差分进化算法（DE）是一种用于优化问题的启发式算法。本质上说，它是一种基于实数编码的具有保优思想的贪婪遗传算法[1] 。同遗传算法一样，差分进化算法包含变异和交叉操作，但同时相较于遗传算法的选择操作，差分进化算法采用一对一的淘汰机制来更新种群。由于差分进化算法在连续域优化问题的优势已获得广泛应用，并引发进化算法研究领域的热潮。 差分进化算法由Storn 以及Price [2]提出，算法的原理采用对个体进行方向扰动，以达到对个体的函数值进行下降的目的，同其他进化算法一样，差分进化算法不利用函数的梯度信息，因此对函数的可导性甚至连续性没有要求，适用性很强。-Differential evolution (DE) is a heuristic algorithm for optimization problems. Essentially, it is based on real-coded greedy thoughts with ensuring quality genetic algorithm [1]. With the genetic algorithm, differential evolution algorithm includes mutation and crossover operation, but compared to the genetic algorithm selection operation on differential evolution algorithm uses one-elimination mechanism to update the population. Due to the advantages of differential evolution algorithm optimization problem in continuous domain has been widely used, and triggered a boom in the field of evolutionary algorithms. Differential evolution algorithm consists of Storn and Price [2] proposed algorithm uses the principle direction of the disturbance of individuals in order to achieve the objective function value of the individual conduct of the decline, like other evolutionary algorithms, differential evolution algorithm does not use function gradient information, so the function can not requir