drive. Here the
optimized output has zero overshoot and steady state error, and also, the time response analysis
has shown tremendous improvement.
Steady state error and tr decreases gradually while moving from FL towards ANN. However, ANN
is a more important tool to find a reasonable solution to a complex issue at faster phase.

Note: second order time domain response

1. Delay Time (td)

It is the time required for the response to reach 50% or half of its final value from the zero instant.

2. Rise Time (tr)

It is the time required for the response to rise from 0% to 100% of its final value. This is applicable
for the under-damped systems. For the over-damped systems, consider the duration from 0% to
90% of the final value.

3. Peak Time (tp)

It is the time required for the response to reach the peak value for the first time.

4. Peak Overshoot (Mp)

Peak overshoot is defined as the deviation of the response at peak time from the final value of
response.

5. Settling time (ts)

It is the time required for the response to reach the steady state and stay within the specified
tolerance bands around the final value. In general, the tolerance bands are 2% and 5%.