基于快速滑窗QR分解的自适应子空间辨识

化工学报 ›› 2008, Vol. 59 ›› Issue (6): 1448-1453.

基于快速滑窗QR分解的自适应子空间辨识

谢磊; 梁武星;张泉灵;张建明;王树青

浙江大学系统与控制研究所，工业控制技术国家重点实验室;电力电子工程及计算机科学学院

出版日期:2008-06-05 发布日期:2008-06-05

Adaptive subspace identification based on fast moving window QR decomposition

XIE Lei; LIANG Wuxing; ZHANG Quanling;ZHANG Jianming; WANG Shuqing

Online:2008-06-05 Published:2008-06-05

摘要/Abstract

摘要：

子空间辨识直接由输入输出数据辨识得到过程状态空间模型，在多变量系统的辨识中取得广泛应用。实现在线子空间辨识算法的关键在于快速、高效的QR分解及SVD分解更新算法。通过将Updating和Downdating操作有效结合，提出了一种快速的滑窗QR分解算法，减少了不必要的重复步骤，进一步提高了计算效率。复杂度分析结果表明，随数矩阵行数增加，快速滑窗QR分解算法比Updating、Downdating两步法可以减少8.3%的计算量。将快速滑窗QR分解算法用于PO-MOESP子空间辨识算法的自适应更新，并通过数值仿真实例验证了算法的有效性。

关键词:

滑窗QR分解, PO-MOESP, 子空间辨识, 自适应辨识

Abstract:

Subspace identification identifies the state space model directly based on the process input and output data.Due to its simplicity and efficiency, subspace identification method is widely used in MIMO process identification.The main obstacle towards implementing adaptive subspace identification is developing online QR and SVD decomposition algorithms.A fast QR decomposition procedure was proposed by simultaneously applying data updating and downdating, which outperformed traditional algorithms in terms of computational efficiency.Complexity analysis showed that fast QR decomposition could save 8.3% computational cost compared with the two-step approach.Numerical simulation study demonstrated the efficiency of the proposed algorithm.

Key words:

滑窗QR分解, PO-MOESP, 子空间辨识, 自适应辨识

谢磊, 梁武星, 张泉灵, 张建明, 王树青. 基于快速滑窗QR分解的自适应子空间辨识 [J]. 化工学报, 2008, 59(6): 1448-1453.

XIE Lei, LIANG Wuxing, ZHANG Quanling, ZHANG Jianming, WANG Shuqing. Adaptive subspace identification based on fast moving window QR decomposition[J]. CIESC Journal, 2008, 59(6): 1448-1453.