Multi-objective Optimization of Planetary Reducer Based on an Improved Genetic Algorithm

Abstract

This paper presents a multi-objective optimization of planetary reducer based on an improved multi-objective genetic algorithm (IMOGA). Minimization of volume, maximization of transmission ratio and efficiency are set as three objectives. However, owing to the difference of difficulty in solving objective functions, the optimization model of planetary reducer has the problem of uneven distribution of competitive pressure, the conventional evolutionary algorithm has poor convergence at the partial Pareto front. Thus, an improved multi-objective genetic algorithm using infeasible solution guidance and hybrid crossover operator of cytoplasm and chromosome is proposed. Experimental results of six test functions verify the effectiveness of the proposed algorithm and show that IMOGA has faster convergence speed and better convergence in comparison with NSGA-II. Ultimately, a planetary reducer optimization problem is solved by IMOGA and NSGA-II. Comparison results illustrate the competitiveness of IMOGA and prove that IMOGA can provide better solutions for designer. The Pareto set of the planetary reducer is distributed in stepped. The solutions on the same step have similar efficiency and different steps have different distribution ranges in transmission ratio and volume.