期刊
ENGINEERING STRUCTURES
卷 186, 期 -, 页码 410-420出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.engstruct.2019.02.041
关键词
System identification; Parameter estimation; Multi-response; Finite element method; Static measurements; Dynamic measurements
资金
- NRF's Campus for Research Excellence and Technological Enterprise (CREATE) programme, Singapore [FI 370074016]
- Land Transport Authority of Singapore (LTA)
In situ measurements have the potential to provide valuable information about the safety and the condition of bridges through implementation of system-identification methodology. A significant amount of research has focused on system identification using either dynamic or static measurements separately. Realizing the complementary relationship between static and dynamic measurements, traditional model updating methods adopt error functions to account for the residual between modeling and measured values for various types of measurements. Behavioral models may be inaccurate due to incomplete representation of modeling and measurement uncertainties. Furthermore, the normalization of error functions may bring additional uncertainty to the identification process. In this paper, an approach based on the model falsification method is proposed to combine both static and dynamic measurements with explicit consideration of both modeling and measurement uncertainties. A measurement selection strategy is also used to help detect abnormal measurements. The approach has been evaluated using a highway flyover bridge in Singapore. Dynamic measurement data include natural frequencies and mode shapes whereas static measurement data include inclinations, deflections and strains. By combining both static and dynamic measurements, this approach leads to falsification of additional model instances and obtains a more precise prediction of parameter values than approaches which interpret static measurements only.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据