Population diversity analysis of The Adaptive Partly Informed PSO algorithm

Particle Swarm Optimization (PSO) is one of the newly developed intelligence optimization algorithms. With its simple concept, few parameters and scalable performance, PSO has become a very promising optimization tool and attracted extensive attention. In this paper, probabilistic analysis on swarm diversity in Adaptive Partly Informed Particle Swarm Optimization algorithm (API-PSO) was conducted. From the analysis results we can learn that the expectation of population diversity in the next moment are affected by the population size, problem-solving dimensions, the topology and the specific optimization function to be solved, ect. We also get the relationship between the current population diversity and the state of next time. This information could serve as the theoretic basis to solve swarm diversity lack, facilitate swarm evolution development and improve algorithm performance.