Abstract: In order to further improve the convergence speed differential evolution algorithm accuracy and stability, using a variety of techniques to increase the population convergence rate and reduce complexity; the use of elite regional learning strategy algorithm global search capability and algorithms further enhance the accuracy of the introduction of immune self-adaptive search strategy in order to achieve variation factor and crossover factor adaptive differential correction algorithm. Through five test functions, the proposed algorithm with the latest literature comparison algorithm, show the superiority of the algorithm in convergence speed, high precision aspects optimization capability dimensional problem.

Key words: differential evolution algorithm, multi-group technology, immune self-adaptive search strategies, elite area learning strategies

摘要： 为了进一步提高差分进化算法的收敛速度、算法精度和稳定性,采用多种群技术来增加算法收敛速度和降低复杂度;利用精英区域学习策略来对算法的全局搜索能力和算法精度进一步提升,引进自适应免疫搜索策略,以实现自适应修正差分算法的变异因子和交叉因子。通过五个测试函数,把本文算法与最新文献中的算法进行对比,表明算法在收敛速度、精度和高维问题寻优能力方面的优越性。

关键词: 差分进化算法, 多种群技术, 免疫自适应搜索策略, 精英区域学习策略