Thermal System Identification Based on Double Quantum Particle Swarm Optimization

Abstract

In order to improve the convergence speed and precision of particle swarm optimization (PSO) and quantum PSO (QPSO), inspired by the idea of quantum physics, a new improved QPSO algorithm named double QPSO (DQPSO) is presented. The particle’s encoding mechanism and the evolutionary search strategy are quantized in DQPSO algorithm, in which the evolution equation of the velocity vector is abandoned, thus the evolution equation is easier, and less parameter are used that makes the algorithm easier to control. Several benchmark multi-modal functions are used to test the proposed DQPSO algorithm, which verified that the new algorithm is superior to standard PSO and QPSO in search capabilities. Then, DQPSO is successfully used to the identification of a thermal system with pure time-delay and non-minimum phase. Finally, the algorithm is applied to the transfer function identification of thermal system based on field operation data.