Optimising predictive control based on neural models

Abstract

This paper presents a Model Predictive Control (MPC) algorithm for on-line economic optimisation of nonlinear technological processes. The economic profit is explicitly expressed in the minimised objective function. The algorithm uses only one dynamic neural model, which is linearised on-line. As a result, an easy to solve on-line quadratic programming problem is formulated. In contrast to the classical multilayer control system structure, the necessity of repeating two nonlinear optimisation problems at each sampling instant is avoided. © Springer-Verlag Berlin Heidelberg 2008.