基于Log-Linear模型的Gauss-Cauchy自适应细菌觅食算法

A Novel Bacteril Foraging Algorithm of Gauss-Cauchy Based on Log-Linear Model

针对基本细菌觅食算法在寻优过程中易在非全局价值点附近大量聚集,导致寻优精度降低、收敛速度过慢、细菌种群多样性降低等一系列问题,提出了一种基于Log-Linear模型的Gauss-Cauchy自适应细菌觅食算法。首先,将Log-Linear模型引入基本细菌觅食算法中用来优化细菌的三个行为;其次,在算法中引入自适应调整细菌的视野和步长的策略,细菌的搜索范围和寻优精度随也可随着算法的进行得到提高;再次,利用Gauss-Cauchy变异来提高细菌种群的多样性。仿真实验结果表明,该改进算法与其他优化算法进行比较,算法的收敛速度提高30%和寻优精度提高78%,并保持了细菌的多样性。

Aiming at some defects of the traditional bacterial foraging algorithm,such as low optimization precision and long runing time,a new bacterial foraging algorithm based on Log-Linear model and Gauss-Cauchy mutation is proposed.Firstly,the Log-Linear model is used to improve the three typical behaviors of the bacterial,which are foraging,chemotaxis ande reproduction and elimination-dispersal.Secondly,a policy which can adaptively adjust vision and step length of the bacterial in the new algo⁃rithm is adopted.Thirdly,the Gauss-Cauchy mutation is leveraged to keep individual diversity and to avoid falling into local extre⁃mum.Simulation results show that compared with other algorithms,the proposed algorithm has a better convergence speed and stability.