Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
4Activity
0 of .
Results for:
No results containing your search query
P. 1
Intelligent Controller for Networked DC Motor Control

Intelligent Controller for Networked DC Motor Control

Ratings: (0)|Views: 227 |Likes:
Published by ijcsis
This paper focuses on the feasibility of Neural Network controller for Networked Control Systems. The Intelligent Controllers has been developed for controlling the speed of the Networked DC Motor by exploiting the features of Neural Networks and Fuzzy Logic Controllers. The major challenges in
Networked Control Systems are the network induced delays and data packet losses in the closed loop. These challenges degrade the performance and destabilize the systems. The aim of the proposed Neural Network Controller and Fuzzy Logic Controller schemes improve the performance of the networked DC motor and also compare the results with the Zeigler-Nichols tuned Proportional-Integral-Derivative Controller. The performance of the proposed controllers has been verified through simulation using MATLAB/SIMULINK package. The effective results show that the performance of networked dc motor is improved by using Intelligent Controller than the other controllers.
This paper focuses on the feasibility of Neural Network controller for Networked Control Systems. The Intelligent Controllers has been developed for controlling the speed of the Networked DC Motor by exploiting the features of Neural Networks and Fuzzy Logic Controllers. The major challenges in
Networked Control Systems are the network induced delays and data packet losses in the closed loop. These challenges degrade the performance and destabilize the systems. The aim of the proposed Neural Network Controller and Fuzzy Logic Controller schemes improve the performance of the networked DC motor and also compare the results with the Zeigler-Nichols tuned Proportional-Integral-Derivative Controller. The performance of the proposed controllers has been verified through simulation using MATLAB/SIMULINK package. The effective results show that the performance of networked dc motor is improved by using Intelligent Controller than the other controllers.

More info:

Published by: ijcsis on Dec 04, 2010
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

05/09/2012

pdf

text

original

 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 8, No.8, 2010
Intelligent Controller for Networked DC Motor Control
B.Sharmila
Department of EIE, Sri Ramakrishna Engineering CollegeCoimbatore, Indiasharmi.rajesh@gmail.com
 N.Devarajan
Department of EEE, Government College of Tech.Coimbatore, India 
 Abstract 
 —This paper focuses on the feasibility of Neural Network controller for Networked Control Systems. The IntelligentControllers has been developed for controlling the speed of theNetworked DC Motor by exploiting the features of NeuralNetworks and Fuzzy Logic Controllers. The major challenges inNetworked Control Systems are the network induced delays anddata packet losses in the closed loop. These challenges degradethe performance and destabilize the systems. The aim of theproposed Neural Network Controller and Fuzzy Logic Controllerschemes improve the performance of the networked DC motorand also compare the results with the Zeigler-Nichols tunedProportional-Integral-Derivative Controller. The performance of the proposed controllers has been verified through simulationusing MATLAB/SIMULINK package. The effective results showthat the performance of networked dc motor is improved byusing Intelligent Controller than the other controllers.
 Keywords- Networked Control Systems (NCS); Network Challenges; Tuning; Proportional – Integral - DerivativeControllers (PID); Fuzzy Logic Controller (FLC); Artificial Neural  Networks (ANN).
I.
 
I
 NTRODUCTION
 Networked Control System is the adaptation of communication network for information exchange betweencontrollers, sensors and actuators to realize a closed controlloop. Networks reduce the complexity in wiring connectionsand the costs of Medias. They are easy to maintain and alsoenable remote data transfer and data exchanges among users.Because of these benefits, many industries and institutions hasshown interest in applying different types of networks for their remote industrial control and automation. Regardless of thetypes of networks, the overall performance of NCS is affected by two major challenges as networked induced delay and datalosses. The challenges of networked DC motor are generallycontrolled by Conventional Proportional – Integral -Derivative Controllers, since they are less expensive withinexpensive maintenance, designed easily, and very effective.But mathematical model of the controller and tuning of PID parameters are difficult and generally not used for non-linear systems. Hence to overcome these challenges auto-tuning andadaptive PID Controller was developed with few mathematicalcalculations. The Intelligent controllers as Fuzzy LogicController and Artificial Neural Networks were used toovercoming the challenges. Thus this paper proposesIntelligent Controller for the compensation of the challenges.The novelty of this paper lies in comparison of the applicationof NARMA-L2 Controller and Mamdani Fuzzy LogicController with conventional PID controller for theimprovement of the performance of networked control DCmotor.There are two approaches to utilize a data network asHierarchical Structure and Direct Structure as shown in Fig. 1and Fig. 2 respectively. In the hierarchical structure the dcmotor is controlled by its remote controller at remote stationwhereas in direct structure the central controller is used for controlling the speed of dc motor. Since the hierarchicalstructure has a poor interaction between central and remoteunit, direct structure is preferred.Recently the stability analysis and control design for NCShave attracted considerable research interest [3], [4], [6] and[11]. The work of Nesic and Teel [2] presents an approach for stability analysis of NCS that decouples the scheduling protocol from properties of network free nominal closed-loopsystem. Nesic and Tabbara [3] extended [2] by stochasticdeterministic protocols in the presence of random packetdropouts and inter transmission time and they also proposedwireless scheduling protocol for non-linear NCS in [6]. Thenetworked predictive control scheme for forward and feedback channels having random network delay was proposed in [4],and [5] addresses the problems of how uncertain delays aresmaller than one sampling period which affects the stability of the NCS and how these delays interact with maximumallowable transfer interval and the selected sampling period.Robust feedback controller design for NCS with uncertainty inthe system model and the network induced delay has beenaddressed in [7]-[8], whereas [9] handles networked inductionmotor speed control by using linear matrix n equality (LMI)method. Ref. [1] measure the networked vehicle control performance using an H infinity norm with linear matrixinequalities conditions and markovian jumping parameters incommunication losses. In case of time varying transmissiontimes, model based NCSs has been proposed for stabilization problem of NCS. The stability analysis and controller synthesis problems are investigated in [11] for the NCSs withrandom packet losses by using H infinity control and linear matrix inequalities. A moving horizon method was developed by [12], which was applied as a quantized NCS in a practicalcontext. Since these methods transmit data specifying only aregion in which the measurements lie, it will reduce the
131http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 8, No.8, 2010
network stabilization of the NCS. However, this method couldreduce the stability of the control system by introducinguncertainty in the control system. The issues of limited bandwidth, time delay and data dropouts was taken intoconsideration when NCSs controllers were designed in [12] – [14]. The networked control system performance depends onthe control algorithm and the network conditions. Severalnetwork conditions such as bandwidth, end-to-end delay, and packet loss rate are major impacts on networked controlsystems. Depending upon the control algorithm and network conditions the overall performance of the networked systemmay vary and hence the stability of the system.II.
 
M
ODELLING
 A networked control system can be divided into theremote unit, the central controller and the data network. Fig. 3shows the general block diagram of the networked controlsystem under investigation. In order to focus our discussion onthe performance of networked closed loop control system withnetwork conditions (delay, data loss), a networked dc motor control system has been illustrated.
 A.
 
 Remote Unit 
The Remote Unit consists of the plant (dc motor), sensor and an interfacing unit. Via the network the remote unit cansend measurements like motor speed, current, temperature,and local environment information, back to the centralcontroller. The electro-mechanical dynamics of the dc motor can be described by the loop equation as first order differentialequations.
baaa
e Ridt di Leu
++==
)(
(1)
Figure 1. Hierarchical Structure.Figure 2. Direct Structure.
 
Figure 3. An overall real-time networked control system.
where u=e
a
is the armature winding input voltage; e
 b
=K 
 b
ω isthe back-electromotive-force (EMF) voltage; L is the armaturewinding inductance; i
a
is the armature winding current; R isthe armature winding resistance; K 
 b
is the back-EMF constantand ω is the rotor angular speed. Based on Newton’s law themechanical-torque balance equation is
a
 Ki Bdt  J 
=++
ω ω 
(2)J is the system moment of inertia; B is the system dampingcoefficient; K is the torque constant and T
l
is the load torque.By letting x
1
= i
a
and x
2
= ω, the electromechanical dynamicsof the dc motor can be described by the following state-spacedescription:
u L x L K  x L R x
b
1)(
211
+=
 (3)
 L x J  B x J  K  x
1)(
212
+=
 (4)The parameters of the motor Table 1 are used for determinethe state space model of dc motor.
TABLE
 
1.
 
DC
 
M
OTOR 
P
ARAMETERS
 J Moment of Inertia 42.6 e-6 Kg-m
2
 L Inductance 170 e-3 HR Resistance 4.67 B Damping Coefficient 47.8 e-6 Nm-sec/radK Torque Constant 14.7 e-3 Nm/A
 b
Back EMF constant 14.7 e-3 Vsec/rad
 B.
 
Central Controller 
The central controller will provide the control signal u
C
(t)to the remote systems. The central controller will monitor thenetwork conditions of the remote unit link and provideappropriate control signals to each remote unit. Similarly theoutput responses are taken as feedback signal y
(t) to thecentral controller. The proposed Intelligent Controllers willcompensate the network-induced delays, data losses andexternal disturbances. The data losses and disturbances occur due to missing or disturbances in input reference signal, controlsignal and feedback signal.
132http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 8, No.8, 2010
C.
 
 Data Network 
There are different ways to define network conditions for  point-to-point (from the central control to a specific remoteunit). Two of the most popular network measures are the point-to-point network throughput and maximal delay boundof the largest data. One factor of interest is the sampling time.To keep the illustration simple, the remote unit receives thedata sent from the central controller as u
(t), which can bemathematically expressed as
)()(
 Rc R
uu
τ  
=
(5)where τ
 is the time delay to transmit the control signal u
C
(t)from the central controller to the remote unit. The remote unitalso sends the sensors signals y
(t) of the remote system back to the central controller y
C
(t), and these two signals are relatedas
)()(
 R
 y y
τ  
=
(6)where τ
C
is the time delay to transmit the measured signalfrom the remote unit to the central controller. There are also processing delays as τ
PC
 and τ
PR 
, at the central and remote unit,respectively which could be approximate small constants or even neglected because these delays are usually smallcompared to τ
C
 and τ
.The functions of network variables such as the network throughput, the network management/policy used, the typeand number of signals to be transmitted, the network protocolused, and the controller processing time, and the network traffic congestion condition are taken as the current network conditions n(t) and let z
-t
be a time delay operator whichdefines the signals as
))(,()(
n z uu
 R
c R
=
(7)
))(,()(
n z  y y
c
 Rc
=
(8)In this paper, we have chosen sampling time as 0.5 ms andsimulations are done.III.
 
M
ODELLING
C
ONTROLLER 
D
ESIGN
F
OR 
 NCSIn this session the proposed Neural Network Controller andFuzzy Logic Controller as the central controller is describedand the results are compared with the PID controller.
 A.
 
 Neural Network Controller 
The proposed scheme utilizes the neural-network  NARMA-L2 Controller. The Neural Network Controller isdesigned to take the error as the input and computes the outputstabilizing signal depending on the input error signal. The block diagram of Neural Network Controller for NCS isshown in Fig. 4.The NARMA-L2 controller, a multilayer neural network has two steps involved as system identification stage andcontrol design stage. In system identification stage, a neuralnetwork model of the plant which has to be controlled isdeveloped and in later stage the neural network plant modelhas been designed to train the controller. The ANN plantspecification has been shown in Table 2.
Figure 4. Neural Network Controller for NCS.TABLE
 
2.
 
ANN
 
P
LANT
S
PECIFICATION
  No. of Inputs 3 No. of Outputs 2 No. of Hidden Layers 2 No. of Training Samples 1000 No. of Training Epochs 200
The error signals are trained for number of epochs by usingthe NARMA-L2 controller and the control signal are generatedfor any challenges in the network.
 B.
 
 Fuzzy Logic controller 
In general, fuzzy logic control is used for the control of a plant where the plant modeling is difficult. For such systemsthat are difficult to model, fuzzy logic controller has beensuccessful by Mamdani. The basic principle of fuzzy logic liesin the definition of a set where any element can belong to a setwith a certain degree of membership. Using this idea, theknowledge of an expert can be expressed in a relatively simpleform and the inference for given inputs can be implementedvery efficiently. Due to these advantages, fuzzy logic controlis an attractive method for NCS whose modeling is verydifficult because of the stochastic and discrete nature of thenetwork. Fig. 5 shows the structure of FLC for a single inputsingle output plant. In Fig. 5 r(t) is the reference input, y(t) isthe plant output, e(t) is the error signal between the referenceinput and plant output and u
C
(t) is the control signal.The FLC consists of three parts as 1) Fuzzifier thatconverts the error signal into linguistic values, 2) Inferenceengine that creates the fuzzy output using fuzzy control rulesgenerated from expert experience and 3) Defuzzifier thatcalculate the control input to the plant from the inferredresults. The input and output signals to the FLC are error signal e(t) and control signal u
C
(t) respectively. In this paper,the trapezoidal fuzzy members are selected for membershipfunctions. Three fuzzy linguistic variables, i.e., Small,Medium and Large are defined. The coefficients of themembership function depend upon the set point and aredetermined by several trial and error experiments with the plant without the network. In order for faster execution of thefuzzy logic controller, the Mamdani’s min-max inferencemethod and the central average defuzzifier are used.The rules used in this paper are asIf e(t) is small then u
C
(t) is small
133http://sites.google.com/site/ijcsis/ISSN 1947-5500

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->