收敛因子非线性变化的鲸鱼优化算法

摘要

The elementary whale optimization algorithm (WOA) has disadvantages in treatment of complex global searching problem such as low solution precision and slow convergence.Aimed at this,an improved whale optimization algorithm (IWOA) was proposed,in which the convergence factor would vary nonlinearly with evolutional iteration times.In this algorithm,a chaotic initialization method instead of random one was used to generate initial population,making it have a better diversity.Inspired by the idea of inertial weighing for particle swarm optimization (PSO),an updating formula of convergence factor was deduced,in which the convergence factor would nonlinearly vary with the evolutional iteration times to compromise the global and local searching ability of the algorithm.A chaotic disturbance strategy was executed onto the present optimum whale individual to expand the search range.Six high-dimensional standard testing functions were chosen to perform numerical test and its result showed that the proposed IWOA algorithm would have a higher solution precision and a quicker convergence speed.%针对基本鲸鱼优化算法在处理复杂全局优化问题时存在解精度低和收敛速度慢等缺点,提出一种收敛因子随进化迭代次数非线性变化的改进鲸鱼优化算法.该算法利用混沌方法替代随机方法初始化种群,使群体具有较好的多样性.受粒子群算法惯性权重启发,设计出一种随进化迭代次数增加而非线性变化的收敛因子更新公式,以平衡算法的全局搜索和局部搜索能力.对当前最优鲸鱼个体执行混沌扰动策略以扩大其搜索范围.选取6个高维标准测试函数进行数值实验,结果表明该算法具有较高的收敛精度和较快的收敛速度.