Improvement of Inertia Weight Declining Strategy Based on Particle Swarm Optimization Algorithm

Abstract

The standard particle swarm optimization algorithm introduces the inertia weighetw, which becomes a calculation method for finding the extremum of the function effectively. It is easy to converge fast. Nowadays, the linear decreasing dynamic, inertia weighting strategy is widely used. Although this improvement is very successful. However, the search process is a nonlinear and complex process. In order to better maintain the balance between global and local search capabilities, this paper proposes a nonlinear decreasing dynamic inertia weighting strategy based on the linear decreasing strategy, using Griewank, Rastrigrin, Sphere, The simulation experiments of four standard test functions of JDSchaffer were carried out, and the inertia weights in the basic particle swarm optimization algorithm were compared with the nonlinear decrement of fixed weight w = 0.95, linear decreasing LDIW and Chen[6] exponential curve respectively. The improved nonlinear weight decrement strategy is superior to other algorithms in terms of convergence speed, convergence precision number of iterations.