Modeling of Stripper Temperature based on improved T-S Fuzzy Neural Network

In the Polyvinyl Chloride (PVC) industry, proper control of stripper temperature is directly related to product quality of PVC resin. Considering multivariable, strong coupling, nonlinear and time-varying characteristics of the temperature control system for PVC stripper the current modeling method is difficult to obtain a relatively accurate mathematical model. Then, this paper studies the stripper temperature modeling method based on improved T-S fuzzy neural network, and proposes new nearest neighbor clustering fuzzy rules. In order to improve the learning performance, hybrid learning algorithm based on T-S fuzzy neural networks is developed. As for non-linear layer parameters, conjugate gradient algorithm is applied, while recursive least squares algorithm is adopted to handle the linear parameters. Simulation results demonstrate the effectiveness and accuracy of the proposed modeling method given.