摘要
花粉算法是一种新型的元启发式智能算法,但存在陷入局部最优解、收敛速度慢、寻优精度低等缺陷.基于此本文提出一种基于协作搜索策略的花粉算法,该算法使得花粉个体能够在一定程度上跳出局部最优值,提高算法的全局寻优能力.最后,对8个标准测试函数进行测试,结果表明,改进后的算法在7个测试函数中能够找到理论最优值,收敛速度、寻优精度、鲁棒性均比花粉算法以及改进的花粉算法有较大的提高.
Flower pollination algorithm is a new kind of meta-heuristic intelligent algorithm.As a result of the existence of local optimal solution,slow convergence rate and low precision,a flower pollination algorithm based on cooperative search strategy is proposed.The flower pollination individual can jump out of the local optimization value to a certain extent and improve the global optimization ability of the algorithm.The test of eight standard test functions made,the results show that the improved algorithm can not only find the optimal value of the theory in seven test functions,but also greatly improve the convergence rate,optimization accuracy and robustness.
出处
《纺织高校基础科学学报》
CAS
2017年第4期561-566,573,共7页
Basic Sciences Journal of Textile Universities
基金
陕西省软科学研究计划项目(2014KRM2801)
西安市教育科学重大招标项目(2015ZB-ZY04)
关键词
花粉算法
协作搜索
寻优性能
适应度值
flower pollination algorithm
cooperative search
optimization performance
fitness
