MODEL PREDICTIVE CONTROL OF A CHEMICAL PROCESS BASED ON AN ADAPTIVE NEURAL NETWORK

摘要

An adaptive neural network-based predictive strategy is applied to a pilot multivariable chemical reactor. The first stage of the project, simulation study, has been investigated and is presented in this paper, together with the description of the adaptive network. A pseudo-linear radial basis function (PLRBF) network is developed to model the real process and its weights are on-line updated using a recursive orthogonal least squares (ROLS) algorithm. The effectiveness of the adaptive control in improving the closed-loop performance has been demonstrated for process time-varying dynamics and model-process mismatch.