基于粒子滤波器的结构损伤识别研究

英文篇名：Particle Filter Based on Structural System Identification
中文刊名：建筑结构
英文刊名：Building Structure
作者：李烨 ; 王建 ; 李青 ; 钱鹏 ; 薛松涛 ; 唐和生
英文作者：Li Ye~1 ; Wang Jian~1 ; Lin Qing~1 ; Qian Peng~1 ; XueSongtao~2 ; Tang Hesheng~2 (1 East China Architecture Design & Research Institute Co. ; Ltd. ; Shanghai 200002 ; China ; 2 Research Institute of Structural Engineering and Disaster Reduction ; Shanghai 200092 ; China)
中文关键词：损伤识别 ; 粒子滤波 ; Bouc-Wen结构模型 ; 数值仿真
英文关键词：damage identification ; particle filter ; Bouc-Wen model ; numerical simulations
出版日期：2007-07-15
机构：华东建筑设计研究院有限公司;同济大学;
年：2007
期：S1
出版单位：建筑结构

摘要

将在信号处理、目标跟踪等领域中引起了众多学者的关注的粒子滤波方法引入到结构系统的损伤识别问题中作为研究对象。以土木结构中常用的Bouc-Wen滞回模型作为非线性结构损伤识别的研究对象,并考虑高斯噪声与非高斯噪声两种情况应用粒子滤波方法进行了非线性参数的识别,将识别结果与传统的EKF方法进行对比分析,仿真结果表明粒子滤波方法在非线性结构系统尤其是非线性、非高斯结构系统中与传统方法相比具有明显的性能优势。
Research in this paper aims to bring the sequential monte carlo(particle filter)method,which has a wide range of application in signal processing,statistics,and econometrics etc,into the structural parameter identification. Particle filter was introduced to the identification of Bouc-Wen Model which used to predict the nonlinear behavior of many structural systems and the results was compared with EKF method.The particle filter simulation results show that in non-Gaussian noise environment particle filter has also good accuracy,while the extended Kalman filter's performance degrades severely as the non-Gaussian effect increasing.

引文

[1] HOUSNER GW,BERGMAN L A,et al.Structural control:past,present and future[J].Engineering Mechanics,1997,123 (9) .
    [2] 胡洪涛,敬忠良.非高斯条件下基于粒子滤波的目标跟踪[J].上海交通大学学报,2004,38(12) .
    [3] GUT A.An Intermediate Course in Probability.Springer-Verlag,1995．
    [4] JAZWINSKI A H.Stochestic processes and filtering theory.Athematics in science and engineering.Academic Press,New York,USA,1970．
    [5] 冯新,李国强,周晶.土木工程结构健康诊断中的统计识别方法综述[J].地震工程与工程振动,2005,25(2) .
    [6] DOUCET A,DE FREITAS J F G,GORDON N J.Sequential Monte Carlo Methods in Practice [M].New York:Springer-Verlag,2001．