摘要
针对标准鲸鱼算法(WOA)及部分衍生算法求解某些算例效果不佳的问题进行了研究与实验,证明了WOA“包围”过程存在零点搜索偏好陷阱;而混沌优化算法(COA)不均衡的搜索特性使得部分衍生WOA融合的混沌初始种群与群智能优化过程难以调和。为了改善上述缺陷,选用了两种混沌系统和气泡网捕猎策略,设计了一套融合式优化算法。算法采用基于适应度的基线式自适应振荡群粒划分策略指导群体行为模式,充分发挥混沌系统作用,平衡探索与收敛性能。对通用/改进算例和工程应用案例求解可知,该算法性能相较于对比组算法更优,且不存在搜索偏好。
The standard whale optimization algorithm(WOA)and some derivative algorithms are studied and experimented to solve the problem of poor results of some examples.It is proved that there is a zero searching preference trap in the“encir-cling”process of WOA.In addition,the unbalanced search characteristics of chaos optimization algorithm(COA)make it difficult to reconcile the chaos initial population and swarm intelligence optimization process.In order to improve the above defects,this paper selected two chaotic systems and bubble net hunting strategy,and designed a set of fusion optimization algorithm.The algorithm adopted the baseline adaptive oscillation group partition strategy based on fitness to guide the group behavior pattern,gave full play to the role of chaotic system,and balanced the exploration and convergence performance.It used the new algorithm to solve the general/improved examples and engineering application cases.It’s obviously that the performance of the algorithm is better than the contrast group algorithm,and there is no searching preference.
作者
林之博
刘媛华
Lin Zhibo;Liu Yuanhua(Business School,University of Shanghai for Science&Technology,Shanghai 200093,China)
出处
《计算机应用研究》
CSCD
北大核心
2021年第10期3060-3066,3071,共8页
Application Research of Computers
基金
国家自然科学基金资助项目(11505114)
上海航天科技创新基金资助项目(SAST2018-22)。