Improved Particle Swarm Algorithm Based Multi-objective Optimization of Diaphragm Spring of the Clutch

Abstract

Considering that diaphragm spring is the core component of the mechanical clutch, the optimization to which plays practical roles in engineering practices, the multi-objective optimization model for the diaphragm spring of the clutch is established in this article. Aiming at the difficulty in local extremum due to pre-maturity of inertia weight and treatment on nonlinear constraint condition of standard particle swarm optimization (PSO), the improved particle swarm algorithm（Improved PSO） based on dynamic weight and hierarchical penalty function in consideration of the degree of congestion is proposed in this article to improve the original particle swarm algorithm. According to the results of calculating examples, the improved particle swarm algorithm can achieve better global searching ability and convergence ability; when compared with the calculating results of the penalty function algorithm, the genetic algorithm and the NSGA-II algorithm, the pressing force of the diaphragm spring with the new algorithm is increased by 3.24%, and the steering separation force is decreased by 20.09%. The diaphragm spring has better pressing force stability and operating lightness, verifying the correctness of the model and the algorithm proposed in this article.