Dynamic Parameter Selection Based on Trigonometric Function in Particle Swarm Optimization
Wenxue Zhang, Tieke Li, Cantao Shi · 2010
This paper presents an improved PSO algorithm (TPSO) with dynamic parameter selection, in which the inertia weight and acceleration coefficients are defined by trigonometric functions of the present iterative generation. Compared with some typical PSO models, TPSO shows its superiority in convergence, robustness and solution quality through experimental tests with 8 widely used benchmark functions.