基于动态自适应策略的改进差分进化算法  

Improved Differential Evolution Algorithm Based on Dynamic Adaptive Strategies

作　　者：王丛佼[1] 王锡淮[1] 肖健梅[1] 

机构地区：[1]上海海事大学电气自动化系,上海201306

出　　处：《计算机科学》2013年第11期265-270,共6页Computer Science

基　　金：上海市教委科研创新重点项目(12ZZ158);上海市教委重点学科建设项目(J50602)资助

摘　　要：针对差分进化算法处理复杂优化问题时存在后期收敛速度变慢、收敛精度不高和参数设置困难的问题,提出了一种基于动态自适应策略的改进差分进化算法(dn-DADE)。首先,新的变异策略DE/current-to-dnbest/1利用当前种群中的精英解引导有效的搜索方向来动态调整可选的精英解,使其在进化后期趋于全局最优解。其次,分别设计了缩放因子和交叉因子的自适应更新策略,使两者在搜索的不同阶段自适应变化,以弥补差分进化算法对参数敏感的不足,进一步提高算法的稳定性和鲁棒性。对14个benchmark函数进行了测试并与多种先进DE改进算法进行了比较,结果显示,dn-DADE算法具有较高的求解精度,收敛速度快,寻优性能显著。To solve problems of DE applied to complex optimization functions, an improved differential evolution algo- rithm （dn-DADE） based on dynamic adaptive strategy was proposed in this paper. Firstly, the elite solutions of current population were utilized in the new mutation strategy （DE/current-to-dnbest/1） to guide the search direction. Second- ly, the adaptive update strategies of scaling factor and crossover factor were designed for making parameter values self- adapting at different search stages to improve the stability and robustness of the algorithm. A set of 14 benchmark func- tions were adopted to test the performance of the proposed algorithm. The results show that dn-DADE algorithm has the advantages of remarkable optimizing ability, higher search precision, faster convergence speed and outperforms se- veral state-of-the-art improved differential evolution algorithms in terms of the main performance indexes.

关 键 词：差分进化 变异策略 动态调整 参数自适应 全局优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]