基于改进多种群PSO算法的火炮随动系统调节器参数优化  

Parameter Optimization of Gun Follower System Based on Improved Multi-group PSO Algorithm

作　　者：韩超 段纬然 贾长治 闫媛媛 HAN Chao;DUAN Wei-ran;JIA Chang-zhi;YAN Yuan-yuan(Army Engineering University Shijiazhuang Campus,Shijiazhuang 050003,China;Unit 32181 of PLA,Shijiazhuang 050003,China)

机构地区：[1]陆军工程大学石家庄校区,石家庄050003 [2]解放军32181部队,石家庄050003

出　　处：《火力与指挥控制》2020年第6期56-61,66,共7页Fire Control & Command Control

基　　金：国家自然科学基金资助项目(51175508)。

摘　　要：针对传统火炮随动系统调节器参数整定难以达到最优的问题,提出一种基于K-均值与惯性权重指数递减的多种群PSO(KEDM-PSO)优化算法。为保证种群的全局搜索能力得到最优的参数,采用将初始种群划分为多个子群协同寻优的策略。综合考虑系统复杂程度、种群规模、解集的多样性及收敛性,采用K-均值算法将初始种群划分为3个子群,使3个子群协同寻优。为保持种群多样性,各子群不断地聚类重组,动态调整子群规模以更好地进化。子群寻优采用惯性权重指数递减策略,使得算法具有初期搜索范围大、速度快,后期惯性权重小,利于收敛、稳定的特点。试验表明该算法是有效可行的。Aiming at the problem that the tuning of traditional artillery follower system regulators is difficult to achieve optimality,a multi-group PSO(KEDM-PSO)optimization algorithm based on K-means and inertia weight index is proposed.In order to ensure the global search ability of the population to obtain the optimal parameters,a strategy of dividing the initial population into multiple subgroups for collaborative optimization is adopted.Considering the complexity of the system,the size of the population and the diversity and convergence of the solution set,the K-means algorithm is used to divide the initial population into three subgroups,so that the three subgroups can be optimized together.In order to maintain the diversity of the population,each subgroup is continuously clustered and reorganized,and the subgroup size is dynamically adjusted to better evolve.Sub-group optimization uses the inertia weight index diminishing strategy,which makes the algorithm have the characteristics of large initial search range,fast speed,and small inertia weight at the later stage,which is conducive to convergence and stability.Experiments show that the algorithm is effective and feasible.

关 键 词：随动系统 参数优化 多种群PSO算法 惯性权重指数递减 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TJ302[自动化与计算机技术—计算机科学与技术]