Online System Identification Based on Quantum-Behaved Particle Swarm Optimization Algorithm

In this paper, we explore the applicability of quantum-behaved particle swarm optimization (QPSO) algorithm, an efficient variant of particle swarm optimization (PSO) algorithm, to online system identification problems. First, quantum particle swarm optimization and particle swarm optimization are introduced. Then these two algorithms and genetic algorithms are applied to online identify parameters of a system described by differential equations respectively. Finally simulation results show that QPSO algorithm and PSO algorithm greatly accelerate the online identification. Convergence speed and accuracy of QPSO and PSO are far better than that of GA algorithm. Moreover the accuracy and convergence speed of QPSO is better than PSO.