差分进化算法由于算法结构简单易于执行,并且具有优化效率高、参数设置简单、鲁棒性好等优点,因此差分进化算法吸引了越来越多研究者的关注。本文概述了差分进化算法的基本概念以及存在的问题,综述了差分进化算法的控制参数、差分策略、种群结构以及与其他最优化算法混合等4个方面改进策略并讨论它们各自的优缺点,为差分进化算法下一步的改进提出了参考方向。
Due to its simple algorithm structure, ease of performance, high optimization efficiency, simple parameter setting, and excellent robustness, the differential evolution (DE) algorithm has attracted increasing attention from researchers. In this paper, we outline the basic concepts of the DE algorithm as well as its limitations, and review four improvement strategies, including a control parameter, differential strategy, population structure, and mixing it with other optimization algorithms. We discuss the advantages and disadvantages of these strategies and suggest directions for future improvements to the DE algorithm.