An Improved Cooperative Quantum Particle Swarm Optimization Algorithm for Function Optimization

Abstract

Based on the PSO, co-evolution and quantum evolution, this paper proposes an improved cooperative quantum particle swarm optimization (ICQPSO) algorithm. In this algorithm, a new definition of Q-bit expression called quantum angle is proposed and all sub-swarms use the optimized cooperation mode, which not only ensures the convergence rate, but also avoids plunging into local optimum. Meanwhile, a comprehensive learning is introduced to strengthen the diversity of population and prevents the stagnation. On this basis, a disturbance mechanism is added, which is furthermore to avoid plunging into local optimum. The new algorithm is tested by four typical functions. Results of simulation experiments show that new algorithm conquers the stagnation effectively, improves the global convergence ability and has better optimization performance than traditional Quantum Genetic Algorithm.