The Analysis and Improvement of Binary Particle Swarm Optimization

In this paper, the binary particle swarm optimization (PSO) is analyzed with bit change rate and velocity expected value, which results is that binary PSO is more and more stochastic, has the powerful ability of global search, but cannot converge to the optimal particle of swarm. So the binary PSO is lack of local exploration which instructs the improvement of BPSO. Based on the analysis, an improved binary PSO is proposed which changes the formula of its probability mapping and the formula of bit obtaining value. The new formulas are favorable of particle´s convergence to the optimal particle and to intensify the local exploration of binary PSO. With 0/1 knapsack problem, the experiment conducted in this paper shows that the improved binary PSO is outperformed to original binary PSO.