Particle Swarm Optimisation with Improved Learning Strategy

In this paper, a new variant of particle swarm optimisation (PSO) called PSO with improved learning strategy (PSO-ILS) is developed. Specifically, an ILS module is proposed to generate a more effective and efficient exemplar, which could offer a more promising search direction to the PSO-ILS particle. Comparison is made on the PSO-ILS with 6 well-established PSO variants on 10 benchmark functions to investigate the optimisation capability of the proposed algorithm. The simulation results reveal that PSO-ILS outperforms its peers for the majority of the tested benchmarks by demonstrating superior search accuracy, reliability and efficiency.