一种随机调整控制参数的鲸鱼优化算法

Whale Optimization Algorithm Based on Stochastic Adjustment Control Parameter

针对标准鲸鱼优化算法在处理复杂优化问题时出现搜索精度低和易出现早熟收敛等缺点,提出一种随机调整控制参数的改进鲸鱼优化算法(EWOA)。受粒子群优化算法中惯性权重的启发,利用随机分布的方式调整控制参数,以平衡鲸鱼优化算法的全局搜索和局部搜索能力。对当前最优个体执行服从正态分布的变异扰动,以避免算法出现早熟收敛现象。此外,采取佳点集方法替代随机方法产生初始个体以提高算法的全局收敛速度。6个标准测试函数的仿真实验结果表明EWOA能有效处理高维复杂优化问题。

To overcome the problems about low precision and premature convergence,an effective whale optimization algorithm（EWOA） based on stochastic adjustment control parameter was proposed to deal with complex optimization problems.Inspired inertia weight of particle swarm optimization algorithm,the stochastic control parameter was introduced to balance the ability of global search and local search.In order to avoid the proposed EWOA falling into local convergence effectively,the normal mutation for updating strategy of the current optimal individual was developed.In addition,to enhance the global convergence,when producing the initial individuals,the good point set method was employed.The experimental results of 6 benchmark functions show good performance of the proposed EWOA in dealing with high-dimensional optimization problems when compared with other methods.