Multi-objective optimization of gear reducers based on NSGAII

CHENG Xiao-sheng, YU Jun-he, ZHAN Hong-fei

(The Faculty of Mechanical Engineering and Mechanics, Ningbo University, Ningbo 315211, China)

Abstract: Aiming at solving the multi-objective optimization problem with constraints and discrete variables in engineering design, non-dominated sorting in genetic algorithms II (NSGAII) was investigated. The methods were presented to handle the discrete variables and constraints by non-dominated sorting based on crowding distance, and Matlab was used as a tool to work out the multi-objective optimization program of NSGAII. The mathematical model of the multi-objective optimization design of two-stage gear reducer was established, the Pareto set was achieved by using the modified NSGAII and the optimal solution was selected according to the method of fuzzy set theory. The volume, the failure probability and the transmission error were decreased compared to traditional method. The results indicate that the modified NSGAII can effectively solve the multi-objective optimization problem with constraints and discrete variables.