A simple and scalable structure of Particle Swarm Optimization based on linear system theory

Abstract Since it was fifirst presented, Particle Swarm Optimization (PSO) has seen numerous improvements as a traditional optimization process. PSO algorithm becomes more complex as a result of the majority of improvement strategies, which use learning model replacement or parameter adjustment to enhance PSO algorithm’s performance. Based on linear system theory, this paper proposes a simple and scalable framework for restructuring Particle Swarm Optimization (RPSO) and provides a new example of the RPSO algorithm framework, Q-RPSO. The framework of RPSO adopts one position updating formula instead of the original position and velocity updating formulas, which is unrelated to the velocity and the current position of PSO. The experiments have been carried out by comparison with the standard PSO algorithm and four PSO variants based on benchmark functions of CEC 2013. The experimental results demonstrate that, whether in terms of global exploration capability or convergence accuracy, Q-RPSO outperforms all competitor algorithms.