自适应差异演化算法及其应用

Adaptive differential evolution and its application

差异演化(Differential Evolution,DE)算法是一种基于群体差异的演化算法,具有良好的优化性能,但是对于高维复杂函数,DE算法易早熟收敛。为此,在对DE算法参数分析的基础上,提出自适应缩放因子及自适应交叉率两个概念,进而提出一种自适应差异演化(Adaptive Differential Evolution,ADE)算法。利用群体差异度对DE算法进行分期,一方面使缩放因子在前期较大,在进化的中期先变小,后增大,在进化的后期,缩放因子较小;另一方面使DE算法的交叉率在前期较小,中期在一定范围内随机取值,进化后期较大。仿真实验结果与工程应用实例表明,ADE算法在收敛速度和全局搜索能力方面得到了较好的平衡,不仅保证了ADE算法的收敛速度,而且具有较好的全局搜索能力。

Differential Evolution（DE）is one kind of evolution algorithm,which based on difference of individuals.DE has exhibited good porformance on optimization.However the algorithm,to the hign dimension and perplexed funtion,will fall into premature convergence.An adptive scaling factor and catastrophe factor are presented in this document,which based on the analysis of the parameters of differential evolution,and then an Adaptive Differential Evolution（ADE）algorithm is proposed.An adaptive scaling factor and crossover ratio are presented in this document.Divided the algorithm into three partitions according to the difference of population.Firstly,it has a comparatively large scaling factor in first partition and decreasing then increasing scaling factor in second partition and a comparatively small scaling factor in last partition.Secondly,it has a comparatively small crossover ratio in first partition and a random value of certain limit in next partition and a comparatively large value in last partition.Results of simulations and engineering optimization design example show that ADE compromise the algorithm convergence speed and global optimal capabilities,not only guarantees the convergence rate,but also has better global optimal ability.