Monday , June 18 2018

Speed Control of an Asynchronous Motor Using PID Neural Network

Department of Information Technology, Haydarpasa Vocational High School
Istanbul, Turkey

Department of Electronics and Computer Education, Faculty of Technical Education, Marmara University

Istanbul, Turkey

Abstract: This paper deals with the structure and characteristics of PID Neural Network controller for single input and single output systems. PID Neural Network is a new kind of controller that includes the advantages of artificial neural networks and classic PID controller. Functioning of this controller is based on the update of controller parameters according to the value extracted from system output pursuant to the rules of back propagation algorithm used in artificial neural networks. Parameters obtained from the application of PID Neural Network training algorithm on the speed model of the asynchronous motor exhibiting second order linear behavior was used in the real time speed control of the motor. The real time control results show that reference speed successfully maintained under various load conditions.

Keywords: PID, neural network, PIDNN, control.

>>Full text
V. Ayhan MARABA, A. Emin KUZUCUOGLU, Speed Control of an Asynchronous Motor Using PID Neural Network, Studies in Informatics and Control, ISSN 1220-1766, vol. 20 (3), pp. 199-208, 2011.

1. Introduction

The basic purpose of control systems is to keep system behaviour at desired values. A controller located in a closed loop control system generates the control signal required to keep the output of the system at the desired value. In classical control methods, the selection of the controller to be used in the control of systems and for the detection of The parameters that are determined in this way cannot always provide the desired system stability due to various factors, such as modelling mistakes, changes in the parameters of the controlled system, and disruptive effects. Due to all of these problems in classical control methods, practitioners began to use artificial neural networks (ANNs) in the control field because they have the ability to learn and generalize, and the derivation of a mathematical equation is not required [1,2]. Today, most of the systems used in industry exhibit non-linear, time-delay behaviour. These systems have excessive overshoot and high settling times, and they are not stable. It is very difficult and demanding to design a controller for such systems using classical methods. Controllers can be designed by using various methods if mathematical models or transfer functions are available that represent the behaviour of the system very well. However, it is rather difficult to develop mathematical models of such systems in practice [3].

Classical Proportional Integral and Derivative (PID) controllers are immensely preferred in many areas of industrial control, especially in the chemical industry due to their simple structure and high durability. Although PID controllers are used for controlling many systems, it is difficult to find optimum parameters which are proportional, integral and derivative (KP, KI, KD) for the control of time-delay and non-linear systems [3].

Fuzzy logic has been used to improve the performance of PID-controlled, non-linear systems. The fuzzy logic controller modified the parameters of the PID controller to get better system response [4]. Complex devices, such as hyper-redundant robots, can be controlled with a PID control algorithm. The determination of the combinations of proportional, integral, and derivative parameters, as well as the optimum values of these parameters, is a time-consuming operation [5].

PID-Neural Network (PIDNN) controllers perform adaptive control. Uncertainties in systemic and environmental factors ensure that the controller exhibits acceptable behavior by means of adaptive control. However, in practice, there are many problems that must be solved, and this is considered to be one of the main disadvantages of such controllers. The main problems are the slow learning rate, the slow approximation to the reference value, and the uncertainties in the parameters of the controller [3].

A PIDNN controller includes the advantages of the PID algorithm and the neural network. Such a controller is not a hybrid structure that consists of artificial neural networks and a PID controller [6]. The PID algorithm exists in the neurons located in the neural structure as an activation function.

There are P-proportional, I-Integral, D- derivative neurons in the PIDNN structure, and the connection weights that arise among these neurons are updated by using a back-propagation training algorithm according to the error value propagated through the system [3].


  1. DANDIL B., Plant Control By Aid Of Artificial Neural Networks, Master Thesis, Firat University Institute of Applied Sciences, 1997, (in Turkish).
  2. PATIC, P. C., R. ZEMOURI, L. DUTA, Recurrent Neural Networks in Linear Systems Controlling, Studies in Informatics and Control, vol. 19, no. 2, 2010, pp. 153-158.
  3. SHU, H., Y. PI, PID Neural Networks for Time-delay Systems, Computers and Chemical Engineering, Volume 24, Issues 2-7, 15 July 2000, pp. 859-862.
  4. ABDEL-HADY, F., S. ABUELENIN, Design and Simulation of a Fuzzy-Supervised PID Controller for a Magnetic Levitation System, Studies in Informatics and Control, vol. 17, no. 3, 2008, pp. 315-328.
  5. IVĂNESCU, M., M. C. FLORESCU, N. POPESCU, D. POPESCU, The Control of The Hyper-redundant Manipulators by Frequency Criteria, Studies in Informatics and Control, vol. 18, no. 3, 2009, pp. 279-288.
  6. SHU, H., X. GUO, H. SHU, PID Neural Networks in Multivariable Control Systems, International Symposium on Intelligent Control, Vancouver, Canada, 2002.
  7. SHU, H.,X. GUO, Decoupling Control of Multivariable Time-Varying Systems Based on PID Neural Network, 5th Asian Control Conference, 2004, Melbourne, Australia.
  8. SHU, H., H. SHU, Simulation of PID Neural Network Control System with Virtual Instrument, 7th International Conference on System Simulation and Scientific Computing, Beijing, China, 2008.
  9. SHU, H., Y. PI, Decoupled Temperature Control System Based on PID Neural Network, ACSE Conference, Cairo, Egypt, 2005.
  10. RONCO E., P. J. GAWTHROP, Neural Networks for Modelling and Control, Centre for System and Control Department of Mechanical Engineering University of Glasgow, Technical Report csc97008, 1997.
  11. LENNART L., System Identification: Theory for the user, 2nd edition, Prentice Hall PTR, 1999.