Fuzzy identification based on improved T-S fuzzy model and its application in power plants

Abstract

Coordinated control system in power plants is a complicated system with nonlinearity and randomcity. It is difficult to build the nonlinear models by the traditional method, so the whole optimal control for thermal processes is impossible. A kind of method of fuzzy identification based on improved T-S (Takagi-Sugeno) model is proposed in this paper. Firstly, the heuristic information and the multiplex nonlinear optimization are combined to configure the structure of fuzzy model. Secondly, the input data space is partitioned into some local regions based on entropy clustering and competitive learning algorithm. Finally, the T-S model for coordinated control system in power plants is built with weighted recursive least-square algorithm. The simulation results show that the proposed improved T-S model can describe the non-linearity of processes accurately, and the relevant algorithms are very simple and fast.