An improved QPSO algorithm and its application in the high-dimensional complex problems

• We propose an improved quantum particle swarm optimization (MSCQPSO) algorithm based on combing the simulated annealing (SA), Co-evolution theory and diversity-guided mutation.
• The MSCQPSO algorithm takes full advantage of the SA to improve the global search ability, diversity mutation to avoid declining the population diversity, the quantum behavior to change updating strategy, the cooperative search to solve the prematurely problem.
• The experiment results show that the MSCQPSO algorithm takes on the fast convergence and high searching accuracy.

In allusion to the deficiencies of the low computational efficiency and local optimal solution of particle swarm optimization (PSO) algorithm, an improved PSO algorithm based on combining the simulated annealing (SA), co-evolution theory, quantum behavior theory and diversity-guided mutation strategy (MSCQPSO) is proposed in this paper. In the proposed MSCQPSO algorithm, the population is divided into multi-populations according to the computed fitness value. The SA and diversity-guided mutation strategy are introduced to enhance the global search ability. The quantum behavior theory is introduced into co-evolution theory to change the updating mode of the particles in order to guarantee the simplification and effectiveness. In order to prove the validity of the proposed MSCQPSO algorithm, the ten high-dimensional complex benchmark functions are selected in here. The experiment results show that the proposed MSCQPSO algorithm takes on the fast convergence, the high searching accuracy, the better robustness for solving the high-dimensional complex problems than the PSO algorithm, the CPSO algorithm and HPSO algorithm.