基于广义卡尔曼滤波的桥梁结构物理参数识别的子结构法

英文篇名：Proposing a Substructure Method of Extended Kalman Filter(EKF) for Identifying Physical Parameters of Bridge Structure
中文刊名：西北工业大学学报
英文刊名：Journal of Northwestern Polytechnical University
作者：吴子燕 ; 丁兰 ; 刘书奎
英文作者：Wu Ziyan ; Ding Lan ; Liu Shukui(Department of Civil Engineering ; Northwestern Polytechnical University ; Xi′an 710072 ; China)
中文关键词：桥梁结构 ; 参数识别 ; 广义卡尔曼滤波 ; 子结构
英文关键词：bridges ; Kalman filtering ; identification(control systems) ; extended Kalman filter(EKF) ; substructure method
出版日期：2010-06-15
机构：西北工业大学力学与土木建筑学院;
年：2010
期：03
出版单位：西北工业大学学报

摘要

文章将子结构法与广义卡尔曼滤波(EKF)算法相结合,推导出基于子结构有限元模型的状态方程和观测方程,对观测方程和状态方程组成的非线性参数系统应用广义卡尔曼滤波,成功识别出子结构的物理参数。该方法可以根据需要建立不同位置的子结构动力方程进行特定位置、并进一步实现整体结构的物理参数识别,为测量响应信息不完备条件下桥梁结构物理参数识别问题提供了一个较好的解决方法。数值算例表明,所提方法识别精度高、收敛速度快、抗噪性能好,适用于大型桥梁结构物理参数识别。
Aim.The traditional EKF method requires that all the response information be measured,which may not be the case for many bridge structures.Hence we propose our substructure method,which we explain in sections 1 and 2 of the full paper.Their core is: "Firstly,we select the substructure with complete response information as the research object.Secondly,the observation equation and the state equation based on finite element model are derived by the motion equations of the substructure.Applying the EKF method to the nonlinear parametric system consisting of the state equation and the observation equation,we identify the physical parameters of the substructure.Combined with EKF,the substructure method proposed in this paper offers an efficient approach for the nonlinear parametric system identification problem with incomplete response information." The simulation results,given in Table 2 and Fig.3,demonstrate preliminarily that the proposed approach is capable of identifying the physical parameters with rapidity and high precision.

引文

[1]Yang J N,Lin S L,Huang HW.An Adaptive Extended Kalman Filter for Structural Damage Identification.Journal of Structur-al Control and Health Monitoring,2006,13:849~867
    [2]Hoshiya M,Saito E.Structural Identification by Extended Kalman Filter.Journal of Engineering Mechanics,ASCE,1984,110(12):1757~1770
    [3]赵昕,李杰.一类加权全局迭代参数卡尔曼滤波算法.计算力学学报,2002,19(4):403~408Zhao Xin,Li Jie.A Weighted Global Iteration Parametric Kalman Filter Algorithm.Chinese Journal of Computational Mechan-ics,2002,19(4):403~408(in Chinese)
    [4]李杰,陈隽.子结构物理参数识别与输入地震动的复合反演研究.振动与冲击,1998,17(1):58~62Li Jie,Chen Jun.Study on Composite Inversion of Ground Motion and Substructural Parameter Identification.Journal of Vibra-tion and Shock,1998,17(1):58~62(in Chinese)
    [5]谢献忠,易伟建.结构物理参数时域识别的子结构方法研究.工程力学,2005,22(5):94~98Xie Xianzhong,Yi Weijian.A Substructure Method for Parameter Estimation in Time Domain.Engineering Mechanics,2005,22(5):94~98(in Chinese)
    [6]Serdar Soyoz.Health Monitoring of Existing Structure.University of California Irvine,2007