微分进化算法的研究综述

微分进化(DE)是比较新的基于群体的随机优化方法.它具有简单、快速、鲁棒性好等特点,已经得到广泛关注.不同于其它进化算法,它的变异算子是由种群中任意选取的多对向量的差值得到的.微分进化主要用于实参数优化问题,在非线性和不可微的连续空间问题上优于其它进化方法.近些年,微分进化的应用领域也是不断扩大.研究目的是总结微分进化的研究进展和应用领域,并对它的进一步研究进行展望.

Research on differential evolution algorithm:a survey

Differential evolution is a relatively new population based stochastic optimization approach.It has been attracting increasing attention for it is simple,fast and robust.Unlike other evolutionary algorithms,the differential evolution uses the difference of randomly sampled pairs of vectors in the population for its mutation operators.DE is applied mainly in real parameter optimization,and outperforms other evolutionary algorithms in nonlinear and non-differentiable continuous space problems.DE has been found an increasing application in recently yeas.An aim of this paper is to summarize DE's researches and applications,and to give some further research issues.