基于窦性变异的改进人工蜂群白骨顶鸡算法及应用-《计算机技术与发展》

文章信息/Info

Title:: Improved Artificial Bee Colony Coot Algorithm Based on Cosine Mutation and Its Application

Author(s):: ZHANG Yu; HE Qing; School of Big Data and Information Engineering,Guizhou University,Guiyang 550025,China

Keywords:: COOT; elite opposition-based learning; artificial bee colony algorithm; cosine mutation strategy; engineering design problem

摘要:: 针对白骨顶鸡算法(COOT)存在求解精度低、收敛速度较慢和易陷入局部最优的问题,该文提出一种基于窦性变异的改进人工蜂群白骨顶鸡算法( ICOOT) 。首先,采用精英反向学
习策略初始化个体位置,增加初始个体寻优多样性;其次,考虑到人工蜂群算法强大的搜索能力,提出一种以全局最优值引导的改进人工蜂群搜索策略,更新白骨顶鸡个体的位置,以提高 COOT 的搜索能力和收敛精度;最后,引入窦性变异策略对最优个体进行扰动,一方面使算法能够有效跳出局部最优,另一方面提高最优个体质量。利用 12 个基准测试函数对 ICOOT 进行寻优性能测试,将 ICOOT 应用于拉力 / 压力弹簧优化工程设计问题,并与其他元启发式算法进行了比较和分析,结果验证了改进的算法的可行性和优越性。

Abstract:: Aiming at the problems of low solution accuracy,slow convergence and local optimality in COOT algorithm,we propose an improved artificial bee colony white-bone top chicken algorithm ( ICOOT) based on cosine mutation. Firstly,the elite opposition-basedlearning strategy is used to initialize the individual position and increase the diversity of the initial individual search. Secondly,considering the powerful search ability of the artificial bee colony algorithm,an improved artificial bee colony search strategy guided bythe global optimal value is proposed to update the positions of the white-boned top hen individuals to improve the search capability andconvergence accuracy of the COOT. Finally,the?
sinus variation strategy is introduced to perturb the optimal individual,which on the onehand makes the algorithm jump out of the local optimal effectively,and on the other hand improves the quality of the optimal individual.Twelve benchmark test functions are used to test the optimization performance of the ICOOT. The ICOOT is applied to the problem oftension / pressure spring optimization engineering design, and is compared and analyzed with other meta - heuristic algorithms, whichverifies the feasibility and superiority of the improved algorithm.