An Analysis of the Behavior of Original Discrete Binary Particle Swarm Optimization on Trap Functions

Binary particle swarm optimization (BPSO) is the discrete binary version of particle swarm optimization (PSO) which is based on a metaphor of social behavior of birds and is a new kind of evolutionary computation paradigm. As BPSO has been successfully applied in a wide range of real-world application fields and some BPSO variants have been proposed, there is a need to guide these two aspects in theory. In this paper, experimental methods are developed to analyze the behavior of BPSO on trap functions and the influence of deceptiveness on BPSO's search behavior is presented.