基于Logistic映射的蜉蝣优化算法
2021-10-24陈嘉豪童楠符强
陈嘉豪 童楠 符强



摘要: 在高维复杂问题上,蜉蝣优化算法存在易陷入局部最优区域且求解精度较差等问题,因而提出基于Logistic映射的蜉蝣优化算法。引入依据Logistic映射的混沌机制,当种群进化停滞时,当前最优蜉蝣通过混沌机制寻找适应度更好的蜉蝣,以激发种群进化能力;建立较劣蜉蝣加速进化机制,激励蜉蝣个体以达到种群寻优要求;采用动态惯性权重均衡算法全局和局部的搜索性能。抽取5个benchmark函数测试算法性能,实验结果验证了所提算法在寻优性能上的有效性。
关键词: 群智能算法; 蜉蝣优化算法; Logistic映射; 扰动; 动态惯性权重
中图分类号:TP301.6 文献标识码:A 文章编号:1006-8228(2021)10-06-05
Mayfly optimization algorithm based on Logistic mapping
Chen Jiahao, Tong Nan, Fu Qiang
(College of Science and Technology, Ningbo University, Ningbo, Zhejiang 315300, China)
Abstract: For high-dimensional complex problems, mayfly optimization algorithm has problems such as easy to fall into local optimal area and poor solution accuracy. Therefore, a mayfly optimization algorithm based on Logistic mapping is proposed. Introducing the Logistic mapping based chaotic mechanism, when the evolution of the population stagnates, the current optimal mayfly uses the chaotic mechanism to find a mayfly with better adaptability to stimulate the evolutionary ability of the population; Establishing a accelerated evolution mechanism for inferior mayfly, the mayfly individual is motivated to meet the requirements of the population optimization; Using the dynamic inertial weight balance algorithm, the global and local search capabilities are balanced. Five benchmark functions are selected to test the performance of the algorithm, and the experimental results verify the effectiveness of the proposed algorithm in optimizing performance.
Key words: swarm intelligence algorithm; mayfly optimization algorithm; Logistic mapping; disturbance; dynamic inertia weight
0 引言
群智能優化算法[1-2]易于编程,寻优能力强,现已广泛应用于实际工程领域中,例如解决旅行商问题和0-1背包问题等[3-4]。
蜉蝣优化算法[5](A mayfly optimization algorithmMA)是一种2020年7月由Konstantinos Zervoudakis和 Stelios Tsafarakis 提出的新型群智能算法,此算法受蜉蝣飞行行为和交配过程的启发,联结粒子群算法[6]和遗传算法[7]的优点,在低维问题上,相比较其余群智能算法具备更好的收敛速度和收敛精度,且算法中的随机飞行和婚礼舞蹈行为能有效平衡种群的全局探索及局部搜索要求。但当遇到高维复杂问题时,蜉蝣算法单纯依靠自身机制仍然较难跳出局部最优区域,且由于蜉蝣在移动时步长太短,导致算法收敛精度不高。
针对上述问题,本文提出基于Logistic映射的蜉蝣优化算法(mayfly optimization algorithm based on Logistic mapping LMA),采用动态惯性权重提高种群的搜索能力,增加算法收敛效率;引入基于Logistic映射的混沌机制,防止种群出现早熟收敛现象;建立较劣蜉蝣加速进化机制,减少较劣蜉蝣在外徘徊的次数,提升种群整体进化速度。……
