基于种群引导和控制参数非线性递减的多目标鲸鱼优化算法
Multi-objective whale optimization algorithm based on nonlinear decline of population guidance and control parameters
薛开开 1刘淳安1
作者信息
- 1. 宝鸡文理学院数学与信息科学学院,陕西 宝鸡 721013
- 折叠
摘要
为了多目标鲸鱼优化算法能快速且有效地得到多目标优化问题的 Pareto前沿面,提出一种基于种群引导和控制参数非线性递减的多目标鲸鱼优化算法(GCNSWOA).该算法利用种群引导比例非线性递减策略,在进化前期可以增强算法全局搜索能力,后期可以改善问题解的分布情况并提高算法跳出局部最优解的概率;控制参数由线性递减变为非线性递减,在进化前期控制参数保持较大值可以提高算法全局搜索能力,避免早熟,在进化后期控制参数快速降低其值可以提高算法局部开发能力并且可以加速算法收敛.实验仿真将 GCNSWOA和 3 种比较算法对 5 个标准多目标测试问题的寻优结果进行对比.结果表明,GCNSWOA寻得的多目标优化问题的 Pareto最优解具有更好的宽广性和均匀性.
Abstract
A multi-objective whale optimization algorithm(GCNSWOA)based on nonlinear decline of population guidance and control parameters was proposed in order to quickly and effectively obtain Pareto front of multi-objective optimization problems.In the early stage of evolution,the algorithm can enhance the global search ability,and in the later stage,it can improve the distribution of the solution and increase the probability of leaping out of the local optimal solution.Control parameters change from linear decline to nonlinear decline.Maintaining a large value of control parameters in the early stage of evolution can improve the global search ability of the algorithm and avoid prematurity.Rapidly reducing the value of control parameters in the late stage of evolution can improve the local development ability of the algorithm and accelerate the convergence of the algorithm.The optimization results of 5 standard multi-objective test problems are compared by GCNSWOA and 3 comparison algorithms.The results show that the Pareto optimal solution of the multi-objective optimization problem obtained by GCNSWOA has better universality and uniformity.
关键词
种群引导/控制参数/非线性递减/非支配排序/多目标鲸鱼优化算法Key words
population guidance/control parameters/nonlinear decline/non-dominated sorting/multi-objective whale optimization algorithm引用本文复制引用
出版年
2025
NETL
NSTL
万方数据