基于共轭增强策略的差分进化算法

为了平衡差分进化算法的全局探测能力和局部搜索能力,提出基于共轭增强策略的差分进化算法.首先,根据个体适应度信息设计基于轮盘赌的个体选择策略,选取适应值较差的个体组建子种群;然后,基于个体的时间和空间知识设计共轭增强方向,在不丧失全局探测能力的前提下实现子种群的局部增强,以提高算法的局部搜索能力;最后,18个标准测试函数的实验结果表明,所提算法在计算代价、可靠性及收敛速度方面均优于所介绍的主流改进差分进化算法和非差分进化算法.

Differential evolution algorithm with conjugate enhancement strategy

To balance the global exploration and exploitation ability of differential evolution, a differential evolution algorithm based on the conjugate enhancement strategy is proposed. Firstly, a roulette-based individual selection strategy is designed according to the fitness of individual to select individuals with the poorer fitness value to form the subpopulation. Then, the temporal and spatial knowledge of the individual is used to design the conjugate enhancement direction, and the local enhancement of subpopulation is realized without losing the global exploration ability to improve the exploitation ability. Finally, experimental results of 18 benchmark functions show that the proposed algorithm is superior to the main-stream differential evolution variants and non-differential evolution algorithms mentioned in this paper in terms of computational cost, reliability and convergence speed.