摘要
将量子粒子群优化(QPSO)引入非线性Volterra系统辨识中,提出基于量子粒子群优化(QPSO)的Volterra级数辨识方法,利用QPSO算法估计出非线性系统的Volterra核函数。将所提方法与传统的最小二乘(LMS)辨识方法进行比较,仿真结果验表明,在无噪声干扰下,所提方法与LMS方法都具有很好的辨识精度和收敛性。而在有噪声干扰下,无论在辨识精度、收敛性和抗干扰性方面,所提方法都优于传统的LMS方法,且随噪声的增强优势越明显。
The quantum particle swarm optimization (QPSO) algorithm was introduced into a nonlinear Voherra system identification, a new Voherra series identification method based on QPSO was proposed. In the proposed method, the QPSO algorithm was used to estimate Voherra kernel functions of a nonlinear system. The proposed method was compared with traditional least mean square (LMS) identification method. The simulation results showed that the two methods have good identification precision and convergence under no noise interference; however, under noise environment, the identification precision, convergence and anti-interference of the proposed method are superior to those of the traditional LMS identification one, especially, when signal-to-noise ratio (SNR) is very small.
出处
《振动与冲击》
EI
CSCD
北大核心
2013年第3期60-63,74,共5页
Journal of Vibration and Shock
基金
国家自然科学基金(50775208,51075372)
湖南省机械设备健康维护重点实验室开放基金(200904)
江西省研究生教育创新基地基金
关键词
量子粒子群优化
VOLTERRA级数
非线性系统辨识
quantum particle swarm optimization (QPSO)
Volterra series
nonlinear system identification