An Improved Particle Swarm Optimization Based on Bacterial Chemotaxis
Ben Niu, Yunlong Zhu, Xiaoxian He, Xiangping Zeng · 2006
Inspired by the phenomenon of chemotaxis in colonies of the bacteria, an improved particle swarm optimization (PSO) is presented by analogy to the way that bacteria react to chemo-attractants or chemo-repellents. The proposed algorithm (PSOBC) alternates between phases of attraction and repulsion. Once the diversity of population is too low, the individuals will be dispersed by repulsion force, while if the diversity of population is too high, the individuals have to be congregated by attraction force. This is accomplished by employing a diversity control method. Comparisons with standard PSO (SPSO) and it variants on a set of benchmark functions indicate that PSOBC not only prevents premature convergence to a high degree, but also keeps a more rapid convergence rate than SPSO