NEURAL GENERALIZED PREDICTIVE CONTROL WITH REFERENCE CONTROL MODEL FOR AN INDUCTION MOTOR DRIVE

A. Merabet, M. Ouhrouche, and R.T. Bui

Keywords

Induction motor, generalized predictive control, artificial neural networks, inputoutput modelling, reference model control

Abstract

In this paper the authors present a new advanced control algorithm for speed and flux tracking of an induction motor. This algorithm, called neural networks generalized predictive control (NGPC), uses a combination of artificial neural networks (ANN) and generalized predictive control (GPC) technique. The later is traditionally used for systems characterized by a slow dynamics, as in industrial process control. The NGPC algorithm is based on the use of ANN as a nonlinear prediction model of the motor. This modelling technique is done by using I/O data with no need of additional information regarding the machine parameters. The outputs of the neural predictor are the future values of the controlled variables needed by the optimization procedure, which is achieved by minimizing a cost function with a reference control model using the Newton-Raphson optimization algorithm. Simulation results show the effectiveness of the proposed control method.

Important Links:



Go Back