Improved particle swarm algorithm with a novel local search
Mingqian Wang, Zhiguo Shi, Haishan Zhao · 2010
This paper presents an improved particle swarm optimization (PSO) algorithm, called IPSO, which employs a novel local search operator. The main idea of IPSO contains three steps. First, we create two trail particles in the local area of a particle. Then, an elitist mechanism is used to select the best one among the current particle and the two trail particles. Third, we replace the current particle with the fittest one. Experimental studies on ten benchmark functions show that the proposed approach IPSO outperforms standard PSO in all test cases.