基于遺傳因子優(yōu)化粒子群算法

劉歡　郝俊　李文娜　陳雅夢

摘 要：針對粒子群在搜索最優(yōu)值時間過長、易陷入局部最優(yōu)、無法挑選出最優(yōu)值的問題，本文在粒子群算法的基礎(chǔ)上，結(jié)合遺傳算法中的變異因子，提出一種基于遺傳優(yōu)化粒子群的算法。首先，該算法采用對數(shù)函數(shù)遞減慣性策略加速粒子跳出局部最優(yōu)，其次，遺傳變異因子增加個體極值的多樣性來尋出最佳值;最后，基于一定的迭代次數(shù)，根據(jù)標準函數(shù)Rastrigin進行尋優(yōu)效果測試驗證，仿真結(jié)果表明，改進后的算法能夠避免進入局部最優(yōu)情況，并且在最佳適應(yīng)度、標準差和尋優(yōu)時長等性能指標優(yōu)越于其他算法。

關(guān)鍵詞：粒子群算法 遺傳因子 適應(yīng)度函數(shù)

Optimizing Particle Swarm Optimization Algorithm based on Genetic Factors

Liu Huan，Sean Sean，Li Wenna，Chen Yameng

Abstract：Aiming at the problem that the particle swarm is too long to search for the optimal value， it is easy to fall into the local optimal， and the optimal value cannot be selected. Based on the particle swarm algorithm and the mutation factor in the genetic algorithm， this paper proposes a method based on genetic optimization of particle swarm algorithm. First， the algorithm uses the logarithmic function decreasing inertia strategy to accelerate the particles out of the local optimum. Secondly， the genetic variation factor increases the diversity of individual extreme values to find the best value; finally， based on a certain number of iterations， it is performed according to the standard function Rastrigin. The optimization results are tested and verified， and the simulation results show that the improved algorithm can avoid entering the local optimal situation， and is superior to other algorithms in performance indicators such as the best fitness， standard deviation， and optimization time.

Key words：particle swarm optimization， genetic factor， fitness function