Abstract: In order to improve the precision of assembly，Computer Aided Selective Assembly can achieve this target.This article proposes a new selective assembly model used in multi-dimension chains and contrasts different multi-object optimization algorithms when they are applied on Computer Aided Selective Assembly.Based on PSO，an improved PSO algorithm is designed to solve the problem of multi-objective optimization.In the improved algorithm，authors use an external collection to update the solutions，assure the algorithm can better converge near the true Pareto-optimal front and ensure the distributing of solutions is better equality.Finally，a fact proves that this model and algorithm can efficiently solve the selective assembly problem.

Key words: selective assembly, multi-objective optimization, Particle Swarm Optimization（PSO）, quality loss cost

摘要： 为了提高装配的精度，可以使用计算机辅助选择装配来选择合适的零件进行装配。提出了一种面向多尺寸链计算机辅助选择装配模型；对比了几种多目标优化算法应用在计算机辅助选择装配中的优缺点；最终选择一种以粒子群优化算法为基础的多目标优化算法，在算法中通过使用外部集的不断更新来保证算法收敛到全局最优解。实例证明，随着迭代次数的增加，外部集中的解逐渐收敛于pareto前沿，而且解的分布比较均匀。

关键词: 选择装配, 多目标优化, 粒子群优化算法, 质量损失