Strategies for modeling and control of nonlinear chemical processes using neural networks

Abstract

A nonlinear predictive control framework is presented, in which nonlinear processes are modeled using neural networks. Several important issues concerning the modeling of nonlinear processes using neural networks are treated, with the emphasis placed on the convergence of neural networks to desired steady states. For nonlinear process predictive control where the neural network model is employed, an important case is examined. A typical nonlinear process, pH control problem, is taken as a case study to demonstrate the proposed approach, some significant results are given.