Quantum-behaved particle swarm optimization with adaptive mutation operator

In this paper, the mutation mechanism is introduced into Quantum-behaved Particle Swarm Optimization (QPSO) to increase the diversity of the swarm and then effectively escape from local minima to increase its global search ability. Based on the characteristic of QPSO algorithm, the two variables, global best position (gbest) and mean best position (mbest), are mutated with Cauchy distribution respectively. Moreover, the amend strategy based on annealing is adopted by the scale parameter of mutation operator to increase the self-adaptive capability of the improved algorithm. The experimental results on test functions showed that QPSO with gbest and mbest mutation both performs better than PSO and QPSO without mutation.