Abstract: After analyzing the low local search ability of electromagnetism-like mechanism（EM） algorithm, a hybrid EM algorithm based on simplex method is proposed. The proposed algorithm utilizes opposition learning strategy to construct the initial population that is scattered uniformly over the entire search space in order to maintain the diversity. Select the best population for local search by simplex method to speed up the convergence rate of the algorithm. The performance of the proposed algorithm tested using four well-known benchmark functions are reported, and the experimental results show that the proposed algorithm is more effective than standard EM algorithm and other evolutionary algorithms.

Key words: electromagnetism-like mechanism algorithm, opposition learning, simplex method, optimization

摘要： 针对类电磁机制算法存在局部搜索能力差的问题，提出一种基于单纯形法的混合类电磁机制算法。该混合算法首先利用反向学习策略构造初始种群以保证粒子均匀分布在搜索空间中。利用单纯形法对最优粒子进行局部搜索，增强了算法在最优点附近的局部搜索能力，以加快算法的收敛速度。四个基准测试函数的仿真实验结果表明，该算法具有更好的寻优性能。

关键词: 类电磁机制算法, 反向学习, 单纯形法, 优化