基于多种策略改进的鲸鱼优化算法

An improved whale optimization algorithm based on multiple strategies

针对标准鲸鱼优化算法收敛速度慢、搜索与开发不平衡、种群之间信息交流匮乏、容易陷入局部最优的问题,提出了一种改进算法。首先,采用Tent混沌映射提高初始种群的分布均匀性,并引入非线性收敛因子,提升算法前期的全局搜索和中后期局部开发的能力,协调了搜索与开发的转换机制。然后,将种群的平均位置向量引入随机搜索过程中,有效改善个体与种群之间缺乏信息交流的问题。接着,将自适应惯性权重引入位置更新公式中,以加快算法的收敛速度,提高求解精度。最后,利用柯西算子对陷入局部最优的个体进行变异扰动。通过15个基准测试函数对改进算法进行仿真实验,实验结果表明,改进后的鲸鱼优化算法具有良好的性能,并通过Wilcoxon秩和检验证明了改进算法的有效性。

To address the issues of the standard whale optimization algorithm,including slow convergence speed,imbalance between exploration and exploitation,lack of information exchange among the population,and susceptibility to local optima,an improved algorithm is proposed.Firstly,the Tent chaotic mapping is employed to enhance the uniformity of the initial population distribution.Secondly,a nonlinear convergence factor is introduced to improve the algorithm s global search ability in the early stage and local exploration ability in the middle and late stages,coordinating the transition mechanism between search and exploitation.Then,the average position vector of the population is introduced into the random search process,effectively addressing the lack of information exchange between individuals and the population.Next,an adaptive inertia weight is introduced into the position update formula to enhance the convergence speed and accuracy of the algorithm.Finally,the Cauchy operator is utilized to perform mutation perturbation on individuals trapped in local optima.Simulation experiments were conducted on 15 benchmark test functions to evaluate the improved algorithm.The experimental results demonstrate that the improved whale optimization algorithm possesses excellent performance,and the effectiveness of the improved algorithm is proven through the Wilcoxon rank-sum test.