Modified adaptive discrete control system containing neural estimator and neural controller

In this paper a modified discrete adaptive control system with neural estim ator and neural controller is presented. The structure of the adaptive cont roller is based on the model presented by Etxebarria (Etxebarria V. Adaptiv e control of discrete systems using neural networks. IEE Proc. Control Theo ry Application, Vol. 141, No. 4, July, 1995) where the stability of the con trol procedure is proved. The Widrow-Hoff procedure of learning and the DAR MA model is used for identifying and adjustment of neural network parameter s, applied to adaptive control of discrete systems. In this paper the proce dure of Etxebarria is modified. The learning rate of the neural network is improved and accelerated using the PD, PI and PID input controllers for inp ut neurons. The effect of adding a momentum term (the past record of the le arning) to the learning rule of the neural network is studied. The results are compared and discussed using the examples of Etxebarria and two other c ase studies. The procedure is extended to multi-input multi-output systems and cases studied are simulated. (C) 2000 Elsevier Science Ltd. All rights reserved.