一种多群进化规划算法

Multi-subgroup Evolutionary Programming

在分析了导致进化规划算法早熟原因的基础上 ,提出了一种改进的多群进化规划算法。在该算法中 ,进化在多个不同的子群间并行进行 ,通过使用不同的变异策略 ,实现种群在解空间具有尽可能分散探索能力的同时 ,在局部具有尽可能细致的搜索能力。通过子群重组实现子群间的信息交换 ,基于典型算例的数字仿真证明 ,该算法具有更好的全局收敛性 ,更快的收敛速度和更强的鲁棒性。

An improved multi-subgroup evolutionary programming(MEP) algorithm is proposed based on the analysis of traditional evolutionary programming premature convergence. In MEP algorithm, evolutions of many subgroups are parallelly performed with different mutation strategies, and then the population can separately explore the solution space and detailedly search the local part all together. Information is exchanged when subgroups are reorganized. Simulations based on benchmarks confirm that MEP algorithm is better than classic evolutionary programming algorithm in the aspects of global optimization, convergence speed and the robustness.