融合多策略人工水母算法及工程应用研究

Research on artificial jellyfish search integrating multi⁃strategy and its engineering application

针对人工水母群算法(JS)在求解较为复杂问题和部分工程问题时存在易陷入局部最优和寻优精度低等问题,提出一种融合多策略的人工水母算法。首先,该算法利用莱维飞行的扰动策略对洋流运动和被动运动阶段位置更新方式进行扰动,增强算法跳出局部最优的能力;其次,对水母跟随洋流和群内被动运动阶段引入指数递减惯性权重,使算法扩大全局搜索范围,提高寻优收敛速度;最后,将黄金正弦引导机制引入至被动运动阶段,替换原有被动运动位置更新方式,提升算法在探索和开发阶段的寻优能力。通过对6个基准测试函数进行测试分析,采用Wilcoxon秩和检验证明改进算法的寻优能力强、收敛速度快,将其应用于压力弹簧设计工程实际问题中,进一步验证了改进算法的适应性和有效性。

Since the artificial jellyfish search is easy to fall into local optima and has low searching accuracy when it is used to address more complex problems and some engineering problems,a new artificial jellyfish search integrating multi⁃strategy is proposed in this paper.The perturbance strategy of Lévy Flight is adopted in the algorithm to perturb the location update mode in the sections of ocean current motion and passive motion,so as to enable the new algorithm to avoid local optima.The exponentially decreasing weight is introduced to the phases that the jellyfishes move with ocean current and passive motion in jellyfish group,which can make the algorithm expand the scope of global search and accelerate the convergence.The golden sine guidance mechanism is introduced into passive motion stage change to replace the original location update mode of passive motion and improve the ability of the improved algorithm in the sections of search and development.The results of test analysis and Wilcoxon rank sum test 6 benchmark test function show that the improved algorithm has strong optimizing ability and fast convergence speed.It has been applied to the design of the spring pressure,by which its applicability and effectiveness have been verified.