An Improved Flower Pollination Algorithm

Abstract

In order to solve the problems of the poor local deeply searching ability, easily falling into local optimum, and the low convergence rate in the late iteration of the Flower Pollination Algorithm (FPA), a Flower Pollination Algorithm based on Adaptive Gauss Mutation and Shuffled Frog Leaping (AGM-SFLFPA) is proposed. First, drawing on the thoughts of Shuffled Frog Leaping Algorithm (SFLA), AGM-SFLFPA sorts the population according to the fitness value of individual, groups them and updates the location of the worst individual in each group. It not only enhances the local depth search ability, but also increases the population diversity. Then, Gauss mutation strategy is introduced, which is automatically performed on the global optimal individuals when the algorithm falls into the local optimal solution. The proposed algorithm not only improves the ability of the individual to jump out of the local optimum, but also increases the diversity of the population and accelerates the convergence rate. In this paper, four standard test functions are used to verify the validity of AGM-SFLFPA from four aspects. The experimental results show that AGM-SFLFPA has better stability and reliability, faster convergence speed and higher precision, which is suitable for solving high dimensional multi-extremum complex function problems.