Optimization Algorithm Based on T-S Fuzzy Model of Self-Adaptive Disturbed Particle Swarm Optimization and Neural Network

Abstract

To solve fuzzy and non-linear features of mechanical equipment. A new computational intelligence method was proposed by combing based on extended T-S fuzzy model of self-adaptive disturbed PSO and BP neural network algorithm. Firstly, the T-S fuzzy model is modified, and then uses the extended T-S model to adjust the PSO parameter. Secondly, the neural network is trained by the modified PSO algorithm. Finally, a wheel disc model is optimized to check that network model, the test results show that it guarantees the performance of the wheel disc, meanwhile the wheel disc structure is obviously optimized, and the algorithm in the paper is a method of viable structure optimization.