多策略改进哈里斯鹰优化算法  

Improved Harris Hawks optimization algorithm with multiple strategies

作　　者：郭雨鑫 刘升[1] 高文欣 张磊[1] GUO Yuxin;LIU Sheng;GAO Wenxin;ZHANG Lei(School of Management,Shanghai University of Engineering and Technology,Shanghai 201620,China)

机构地区：[1]上海工程技术大学管理学院,上海201620

出　　处：《微电子学与计算机》2021年第7期18-24,共7页Microelectronics & Computer

基　　金：国家自然科学基金(61673258);上海市自然科学基金(19RZ1421600)。

摘　　要：为解决基本哈里斯鹰算法(Harris hawks optimization,HHO)易陷入局部最优和收敛精度低的问题,提出多策略优化的哈里斯鹰优化算法(Multi-Strategy Harris hawks optimization,MHHO).在探索阶段,引入柯西分布函数变异全局位置,增加种群多样性;在过渡阶段,利用随机收缩指数函数非线性化能量方程,更好地协调全局探索和局部开采;在开采阶段,引入自适应权重因子更新局部位置,提高局部开采能力.通过求解多个单峰、多峰和高维度测试函数,结果表明融合三种策略的MHHO算法具有更好的寻优精度和稳定性.In order to solve the problems of the basic Harris hawks optimization(HHO)that are easy to fall into local optimization and low convergence accuracy,a multi-strategy optimization Harris hawks optimization(MHHO)is proposed.In the exploration stage,the Cauchy distribution function is introduced to mutate the global position which increases population diversity.In the transition stage,the random contraction exponential function is used to nonlinear the energy equation in order to coordinate global exploration and local mining effectively.In the mining stage,the adaptiveweight factor updates the local position and elevates the local mining capacity.By solving multiple single peak,multi peak and high-dimensional test functions,the results show that the MHHO algorithm combining the three strategies has better optimization accuracy and stability.

关 键 词：哈里斯鹰优化算法 柯西变异 随机收缩指数函数 自适应权重 高维优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]