基于粒子群量子操作算法的同步发电机非线性模型参数辨识
Parameters Identification of Nonlinear Model of Synchronous Generator Based on Particle Swarm Optimization Algorithm With Quantum Operation
-
摘要: 为解决同步发电机三阶非线性模型参数辨识问题,在将参数辨识问题转化为非线性优化问题的基础上,引入粒子群优化(particle swarm optimization,PSO)算法。考虑到PSO收敛速度慢、参数辨识精度低,而量子粒子群优化(quantumparticle swarm optimization,QPSO)算法收敛速度快、具有较好的全局搜索能力,将量子操作引入到PSO算法中,提出了粒子群量子操作(particle swarm optimization withquantum operation,PSO-QO)优化算法。仿真试验结果表明,与PSO算法、QPSO算法相比,PSO-QO算法收敛速度快、参数辨识精度高、算法更为稳定;与经典扩展卡尔曼滤波(extended Kalman filter,EKF)方法相比,PSO-QO具有更强的鲁棒性,尽管强噪声条件下参数辨识精度有所下降,但效果仍优于EKF。Abstract: With converting the problem of parameters estimation into parameters optimization problem,the paper introduced particle swarm optimization（PSO） algorithm to identification parameters of three order nonlinear synchronous generator model.Focus on the drawback of PSO in convergence and identification accuracy and the advantage of quantum particle swarm optimization（QPSO） in convergence speed and global search ability,the paper proposed a improved algorithm named Particle Swarm Optimization with Quantum Operation for parameters identification of synchronous generator.The simulation test results show that PSO-QO convergences much more fast than PSO and QPSO,meanwhile with a high accuracy.
下载:
