MODAL PARAMETER IDENTIFICATION OF STRUCTURAL MODEL BASED ON OUTPUT SEISMIC RESPONSE DATA
The objective of the paper is to present a method for modal parameter identification of structural shaking table test model by using the simulated earthquake response data. An adaptive on-line system identification method, the adaptive forgetting through multiple models, is introduced to investigate whether the time-varying response occurred or not during the process of vibration. Then, the whole response data such as an acceleration time history is divided into three segments according to the adaptive system identification result if time-varying response occurs. Lastly, the third segment data is used to identify the modal parameters by using the off-line system identification method of ARX (Auto-Regression with eXogenous variables) model to understand the dynamic properties of structural model after one strong motion input. On the contrary, the whole response time history is adopted to identify the modal parameters if the response is time-invariant. The method is applied to the shaking table test data of a 12-storey RC-frame structure model (scale 1:10) to identify its modal frequencies, damping ratios and mode shapes, and it is shown that the dynamic properties can be easily evaluated by the method. Furthermore, the results obtained from the seismic response data are compared with the modal analysis results that are identified from response data under the excitation of white-noise input, and it is found that the method is with high availability and identification accuracy. The results identified from the simulated seismic response data can be as the useful supplements of modal analysis of structural model experiment.
Modal parameter identification Transfer-function Structural model test ARX model
Maosheng Gong Lili Xie Junwu Dai
Institute of Engineering Mechanics, CEA, China lnstitute of Engineering Mechanics, CEA, China
国际会议
上海
英文
536-543
2007-12-04（万方平台首次上网日期，不代表论文的发表时间）