01370305

Component

Scour has been the most detrimental cause for bridge failures in the United States and around the world. Considerable efforts have been devoted to developing scour sensors that directly detect scour occurrence through underwater installation and operation. However, the cost and difficulty for installing and maintaining such a system may limit the wide application of these sensors. Noting that scour damages can lead to the change in bridge dynamic properties, alternative approach could be adopted for detecting scour by using vibration-based measurement, which is inexpensive and easily obtained at the upper portion of bridges through fixed installation or portable deployment. Nevertheless, this approach is the inverse solution to detecting scour problem and could often result in unidentification condition or non-uniqueness of solution. Besides, the change in dynamic properties of a bridge system could result form the environmental impacts, e.g., the change in the temperature or humidity, which needs to be sorted out. To overcome these drawbacks, this paper presents a new probabilistic computational framework for bridge scour damage identification on Bayesian inference. Within this probabilistic framework, uncertain model parameters regarding the scour damage’s extent and location, as well as its environmental factors, are formulated and represented by their prior statistical distributions based on engineering judgment. The statistical distributions of these model parameters can be then updated to represent the system’s damage and conditions through Bayesian inference with the available in-situ vibration-based measurements. The applicability of this probabilistic computational framework for bridge scour damage identification is demonstrated and explored through simplified models under several simulated scenarios, which include particularly the cases with the change of the temperature and humidity. Finally, the needs for the future research are also identified.

This paper was sponsored by TRB committee AHD35 Bridge Management

01362476

12-4195

English

17p

2012

Transportation Research Board 91st Annual Meeting

Digital/other

Figures; References; Tables

Bayes' theorem; Humidity; Mathematical models; Scour; Temperature

Bridges and other structures; Data and Information Technology; Highways; I24: Design of Bridges and Retaining Walls

Transportation Research Board Annual Meeting 2012 Paper #12-4195

TRIS, TRB

Feb 8 2012 5:22PM