摘要
针对量子粒子群算法(QPSO)在迭代后期出现种群多样性缺失和容易陷入局部最优的问题,提出了一种基于交叉操作的改进算法;在改进算法中,考虑了粒子的历史最优位置和次优位置,用以扩大粒子的搜索范围;同时,将遗传算法的交叉操作运用到位置的更新中,以增加种群的多样性,进而提高算法的收敛性;在性能测试中,将改进算法与原始的量子粒子群算法、基于差分进化的QPSO和基于黑洞探索的QPSO在收敛精度和鲁棒性方面进行了比较;最后,运用改进算法对一类具有投资数量限制的投资组合问题进行了求解,并与遗传算法、粒子群算法和标准的量子粒子群算法的寻优结果进行了对比。
According to the shortcomings of quantum behaved particle swarm optimization algorithm(QPSO),for instance,the lack of population diversity and getting trapped in local optima easily during the later stage of iteration,an improved algorithm based on cross operation is proposed.In the improved algorithm,particle’s history best position and suboptimal position are considered to expand its search space.Moreover,cross operation in genetic algorithm is used to renew particle’s position for enhancing population diversity and algorithm’s convergence.Through performance test,the improved algorithm is compared with the original quantum behaved particle swarm optimization algorithm,QPSO with differential evolution and QPSO based on black hole exploration in convergence accuracy and robustness.Finally,the improved algorithm is used to solve a kind of portfolio problems with quantity constraints,and the related optimization results are compared with genetic algorithm,particle swarm optimization algorithm and the standard quantum behaved particle swarm optimization algorithm.
作者
何光
李高西
HE Guang;LI Gao-xi(School of Mathematics and Statistics,Chongqing Technology and Business University,Chongqing,400067,China)
出处
《重庆工商大学学报（自然科学版）》
2020年第6期83-87,共5页
Journal of Chongqing Technology and Business University:Natural Science Edition
基金
国家自然科学基金(11901068)
重庆市基础与前沿研究计划项目(CSTC2016JCYJA0564)
重庆工商大学博士科研启动项目(2015-56-08)
重庆工商大学青年项目(1552004).
关键词
量子粒子群优化算法
交叉操作
多样性
投资组合
quantum-behaved particle swarm optimization algorithm
cross operation
diversity
portfolio
