Intelligent predictive control with locally linear based model identification and evolutionary programming optimization with application to fossil power plants

Abstract

In this paper, an intelligent predictive control algorithm based on Locally Linear Model Tree (LOLIMOT) is implemented to control a fossil fuel power unit. The controller is a non-model based system that uses a LOLIMOT identifier to predict the response of the plant in a future time interval. An evolutionary programming (EP) approach optimizes the identifier-predicted outputs and determines input sequence in a time window. This intelligent system provides a predictive control of multi-input multi-output nonlinear systems with slow time variation. © Springer-Verlag Berlin Heidelberg 2005.