¹School of Information Science and Engineering, Central South University, Changsha, China.
²Institute of Information and System Science, Beifang University of Nationalities, Yinchuan, China.

---

Abstract—An Dynamically Quantum Particle Swarm Optimization Algorithm with Adaptive Mutation (AMDQPSO) is given, the algorithm can better adapt to the problem of the complex nonlinear optimization search. The concept of the evolution speed factor and aggregation degree factor are introduced to this algorithm, and the inertia weight was constructed as a function of the evolution speed factor and aggregation degree factor, so that the algorithm has the dynamic adaptability in each iteration. This paper introduces the concept of the rate of cluster focus distance changing, and gives a new perturbations method. When the algorithm is found to sink into the local optimization, the new adaptive mutation operator and mutation probability are implemented at the best position of the global optimization. so that the proposed algorithm can easily jump out of the local optimization. The test experiments with six well-known benchmark functions show that the AMDQPSO algorithm improves the convergence speed and accuracy, strengthens the capability of local research and restrains the prematurity.

Index Terms—Quantum particle swarm optimization (QPSO), adaptive mutation, the rate of cluster focus distance changing, inertia weight.

[PDF]

Cite: Chenyang Gao, Ning Chen, Yuelin Gao, Jiajiang Zhang, "A Dynamically Quantum Particle Swarm Optimization Algorithm with Adaptive Mutation," Journal of Software vol. 11, no. 4, pp. 338-346, 2016.