论文部分内容阅读
针对多模态优化问题,提出一种动态小生境半径两阶段多模态差分进化算法.基于构象空间退火思想,设计一种两阶段退火策略来动态调整小生境半径,并根据退火过程将整个优化过程分为两个阶段.在第1阶段,通过差分限制变异策略生成高质量的新个体来维持种群的多样性,促进多模收敛;在第2阶段,利用种子邻近变异策略对已探测到的生境高度搜索,加快算法的收敛速度.实验结果表明,所提出算法能够有效实现从全局探测到局部增强的自适应平滑过渡,是一种有效的多模态优化算法.
Aiming at the multi-modal optimization problem, a two-stage multi-modal differential evolution algorithm of dynamic niche radius is proposed. Based on the idea of conformational space annealing, a two-stage annealing strategy is proposed to dynamically adjust the niche radius. According to the annealing process, The process is divided into two stages: in the first stage, the high-quality new individuals are generated through the differential restriction mutation strategy to maintain the diversity of the population and promote the multi-mode convergence; in the second stage, using the seed proximity mutation strategy, Habitat height search speeds up the convergence of the algorithm.The experimental results show that the proposed algorithm can effectively achieve the adaptive smooth transition from global detection to local enhancement and is an effective multi-modal optimization algorithm.