The temperature system identification of the PVC stripper tower top based on PSO-FCM optimized T-S model

Abstract

In view of the characteristics of T-S model, such as easily expressing complex dynamic systems and the characteristics of PSO algorithm which could find the optimal solution of complex problems easily. This paper will presents a new identification method based on the T-S model in which FCM parameters is optimized by PSO. The mathematical model of the temperature system of the PVC stripper tower top will be built by this method. First, an adaptive number of clusters of C-means clustering fuzzy (FCM) algorithm is used to find the appropriate number of clusters in FCM, and both the number of fuzzy rules and the premise parameters of the model can are determined. Using PSO algorithm to optimize the FCM algorithm, then getting the best membership matrix by the FCM algorithm based on PSO in the end. Then, a least square algorithm is applied to determine the parameters of consequent part of T-S model. The simulation result shows the effectiveness and feasibility of the modeling method.