An improved quantum-behaved particle swarm optimization method for solving constrained global optimization problems

Abstract

A standard quantum-behaved quantum particle swarm optimization (QPSO) method outperforms a standard PSO approach in search ability and only needs a few parameter settings. To improve the capabilities of a standard QPSO algorithm, this study develops (1) a Cauchy mutation operator to increase the diversity of particles in a population, (2) an operator based on evolution generations to update a contraction expansion coefficient and (3) an elitist strategy to remain the strong particles. The proposed IQPSO algorithm is applied to solve constrained global optimization problems. This study compares the numerical results obtained using the IQPSO algorithm with those obtained using evolutionary algorithms and particle swarm optimization methods. Numerical results show that the proposed IQPSO approach can obtain the global optimal solution for a CGO problem and outperforms to some published algorithms.