基于混沌映射的混合花朵授粉优化算法

Mixed Flower Pollinate Algorithm Based on Chaotic Mapping

在花朵授粉算法的优化过程中,由于问题本身的局部极小性和复杂性,收敛精度不稳定,为了解决这一问题,本文提出一种新的混合花朵授粉算法,该算法引入Logistic映射,在局部寻优和全局搜索的过程中,周期性添加新的花朵个体,增加原有算法的种群多样性,有助于算法跳出局部极值。将该混合花朵授粉算法在函数优化中与原有基本算法进行仿真对比,结果表明其在收敛精度方面优于原有算法。

In the optimization process of the flower pollinate algorithm,the convergence accuracy is unstable because of the local minimum and complexity of the problem itself. In order to solve this problem,this paper proposes a new mixed flower pollinate algorithm. The Logistic mapping is integrated into the algorithm to add periodically new flower individuals. And then,the new flower pollinate algorithm can increase the population diversity in the process of local optimization and global search. In the function optimization simulation process,compared with the original flower pollinate algorithm,the convergence accuracy of this new flower pollinate algorithm has obvious advantages.