A Computationally Efficient Nonlinear Predictive Control Algorithm with RBF Neural Models and Its Application

摘要

This paper details a computationally efficient (suboptimal) nonlinear Model Predictive Control (MPC) algorithm with Radial Basis Function (RBF) type neural network models and discusses its application to a polymerisation reactor. Neural model of the process is used on-line to determine the local linearisation and the nonlinear free trajectory. Unlike the nonlinear MPC technique, which hinges on non-convex optimisation, the presented algorithm is more reliable and less computationally demanding because it results in a quadratic programming problem, whereas its closed-loop control performance is similar.