Numerical analyses of three inertia-weight-improvement-based particle swarm optimization algorithms

As one widely applied swarm intelligent algorithm, particle swarm optimization (PSO) algorithm has obtained the attention of various scholars with its advantages of easy implementation, high precision and fast convergence. Firstly, aiming to solve the problems that PSO has low searching speed and PSO is easy to fall into local optimal solution especially when dealing with high-dimension model, this paper modifies the PSO from the point of inertia weight improvements. The three improved inertia weights separately update the searching speed of particles with different principles, which are called linear-decline inertial weight, stochastic inertial weight and adaptive inertia weight. Then, three problems of evaluation non-linear function extremum are adopted in the simulation part to verify the convergence speed and optimization ability of three modified PSO algorithms. The numerical analyses demonstrate that these three improved inertia weights can guide particles to find the optimal solution more precisely and quickly. Thus, three proposed inertia-weight-improvement-based PSO algorithms have certain significance.