Volterra series identification method based on quantum particle swarm optimization
Li Zhinong1, Jiang Jing2, Chen Jingang2, Wu Guanhua1, Li Xuejun3
1. Key Laboratory of Nondestructive Testing, Ministry of Education, Nanchang Hangkong University, Nanchang 360063, China2. School of Mechanical Engineering, Zhengzhou University, Zhengzhou 450001, China3. Hunan Province key Lab of Health Maintenance for Mechanical Equipment,Hunan University of Science and Technology, Xiangtan, 411201
Abstract:The quantum particle swarm optimization (QPSO) algorithm is introduced into the nonlinear Volterra system identification, a new Volterra series identification method based on the QPSO is proposed. In the proposed method, the QPSO algorithm is used to estimate nonlinear system Volterra kernel function. The proposed method is compared with traditional least mean square (LMS) identification method. The simulation result shows that the two methods have good identification precision and convergence under free-noise interference. However, under noise environment, whether in identification precision, convergence or anti-interference, the proposed method is superior to the traditional LMS identification method, especially at signal-to-noise ratio (SNR) is very small.