Self tuning control of nonlinear systems: a meural net based approach

Abstract

In recent years there has been considerable interest in the area of artificial neural networks. One of the application areas of neural networks is in adaptive control systems. In this work, use of neural networks for direct self tuning control of deterministic and stochastic nonlinear systems has been proposed and investigated using simulation study. Minimum variance and generalized minimum variance control schemes have been used. Methods to overcome the slow convergence problem associated with the backpropagation training algorithm have also been proposed and implemented.