论文部分内容阅读
文章介绍了一种在噪声环境下,多自由度系统的参数识别方法。此方法包括一个迭代程序。这个程序通过最小二乘法,近似得到初始的参数估计值,此值送到一个相应的卡尔曼滤波器,从而得到改进的系统状态估计值。然后,将这个改进的估计值再用最小二乘法,得到系统参数的精确估计值。如此重复进行,直到参数收敛到可接受的范围内。参数误差可通过在滤波器中把伪噪声加到系统方程上的办法得到补偿。每次迭代后噪声强度都进行修正。经对两自由度系统的模拟研究表明:这种方法以相当低的计算费用,可获得可靠的系统参数的估计值。即使在数据记录时问短的情况下,也是如此。
This paper introduces a method of parameter identification in multi-degree-of-freedom system under noisy environment. This method includes an iterative procedure. This procedure approximates the initial estimate of the parameter by a least-squares method, which is fed to a corresponding Kalman filter to obtain an improved estimate of the system state. Then, using this improved estimate, the least squares method is used to obtain an accurate estimate of the system parameters. This is repeated until the parameters converge to an acceptable range. The parameter error can be compensated for by adding pseudo-noise to the system equation in the filter. The noise intensity is corrected after each iteration. Simulation studies on the two degrees of freedom system show that this method can obtain reliable estimates of system parameters with relatively low computational costs. This is true even when data recording is short.