Individual predicted integral-controlled particle swarm optimisation

Abstract

Integral-controlled particle swarm optimisation (ICPSO) is a novel variant of particle swarm optimisation by incorporating two integral controllers into the methodology. Although the population diversity of ICPSO is large enough, the performance is not always well when dealing with multi-modal problems, the reason is partly due to the wrong guiding direction dominated by the best location found by entire swarm. Therefore, in this paper, a new strategy, quadratic interpolation method, is introduced to estimate the exact local optima in the best location neighbours. Simulation results show this new variant provide an efficient local search capability than other three variants of particle swarm optimisation when solving multi-modal numerical problems.