融合高斯变异和Powell法的花朵授粉优化算法

发布时间：2019-03-31 13:25

【摘要】：花朵授粉算法(flower pollination algorithm,FPA)是最近提出的一种新型群智能优化算法,由于其较好地解决了全局搜索和局部搜索的平衡性问题,且具有参数少,易实现等特点,已得到广泛应用和研究,但现有研究对其参数的研究较少,同时该算法也存在演化后期收敛速度慢且易陷入局部极小等缺陷,使其应用范围受到制约。为了提升FPA算法的整体性能,对其控制步长的缩放因子的取值进行了修正;提出了把高斯变异和Powell法融入到花朵授粉算法中的混合算法GMPFPA(flower pollination algorithm combination with Gauss mutation and Powell search method)。改进算法首先利用高斯变异对全局搜索进行扰动,增强种群的多样性,提高全局探测能力,然后引入局部寻优能力强大的Powell法提升其局部开发能力。通过12个高维经典测试函数对比实验,验证了改进算法的有效性和优越性。
[Abstract]:Flower pollination algorithm (flower pollination algorithm,FPA) is a new group intelligence optimization algorithm proposed recently. Because it solves the balance problem of global search and local search well, it has the characteristics of few parameters and easy to realize. It has been widely used and studied, but there are few researches on its parameters. At the same time, the algorithm has some defects, such as slow convergence rate and easy to fall into local minimum in the late stage of evolution, which restricts its application scope. In order to improve the overall performance of the FPA algorithm, the scaling factor of its control step is modified, and a hybrid algorithm, GMPFPA (flower pollination algorithm combination with Gauss mutation and Powell search method)., which integrates Gao Si mutation and Powell method into the flower pollination algorithm is proposed. The improved algorithm first uses Gao Si mutation to disturb the global search, enhances the diversity of the population and improves the global detection ability. Then, the Powell method with strong local optimization ability is introduced to improve its local development ability. The effectiveness and superiority of the improved algorithm are verified by comparison of 12 high-dimensional classical test functions.
【作者单位】：江西财经大学信息管理学院;河池学院计算机与信息工程学院;
【基金】：国家自然科学基金No.F020204 广西自然科学基金No.2013GXNSFBA019022 广西高校科学技术研究项目Nos.KY2015LX332,KY2015LX334 江西省研究生创新项目No.YC2015-B054 河池学院计算机网络与软件新技术重点实验室资助项目No.2013-03;河池学院教改项目No.2014EB022;河池学院基金项目No.XJ2015QN003~~
【分类号】：TP18

【相似文献】