An improved diversity-guided particle swarm optimisation for numerical optimisation (Q2341494)

Summary: Particle swarm optimisation (PSO) is a global optimisation technique, which has shown a good performance on many problems. However, PSO easily falls into local minima because of quick losing of diversity. Some diversity-guided PSO algorithms have been proposed to maintain diversity, but they often slow down the convergence rate. In this paper, we propose an improved diversity-guided PSO algorithm, which employs a local search to enhance the exploitation. In addition, a concept of generalised opposition-based learning is utilised for population initialisation and generation jumping to find high quality of candidate solutions. Experiments are conducted on a set of benchmark functions. Results show that our approach obtains a promising performance when compared with other PSO variants.