基于多策略混合鲸鱼-蚁群优化算法的装配序列优化  

ASSEMBLY SEQUENCE OPTIMIZATION BASED ON EOBSWOA-ACO ALGORITHM

作　　者：黎响 王永[1] 田德[1] Li Xiang;Wang Yong;Tian De(School of New Energy,North China Electric Power University,Beijing 102206,China)

机构地区：[1]华北电力大学新能源学院,北京102206

出　　处：《太阳能学报》2025年第2期565-575,共11页Acta Energiae Solaris Sinica

基　　金：国家重点研发计划(2022YFE0207000)。

摘　　要：装配序列规划(ASP)是风电机组设计和制造的关键内容,对产品的生产效率和成本有重要影响。SP问题是一个典型的NP完全问题,需使用有效的方法来搜索最优或近优的装配序列,但常用智能优化算法的参数值获取比较困难,导致在搜索效率和收敛精度上存在一定局限性。为此,提出一种求解SP问题的多策略混合鲸鱼-蚁群优化算法。在计算过程中,使用增加精英反向学习策略(OBL)、差分进化算法(DE)的多策略混合鲸鱼算法优化蚁群算法的参数,然后再采用蚁群算法搜索最优或近优的装配序列。计算实验表明:多策略混合鲸鱼-蚁群优化算法降低了参数设置的复杂性,在求解SP问题上,与传统蚁群算法相比,算法的收敛速度和寻优能力得到很大提高。Ssembly sequence planning(SP)is the key content of wind turbine design and manufacturing,which has an important impact on the production efficiency and cost.SP problem is a typical NP-complete problem,which needs to use effective methods to find the optimal or near-optimal assembly sequences.However,it is difficult to obtain the parameter values of the general intelligent optimization algorithms,which leads to the limitations of these algorithms on search efficiency and convergence accuracy.To tackle the problem,a multi-strategy hybrid whale-ant colony optimization algorithm for SP problem is proposed.In the calculation process,the parameters of the ant colony algorithm are optimized by the multi-strategy hybrid whale algorithm,which adds the elite reverse learning strategy and differential evolution(DE)algorithm,and then the ant colony algorithm is used to search the optimal or near-optimal assembly sequence.Computational experiments show that the multi-strategy hybrid whale-ant colony optimization algorithm reduces the complexity of parameter setting.Compared with the traditional ACO,the convergence speed and optimization ability of the algorithm are greatly improved in solving SP problems.

关 键 词：装配序列规划 风电机组 参数 多策略混合鲸鱼-蚁群算法 

分 类 号：TK83[动力工程及工程热物理—流体机械及工程]