【摘要】 多级转子装配过程中存在着最终产品质量一致性差的问题，为更好指导实际生产，多级转子在选配时不仅要考虑零件的配对问题(离散变量),还需要考虑零件之间的装配相位(连续变量)。选用粒子群(particle swarm optimization, PSO)算法对其进行优化，提出了一种混合变量编码方式，结合自适应权重、压缩因子等策略对其进行改进，据此提出基于混合变量的多目标优化粒子群算法(mixed-variables and multi-objective PSO, MMPSO)。最后通过实例分析验证，对某一小批量多级转子进行优化选配，结果表明MMPSO提高了在3个优化目标的精度与搜索速度。更多还原

【Abstract】 The multi-stage rotor assembly process suffers from poor consistency in the quality of the final product. In order to better guide the actual production, multi-stage rotor in the selection of not only to consider the pairing of parts(discrete variables), but also need to consider the assembly phase between the parts(continuous variables). Based on this, particle swarm optimization(PSO) is chosen to optimize it. A mixed-variables coding method is proposed, which is improved by combining adaptive weights, compression factors, and other strategies, according to which the mixed-variables and multi-objective particle swarm optimization(MMPSO) is proposed. Finally, an example computational analysis is carried out to optimize the selection of a small batch of multistage rotors, and the results show that MMPSO improves the accuracy and search speed in the three optimization objectives.更多还原