摘要
采用基于个体排序的随机自适应Gaussian-Cauchy混合变异策略,将Gaussian和Cauchy变异算子结合起来以达到全局探索和局部搜索之间的动态平衡.改进的进化策略还使用重组算子、约束条件处理、精英保留策略以进一步提高算法的性能.将改进的混合策略算法应用于求多峰值函数极值问题,数值仿真实验结果显示了该算法的有效性.
In this paper,we employed random adaptive Gaussian-Cauchy hybrid mutation strategy based on the rank order of individuals.Gaussian and Cauchy mutation operator,constraint conditions and the elites saving strategy were also adopted in the improved evolution strategy to further enhance the performance.The proposed algorithm was applied to solving multi-peaks function extreme values learning.Numerical simulation results showed the effectiveness of the proposed algorithm
出处
《周口师范学院学报》
CAS
2009年第5期102-104,共3页
Journal of Zhoukou Normal University
关键词
进化算法
进化策略
函数极值
evolution algorithm
evolution strategy
function extreme values
