Computational Technology Resources - CCC

A. Silva, A. Meixedo, P.A. Montenegro, D. Ribeiro, "A Data-Driven Methodology for Damage Detection in a Short-Span Filler-Beam Railway Bridge", in P. Iványi, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Eighteenth International Conference on Civil, Structural and Environmental Engineering Computing", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 10, Paper 2.7, 2025,

Keywords: structural health monitoring, damage detection, railway bridges, sparse autoencoders, deep learning, filler-beam bridge.

Abstract

Ensuring the structural integrity of railway bridges is a vital concern in infrastructure management, particularly for short-span filler-beam bridges that are prone to degradation under repetitive loading. This work proposes a hybrid data-driven methodology to detect early-stage damage in such structures, using the Cascalheira bridge in Portugal as a case study. The approach integrates signal processing techniques (Continuous Wavelet Transform and Principal Component Analysis), deep learning (Sparse Autoencoders), and statistical tools (Mahalanobis distance and outlier analysis) to extract and refine damage-sensitive features from simulated acceleration responses. A comprehensive numerical model that accounts for train-bridge dynamic interactions and realistic track irregularities supports the simulation framework. Results demonstrate that the proposed method achieves reliable damage identification with a low false positive rate, even under significant environmental and operational noise. This robust and scalable strategy offers a promising advancement for indirect Structural Health Monitoring systems in railway infrastructure.

download the full-text of this paper (PDF, 5874 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 10 PROCEEDINGS OF THE EIGHTEENTH INTERNATIONAL CONFERENCE ON CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING COMPUTING Edited by: P. Iványi, J. Kruis and B.H.V. Topping Paper 2.7 A Data-Driven Methodology for Damage Detection in a Short-Span Filler-Beam Railway Bridge A. Silva¹, A. Meixedo¹, P.A. Montenegro¹ and D. Ribeiro^1,2 ¹CONSTRUCT-iRAIL, Faculty of Engineering, University of Porto, Portugal ²iBuilt, ISEP, Polytechnic of Porto, Portugal Full Bibliographic Reference for this paper A. Silva, A. Meixedo, P.A. Montenegro, D. Ribeiro, "A Data-Driven Methodology for Damage Detection in a Short-Span Filler-Beam Railway Bridge", in P. Iványi, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Eighteenth International Conference on Civil, Structural and Environmental Engineering Computing", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 10, Paper 2.7, 2025, Keywords: structural health monitoring, damage detection, railway bridges, sparse autoencoders, deep learning, filler-beam bridge. Abstract Ensuring the structural integrity of railway bridges is a vital concern in infrastructure management, particularly for short-span filler-beam bridges that are prone to degradation under repetitive loading. This work proposes a hybrid data-driven methodology to detect early-stage damage in such structures, using the Cascalheira bridge in Portugal as a case study. The approach integrates signal processing techniques (Continuous Wavelet Transform and Principal Component Analysis), deep learning (Sparse Autoencoders), and statistical tools (Mahalanobis distance and outlier analysis) to extract and refine damage-sensitive features from simulated acceleration responses. A comprehensive numerical model that accounts for train-bridge dynamic interactions and realistic track irregularities supports the simulation framework. Results demonstrate that the proposed method achieves reliable damage identification with a low false positive rate, even under significant environmental and operational noise. This robust and scalable strategy offers a promising advancement for indirect Structural Health Monitoring systems in railway infrastructure. download the full-text of this paper (PDF, 5874 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions