Damage detection and quantification in deck type arch bridges using vibration based methods and artificial neural networks

Jayasundara, N. and Thambiratnam, D. P. and Chan, T. H. T. and Nguyen, A. (2019) Damage detection and quantification in deck type arch bridges using vibration based methods and artificial neural networks. Engineering Failure Analysis (Article 104265). ISSN 1350-6307

[img] Text (Accepted Version)
Revised paper.pdf
Restricted - Available after 1 December 2020.


Abstract

Vibration based methods can be used to detect damage in a structure as its vibration characteristics change with physical changes in the structure. Extensive research has been carried out on the use of such methods to detect damage in a number of simple and some complex structures. Arch bridge is a popular type of bridge with rather complex vibration characteristics which pose a challenge for using existing vibration based methods to detect damage in them. Further, its complex form of damage detection, even with modified vibration based methods makes the quantification process harder and challenging. This paper develops and applies a vibration based method especially suited for arch bridges to detect, locate and quantify damages in the structural components. In the proposed method, modified forms of the modal flexibility (MMF) and modal strain energy (MMSE) based damage indices coupled with the Artificial Neural Network (ANN) technology is used to provide an overall damage assessment. The procedure to detect and locate damage was experimentally validated and applied to a full scale long span arch bridge under a range of damage scenarios. Damage indices obtained from noise polluted vibration data are then used as input data for training and validation of the neural networks. Two neural networks were trained separately using MMF and MMSE damage indices and a network fusion approach is used to obtain unambiguous and accurate results for detecting, locating and quantifying damages. The trained neural network system was then successfully applied to identify unknown damages using only vibration data of damaged structural elements of arch bridges. The findings of this paper will contribute towards the safe and efficient operation of arch type bridges.


Statistics for USQ ePrint 37495
Statistics for this ePrint Item
Item Type: Article (Commonwealth Reporting Category C)
Refereed: Yes
Item Status: Live Archive
Additional Information: Published online: 4 November 2019. Accepted version embargoed until 1 December 2020 (12 months), in accordance with the copyright policy of the publisher.
Faculty/School / Institute/Centre: Current - Faculty of Health, Engineering and Sciences - School of Civil Engineering and Surveying (1 July 2013 -)
Faculty/School / Institute/Centre: Current - Institute for Advanced Engineering and Space Sciences - Centre for Future Materials (1 Jan 2017 -)
Date Deposited: 18 Dec 2019 01:13
Last Modified: 17 Jan 2020 02:58
Uncontrolled Keywords: bridge failures; arch bridges; vibration based damage detection (VBDD); artificial neural networks (ANN); non-destructive testing
Fields of Research : 09 Engineering > 0905 Civil Engineering > 090505 Infrastructure Engineering and Asset Management
09 Engineering > 0905 Civil Engineering > 090506 Structural Engineering
Identification Number or DOI: 10.1016/j.engfailanal.2019.104265
URI: http://eprints.usq.edu.au/id/eprint/37495

Actions (login required)

View Item Archive Repository Staff Only