Simplified particle swarm algorithm based on nonlinear decrease extreme disturbance and Cauchy mutation

Abstract

Concerning the drawbacks that particle swarm optimisation algorithm is easy to fall into the local optima, and has low solution precision, the simplified particle algorithm which based on the nonlinear decrease extreme disturbance and Cauchy mutation is proposed. The algorithm simplifies particle updating formula, and uses logistic chaotic sequence to initialise the particle position, which can improve the global search ability of population; nonlinear decrease extreme disturbance strategy enhanced the diversity of the population and avoid the particles trapping in local optimum; a novel Cauchy mutation is used for the optimal particle variation to generate more optimal guiding particle movement. The experimental simulation on seven typical test functions shows that the proposed algorithm can effectively avoid falling into local optimal solution, the search speed and optimisation accuracy have improved significantly. The algorithm is suitable to solve the function optimisation problem.