花朵授粉算法的优化

Optimization of flower pollination algorithm

针对花朵授粉算法(FPA)寻优过程中局部深度搜索能力弱、易陷入局部最优、后期收敛速度慢的问题,提出一种基于自适应高斯变异的混合蛙跳花朵授粉算法(AGM-SFLFPA)。借鉴混合蛙跳算法(SFLA)思想,对种群个体按照适应度值进行排序、分组并更新各分组中最差个体的位置,增强算法的局部深度搜索能力并增加种群多样性;通过公示牌动态监测算法是否陷入局部最优,当陷入时,将自动对全局最优个体执行高斯变异操作,提高个体跳出局部最优的能力、增强种群多样性、加快收敛速度。通过6个典型的标准测试函数从4个方面验证该算法的有效性,验证结果表明,AGMSFLFPA具有更好的稳定性和可靠性、更快的收敛速度及更高的寻优精度,适用于高维复杂多极值函数求解问题。

To solve the problems of the poor local deeply searching ability,easily falling into local optimum,and the low convergence precision of the flower pollination algorithm(FPA)in the process of iterative solution,a flower pollination algorithm based on adaptive Gauss mutation and shuffled frog leaping(AGM-SFLFPA)was proposed.The population was sorted according to the fitness value of individual and they were grouped based on a certain rule using the thought of shuffled frog leaping algorithm(SFLA).The worst flower of each group was updated using the update strategy of SFLA,which not only enhanced the local depth search ability of the algorithm,but also increased the population diversity.The public signs were introduced to monitor whether the iterative process was into a local optimal solution.The Gauss mutation operation was performed on the global optimal individuals when the iterative process fell into the local optimal solution,which not only improved the ability of the individual to jump out of the local optimum,but also enhanced the population diversity and accelerated the convergence rate.To verify the effectiveness of AGM-SFLFPA,six typical standard test functions and tests were selected from four aspects respectively.Experimental results show that AGM-SFLFPA has better stability and reliability,faster convergence speed and higher precision.And it is suitable for solving high dimensional multi-extremum complex function problems.