基于遗传算法的产品装配序列规划研究

为解决产品实际装配过程装配序列规划的问题,建立了描述装配过程中零部件间装配关系的几何干涉关系矩阵。设计了以装配换向次数、装配工具更换次数、装配体稳定性和基础件位置为目标函数的遗传算法,并改进了该遗传算法相应的交叉算子、变异算子和适应度函数,以确保算法的有效性。在初始种群生成时,输入由DELMIA软件虚拟拆卸得到的几何可行装配序列,改善种群的组成成份。将该遗传算法应用于某链轮的装配序列规划中,验证了该算法的有效性。

Product Assembly Sequence Planning Based on Genetic Algorithm

The improved genetic algorithm,based on the interference matrix which was establish-ed to describe the geometric interference relations in the components and the objective function considering of the assembly reorientation times,assembly tool change times,the assembly body stability and the assembly basis parts position in the sequence,was used to solve the assembly sequence planning problem.Appropriate crossover operator,mutation operator and fitness function were improved to plan the product assembly sequence which was closer to the actual assembly process.While the primary population is forming,the making up composition of feasible assembly sequence for population improvement that obtained from virtual disassembly sequence with DELMIA software could be inputted.A case study was given to illustrate the availability of the improved algorithm.