Scribd
Upload
172 views37 pages

FLC&NN Control of DC Motor

This document describes a study on controlling the speed of a DC motor using fuzzy logic control and neural network compensation. It includes the objectives of applying fuzzy logic control, modeling the three-phase rectifier and DC motor, developing fuzzy logic controllers for speed and current, compensating for rectifier non-linearities using fuzzy logic and neural networks, and comparing the performance of PI, fuzzy logic, and neurofuzzy control schemes in simulation. Results show that fuzzy logic control with neural network compensation provides the fastest setting time, lowest overshoot and steady state error compared to PI control. Future work could explore neural network control and neurofuzzy control approaches.

Uploaded by

R.SRIKANTH
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
172 views37 pages

FLC&NN Control of DC Motor

This document describes a study on controlling the speed of a DC motor using fuzzy logic control and neural network compensation. It includes the objectives of applying fuzzy logic control, modeling the three-phase rectifier and DC motor, developing fuzzy logic controllers for speed and current, compensating for rectifier non-linearities using fuzzy logic and neural networks, and comparing the performance of PI, fuzzy logic, and neurofuzzy control schemes in simulation. Results show that fuzzy logic control with neural network compensation provides the fastest setting time, lowest overshoot and steady state error compared to PI control. Future work could explore neural network control and neurofuzzy control approaches.

Uploaded by

R.SRIKANTH
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 37

Contents

Objective
Drive Parameters
System block diagram
Modeling & speed control of DC motor
Three phase controlled rectifiers
Fuzzy Logic &fuzzy control
Neural compensation
Closed loop speed control
Speed control of dc motor using fuzzy control and
neural network compensation
Results & conclusion

Objectives
Applications of fuzzy logic control
Linearizing the rectifier transfer characteristics
Speed and current loops with FLC
Simulation study of PI and proposed scheme
Superiority of FLC and NN over conventional
PI controller

Drive Parameters
Supply voltage (V) -110 volts
Rated motor current (Ia) - 20A
Speed of the motor (r.p.s) 1800
Armature resistance (Ra) 0.6 ohms
Armature Inductance (La) 8mH
Moment of Inertia (J) 0.0465 Kg-m^2
Friction Coefficient (B) 0.004N.m sec/rad
Line voltage (VL) 90 volts
Shaft power 2.5hp
Load Torque (TL) 2.78*10^(-4) *W^2

System Block Diagram

Modeling and speed control


simulation of DC motor

Open loop Armature current and Speed


Response

Three phase controlled rectifiers


Controlled rectifiers are used to get variable dc voltage from an ac source of
fixed voltage.
Controlled rectifiers fed dc drives are also known as Static Ward-Leonard
system.
Simulink implementation of three phase converter

Phase controlled converter fed DC


motor
The converter is used in motoring mode only with fixed field
excitation .The firing angle alpha, required for the rectifier is
generated by cosine wave crossing method with control voltage Vc
obtained form the current controller.
The waveform shown below is without compensation rectifier non linear characterisitcs

Rectifier characteristics

Expressions
forconduction
Rectifier
current
For Continuous
mode when
supplied toand
dc motorvoltage

For Discontinuous conduction mode when supplied to dc


motor

Timing signals for thyristors to fire at


alpha=80degrees

Fuzzy logic
Fuzzy Logic is a problem-solving control system methodology.
It provides a simple way to arrive at a definite conclusion based upon
vague, ambiguous, imprecise, noisy, or missing input information .
Fuzzy Logic approach to control problems mimics how a person
would make decisions, only much faster.
Fuzzy Logic controller processes user-defined rules governing the
target control system.
Fuzzy Logic is not limited to a few feedback inputs and one or two
control output, nor is it necessary to measure or compute rate of
change parameters in order for it to be implemented.
Fuzzy Logic can control nonlinear systems that would be difficult or
impossible to model mathematically.

Fuzzy control
Fuzzy Control applies fuzzy logic to the control of processes by utilizing different categories, usually error and change in error, for
the process state and applying rules to decide a level of output, i.e. a suitable control action.
The linguistic variables used for both input and output variables are often of the form negative large, positive small, zero etc.
A typical rule base for a two-input, single-output system with three membership functions per variable is shown in Table below

DE

N
Z
P

E N Z P

N
N
Z

Z
Z
P

Z
P
P

Simple Rule
Base

Fuzzy logic tool box

Fuzzy speed and current control

Here the speed and current control of membership function


implementation is seen

Rule base matrix

NB

NS

PS

PB

NVB

NVB

PB

PB

PB

PB

NB

NVB

NM

NVB

NS

NVS

NVS

NVS

NS

NVB

NM

NS

NS

NS

NVB

NB

NM

NM

NS

PS

NVB

NVB

NVB

NVB

NM
NB

NM

PM

NB
NB

PB

NVB

NVB

NVB

NB

NB

PVB

NVB

NVB

NVB

NVB

NB

Ia

NB

Fuzzy compensation
The converter which drives the separately excited dc motor may
operate in continuous of discontinuous conduction mode.
The non-linear Vd(pu)- Ia(pu) relation adversely affects the gain
characteristics for the current control loop.
The fuzzy control is well-suited in a non-linear systems especially
where parameter variation problem exists.
Therefore this fuzzy method of del-alpha angle compensation is
used to linearise the converter transfer characteristics in
discontinuous mode.
The input being the armature current (Ia) pu with 9 fuzzy subsets,
the angle (alpha) with 5 fuzzy subsets and the output change in
alpha (dalpha) with 11 fuzzy subsets.

Simulink Implementation
of Fuzzy compensation

Neural network compensation


The required data to train the neural network are obtained by
plotting the relationship between rectifier output voltage and current
given by the expressions.
The input of the neural network are Ia and alpha and the output of
the network is del-alpha.
A two layer network is chosen to do this operation.
Five neurons are chosen for both the hidden layers.
The Neural Network is trained using back-propagation by neural
network training tool GUI.
After training, the network is used as del-alpha compensator in the
system block diagram.

Neural network tool box


implementation
A GUI Network/Data Manager window has its own work area, separate from the
more familiar command-line workspace.
Thus, when running a GUI Network , you can create a network, view it, train it,
simulate it, and export the final results to the workspace. Similarly you can import
data from the work space for use in the GUI
Generally here the problem is the compensation of alpha of phase controlled rectifier
fed dc drive

Here for NN compensation, I used the graphical user interface, nntool

Here we have to create a network and perform compensation of del-alpha.


Now give the input vector containing the different values of Ia , alpha. Once
the network is created, train it and can save the network, its output, etc., by
exporting it to workspace.
To start type nntool in the command window. The following window appears.

Now import data from the workspace and select the given data for input and
target

Now create a network and call it delalpha. And set the network type feed
forward backpropagation and press create.

Now new network is created and delalpha is added to network manager.

Now select the network delalpha and train the network .

Now after the training finished generate a simulink block by giving the
network descriptions.

Attaining the neural block


Now just by generating simulink block diagram and placing that block
used for the compensation

Closed loop speed control

Output waveform

Simulink model of fuzzy logic speed


control of dc motor with fuzzy
compensation

Output wave form

Simulink model of fuzzy logic speed


control of dc motor with neural
compensation

Output waveform

Results and Conclusion


Parameter

PI Controller

Fuzzy Logic
Controller and
Compensation

FLC and
Neural
Network
Compensation

Setting Time Ts
Over shoot(%)

0.2sec
6%

0.16sec
0 rad/sec

0.14sec
0rad/sec

Rise time tr
Steady State
Error

0.17 sec
3 rad/sec

0.1 sec
0 rad/sec

0.09sec
0 rad/sec

The delalpha compensation is provided for linearization of the transfer characteristics


in the discontinuous mode of conduction
The comparison of results shows that the Fuzzy logic controller and neural network
compensation is superior than PI controller.

Future Scope
Till now the speed of dc motor is controlled using PI, Fuzzy controllers.
The next approach is to control the speed using a neural network controller and
neurofuzzy control.
Another new approach is using a DSP controller. With the DSP controller an
intelligent control approach is possible to reduce the overall system cost and to
improve the reliability of the drive performance.

References
1. Gilbert C.D. Sousa and B.K. Bose, A Fuzzy Set Theory Based Control of a
Phase-Controlled DC Machine Drive, lEEE Trans,Ind. Appl, vol. 30, Jan 1994.
2. C.L. Chen and C.T. Hsieh, User friendlydesign of Fuzzy Logic Controller, IEE.
Control Theory and applications.
3. J.S. Mapes and B.K. Bose, Linearisation of the transfer characteristics of a phasecantrolled converter under discontinuous conduction, IEEE Trans, Ind. Appl, vol.
1A-i4, pp, 559-564, 1978.
4. J.M. Zurada, Jntroduction to Artificial Neural Systems, Jaico Publishing House,
First Edition, 1994.
5. Drainkov.D, Hel1endoorw.H & Reinfiank.M -An Introduction to Fuzzy Control,
Narosa.publishing house, I Reprint, 1996.
6. Muhammad H. Rashid, Power Electronics Circuits, Devices and Applications,
Prentice
Hall of Jndia, I1 edition, June 1997.

Thank you

You might also like