Multiobjective Optimization Using Clustering Based Two Phase PSO

A clustering based two phase PSO strategy CTPPSO was developed to solve multiobjective optimization problems (MOPs) in this paper. The basic idea is that the initial population was constructed according to the distribution of the particles. The sub-populations which represent the groups of particles specialized on niches were dynamically identified using density-based clustering algorithms. The particle evolution was bounded in each niche. No information was exchanged among different niches, and then the population diversity was kept. Benchmark function optimization and MOPs experimental results demonstrate the effectiveness and efficiency of the proposed strategy.