Abstract
Abstract
The quantum-behaved particle swarm optimization (QPSO) has better global search capability and is regarded as an extremely effective improvement to the particle swarm optimization (PSO), however, there is still a population diversity decay in its operation. To promote the global search capability of QPSO, based on the weighted quantum-behaved particle swarm optimization (WQPSO) with the concept of weighted average optimal position, an improved QPSO based on particle aggregation-driven bias-average weighting is proposed, which analyses the action mechanism and parameter control law of bias-average weighting from the dimension of Euclidean distance, derives the judgment condition to avoid excessive aggregation of particle population, adopts average distance of particle population as the particle aggregation metric parameter to adjust the weighted bias centre, thus effectively increasing the traversal of the algorithm search space and avoiding premature convergence of the algorithm. The effectiveness of the improved algorithm proposed in this paper is demonstrated by applying several conventional test functions and comparing the analysis with PSO, QPSO and WQPSO.