求解函数优化问题的混合进化规划算法

Hybrid Evolutionary Programming Algorithm for Function Optimization Problems

提出一种混合进化规划算法,将进化规划与免疫进化中的克隆扩增相结合.该算法一方面用自适应变异步长的进化规划来有效地控制种群的整体进化,以在全局范围内进行搜索;另一方面,对于当前代中最优个体本身,利用免疫进化中的克隆扩增算子,来进行小邻域的局部细搜,从而形成两层领域搜索机制,以保证全局和局部搜索能力.仿真结果表明,该算法收敛速度快,搜索精确度高,并具有良好的全局搜索能力.

To propose a Hybrid Evolutionary Programming Algorithm （HEP）. HEP combines evolutionary programming method with immune evolutionary algorithm. In one hand, this algorithm employs self - adaptive evolutionary programming to control the global search, in other hand, it employs the clone - increase of immune evolutionary algorithm to perform the local search, so that two- level neighborhood search mechanism is realized to ensure the global and local search capabilities of the algorithm. The simulation results shows that the algorithm converges quickly, and has satisfactory capabilities of global and local search.