协同进化扩展紧致量子进化算法  

Coevolutionary quantum evolutionary algorithm based on extended compact

作　　者：刘振[1,2] 胡云安[1,2] 彭军[1,2] 

机构地区：[1]海军航空工程学院控制工程系,山东烟台264001 [2]海军航空工程学院接改装训练大队,山东烟台264001

出　　处：《控制与决策》2014年第2期320-326,共7页Control and Decision

基　　金：国家自然科学基金项目(61174031;60674090)

摘　　要：为了提高量子进化算法的全局收敛性能,基于协同进化的思想,并结合扩展紧致遗传算法,提出了协同进化扩展紧致量子进化算法(CECQEA).该算法利用多粒度机制进行量子染色体的旋转,并依据边缘积模块(MPM)进行交叉和变异以避免优良模式的破坏;在每一个子种群内对个体依据MPM进行自调整操作,同时进行种群的分裂、合并及优良个体的迁移操作.通过对算法收敛性的分析可看出,CECQEA能够收敛到满意解集;经基准函数以及背包问题的仿真测试分析可看出,算法收敛效果更加明显.In order to promote the convergence performance of the quantum evolutionary algorithm, a coevolutionary extended compact quantum evolutionary algorithm is proposed. Quantum chromosome can be rotated according to the multi-granularity mechanism. Chromosome crossover and mutation are based on the MPM in order to avoid loss of better pattern, and the self-adjust operation is performed in the subpopulation. Spilt and merge operation are also executed in the subpopulation and some superior individuals migrate between subpopulation, and the trait of convergence is analyzed in the paper. The proposed algorithm can reach the satisfactory solution set. Simulation results of benchmark functions and the knapsack problems show that the proposed algorithm performs well than other algorithms.

关 键 词：量子进化算法 分布估计 扩展紧致 协同进化 

分 类 号：TP15[自动化与计算机技术—控制理论与控制工程]