中文摘要：

提出一种基于强度Pareto进化算法2（SPEA2）的复杂机械产品多目标多质量要求下的选择装配方法．以装配精度和装配率作为质量要求的评价方式，建立多质量要求下的选择装配模型．采用特征尺寸为基因的编码方式构建染色体，尺寸链复杂度描述质量要求的主次关系，零件复杂度确定进化操作中的特征尺寸的关联约束基准，综合Pareto支配强度及拥挤度生成适应度函数作为复杂机械产品选择装配多目标多质量要求下的个体评价规则．以某汽车发动机汽缸装配为例，优化结果以Pareto边界集表示，验证了算法的可行性与有效性。

英文摘要：

Studies of selective assembly problem were criticized for the reason that most models were confined to single dimension chain and simple structure of assemblies. To deal with this problem, a selective assembly method of complicated mechanical product under the requirement of multi-objective based on the strength Pareto evolution algorithm 2 （SPEA2） was proposed. A selective assembly model under the re- quirement of multi-objective was established with assembly precision and assembly yield which were taken as the criteria of evaluation. A coding method taking characteristic dimensions as genes to build chromo- some was used. The primary and secondary relations of the dimension chains were described using the complexity of dimension chains. The complexity parts was used to describe the interconnection constraint reference of the characteristic dimension in the operation of evolution respectively. The strength value and density information were taken into account to build the fitness function, which was taken as the evalua- tion rule of the individuals. The assembly of an automobile engine cylinder was taken as an example to prove the efficiency and feasibility of the algorithm, and the optimization results were stored in the Paretooptimal front set.