The Study Of Extenics-Based Multi Particle Swarm Optimization

PSO (Particle Swarm Optimization) algorithm has been concerned and researched widely by many scholars since it is proposed, which has been applied to solve various optimal problems in more and more fields. Along with the further research and the increasing complication of the awaiting solutions, all kinds of improvement measures has been proposed successively to perfect the PSO algorithm, which has made the application scope and the performance of PSO further improved. The modified PSO algorithms are not only limited to the parameters amendment, whose combinations with other optimization algorithms have constituted an important research direction of the PSO.Extenics is a new subject established originally by Chinese scholar, which is mainly used to solve contradiction problems in social lives. Extenics has its own theoretical framework and application research since it is proposed. In the recent years, the study on Extenics binding other optimization algorithms has got some advance, such as the extension genetic algorithm, which open a new thinking on how to combine Extenics with intelligent optimization algorithms.This paper introduces the Extenics into PSO, which proposed a MPSO algorithm based on Extenics. Comparing with the traditional PSO algorithm, it possesses more biological characteristics and is much more closed to the real rules of birds swarm's foraging. It makes the particles on different layers adapt different evolution measures. Then using extension transformation to make the bad particle put into service again. Through test function, algorithm efficiency has been validated.