基于多目标遗传算法去NSGA-Ⅱ的RV减速器选配优化及实现

Optimization and Realization for Selective Assembly of RV Reducer Based on Multi-objective Genetic Algorithm NSGA-Ⅱ

作为工业机器人核心传动部件的RV减速器装配过程面临着装配结构复杂、装配尺寸链多且装配精度要求高等难题,传统的人工选配方法存在着装配效率低、精度差且合格率低等问题。在分析RV减速器结构特点和传动精度影响因素的基础上,提出了减少误差项的两级关键装配尺寸链。以选配合格率和装配精度作为评价标准,构建了RV减速器的多目标选配数学模型。针对选配优化样本空间大且计算复杂的特点,使用以零件位置为基因片段的矩阵实数编码方式,并提出交叉算子映射法修复无效染色体,采用运行速度快、解集收敛性好的NSGA-Ⅱ多目标遗传算法进行寻优。优化结果表明选配合格率从56.67%提升到96.67%,装配精度提升约22%。显然,文中提出的算法具有较好的可行性和实用性。

The assembly process of the RV reducer,being a core transmission component of industrial robots,is faced with complex assembly structure,multiple assembly dimension chains and high assembly precision requirement.Traditional manual matching method has low assembly efficiency,poor precision and low qualified rate.Based on the analysis of the structural characteristics of the RV reducer and the influencing factors of transmission accuracy,the two key assembly dimension chains to reduce the error terms are proposed.The matching qualification rate and matching accuracy are taken as evaluation criteria and a multi-objective matching mathematical model of the RV reducer is constructed.Aiming at the characteristics of large sample space and complex calculation,the matrix real number coding mode with part position as gene fragment is used.A cross operator mapping method is proposed to repair invalid chromosomes and NSGA-Ⅱmulti-objective genetic algorithm with fast running speed and good convergence of solution set is used for optimization.The optimization results show that the matching qualification rate is increased from 56.67%to 96.67%and the assembly precision is increased by about 22%.The algorithm proposed in this paper has good feasibility and applicability.