Bias Compensation Recursive Least Squares Identification for Output Error Systems with Colored Noises
-
摘要: 借助于偏差补偿原理和预滤波思想, 推导了有色噪声干扰输出误差系统参数估计的偏差补偿递推最小二乘 (Bias compensation recursive least squares, BCRLS) 辨识方法. 该方法降低了辨识对输入信号平稳性的要求, 实现了偏差补偿方法参数估计的递推计算, 可以用于在线辨识. 提出的递推 BCRLS 辨识方法优于非递推偏差补偿最小二乘算法, 提高了参数估计精度. 仿真试验证实了算法的有效性.Abstract: Based on the bias compensation principle and pre-filtering idea, this paper derives a bias compensation recursive least squares (BCRLS) identification algorithm for output error systems with colored noises. The algorithm proposed, which does not require the input signals to be stationary and ergodic, carries out the recursive computation of the bias compensation methods and can be on-line implemented. The BCRLS algorithm has advantage over the non-recursive bias compensation algorithms and can give highly accurate parameter estimation. The simulation results confirm the theoretical results.
-
计量
- 文章访问数: 4026
- HTML全文浏览量: 143
- PDF下载量: 1874
- 被引次数: 0
下载:
