融合多策略的改进秃鹰搜索算法
Improved bald eagle search algorithm fused with multiple strategies
郭云川 1张长胜 1段青娜 2罗运河 2程倩 2钱斌 1胡蓉1
作者信息
- 1. 昆明理工大学信息工程与自动化学院,昆明 650500
- 2. 红云红河烟草(集团)有限责任公司,昆明 650202
- 折叠
摘要
针对秃鹰搜索算法(BES)存在全局搜索性能与局部开发能力不协调、易陷入局部最优等缺陷,提出一种融合多策略的改进秃鹰搜索算法(IBES).采用凸型自适应控制因子使算法在迭代寻优过程中可根据搜索进程动态调整位置更新方程以修正模型,实现自适应寻优,有效平衡算法的全局搜索性能和局部开发能力;引入折射反向学习机制可对问题当前解在其解空间内进行折射反向学习找到与之对应的折射反向解,增加寻到最优解的概率,提升算法求解精度和收敛速度;同时,利用定向重组与诱导突变策略实现种群个体多维信息的重组和突变,提升个体质量和种群多样性,增加算法跳出局部最优的机率,提高搜索精度.以最优值、平均值、标准差和平均收敛代数作为算法性能的评价指标,对10个不同基准测试函数进行数值仿真实验,实验结果验证了所提改进方法的有效性及IBES算法的优越性.此外,经IBES算法优化后的PID神经网络控制器响应速度快、超调量小、调节时间短,进一步验证了算法的实用性.
Abstract
An improved bald eagle search algorithm fused with multiple strategies(IBES)is proposed to address the shortcomings of the bald eagle search algorithm(BES),such as the global search performance is not coordinated with the local exploitation capability and it is easy to fall into local optimum.The use of convex adaptive control factors enables the algorithm to dynamically adjust the position update equations to modify the model according to the search process during the iterative optimization,thus achieving adaptive optimization and effectively balancing the global search performance and local exploitation capability of the algorithm.The refracted opposition-based learning mechanism is used to discover the corresponding solution by refracting the current solution of the problem in its solution space,which increases the probability of finding the optimal solution and improves the solution accuracy and convergence speed of the algorithm.At the same time,the directional recombination and induced mutation strategy is used to achieve the recombination and mutation of the multi-dimensional information of population individuals,improve the individual quality and population diversity,increase the probability of the algorithm escaping from local optimum,and raise the searching precision.The optimal value,mean value,standard deviation and average convergence algebra are used as evaluation indexes of the algorithm performance,and numerical simulation experiments are conducted for 10 different benchmark test functions.The experimental results verify the effectiveness of the proposed improved method and the superiority of the IBES.In addition,the PID neural network controller optimized by the IBES has a fast response,small overshoot,and short regulation time,which further verifies the practicality of the algorithm.
关键词
秃鹰搜索算法/凸型自适应控制因子/折射反向学习机制/定向重组与诱导突变策略/PID神经网络控制器Key words
bald eagle search algorithm/convex adaptive control factors/refracted opposition-based learning mechanism/directional recombination and induced mutation strategy/PID neural network controller引用本文复制引用
基金项目
国家自然科学基金项目(51665025)
国家自然科学基金项目(61963022)
出版年
2024
NETL
NSTL
维普
万方数据