摘要： 提出一种基于自适应选择维度的记忆进化算法。该算法设置一个三维数组保存有用的进化信息，用于引导后续的进化过程，增强局部搜索能力，在变异过程中结合记忆信息自适应地选择维度进行变异，加强变异的有效性，当代种群中的最优个体通过自学习提高算法求解精度。标准函数仿真结果表明，该算法适合求解高维优化问题，局部收敛速度快，全局收敛能力强，算法稳定性高。

关键词: 进化算法, 变异记忆, 自适应选择, 自学习, 维度选择

Abstract: This paper proposes a Memory Evolutionary Algorithm Based on Adaptive Selecting Dimension(MEABASD). It sets a three- dimensional array to save useful evolutionary information in order to guide the evolution of the follow-up, which can enhance the local search ability. In mutation process, combining with memory information, it adaptively selects the dimension to mutation to strengthen the effectiveness of mutation. The best contemporary populations does self-learning operator to improve the precision of the algorithm. Simulation results on standard test functions show that the algorithm is suitable for the high-dimension optimization problem, and it has the characteristics of rapid convergence, powerful global search capability and high stability.

Key words: evolutionary algorithm, mutation memory, adaptive selection, self-learning, dimension selection

中图分类号: 

• TP301.6