摘要
为解决无约束全局最优问题,提出自适应双变异模式差分进化算法。该算法的变异规则结合差分进化算法中的两种基本变异模式,通过采用自适应缩放因子和交叉概率,来改善种群的多样性,平衡全局搜索和局部寻优能力。对高维benchmark典型函数进行数值仿真,与另外5种算法进行比较,比较结果表明,该算法具有较高的搜索精度、收敛速度以及较强的跳出局部最优解的能力。
An adaptive double mutation mode differential evolution algorithm was proposed for solving unconstrained global opti- mization problems. In the improved algorithm, the mutation rule was combined with two basic mutation modes. In addition, to improve the diversity of the population and balance the search capabilities between global and local search, the adaptive scaling factor and the crossover probability were used. This modification was used to enhance the local search ability and increase the convergence rate. Numerical experiments and comparisons on a set of well-known high dimensional benchmark functions indicate that the improved algorithm outperforms other existing differential evolution algorithms in terms of final solution quality, conver- gence rate, and the ability to jump out of local optimal solution.
出处
《计算机工程与设计》
北大核心
2015年第8期2250-2254,2270,共6页
Computer Engineering and Design
关键词
差分进化算法
进化模式
自适应
收敛速度
高维问题
differential evolution algorithm
evolution mode
self-adaptive
velocity of convergence
high-dimensional problem
