PSI - Issue 78

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 78 (2026) 875–882

© 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of XX ANIDIS Conference organizers Keywords: Prestressed concrete bridge; Structural Health Monitoring; Model-based damage detection; Artificial Intelligence; Bayesian Neural Network Abstract In Structural Health Monitoring (SHM), maintaining infrastructure integrity under seismic conditions presents significant chal lenges, particularly due to uncertainties in the structural properties required for accurate numerical modeling. To address these challenges, this study proposes a Bayesian Neural Network (BNN) framework designed to detect and quantify seismically induced damage states in structures over time. By integrating multi-source data from various monitoring sensors, the BNN leverages recent advancements in artificial intelligence to provide probabilistic predictions. Unlike traditional neural networks, the BNN is partic ularly e ff ective when working with limited datasets and excels at updating predictions as new information becomes available. In this work, a prestressed concrete box-girder bridge with vertically prestressed internal joints is used as a representative case. A Finite Element Model (FEM) of the bridge is developed and calibrated using data from Ambient Vibration Tests. The SHM system comprises a sensor network distributed along the girder and substructure, enabling monitoring of both dynamic and static responses under seismic loading. The BNN is trained on SHM data to infer local variations in structural sti ff ness, especially targeting the sub structure, to solve the inverse problem of seismic damage detection, localization, and quantification. Parametric analyses show that the BNN can e ff ectively detect damage patterns resulting from seismic events, providing confidence intervals for each prediction. Importantly, the Bayesian learning process allows for continuous model refinement as new seismic or post-event inspection data are collected, enhancing the long-term robustness and reliability of SHM-based decision-making in seismically prone regions. XX ANIDIS Conference Bayesian neural networks for seismic damage detection in bridges using monitoring data Francesco Mariani a, ∗ , Laura Ierimonti a , Filippo Ubertini a , Ilaria Venanzi a a University of Perugia, Via G. Duranti 93, 06125, Perugia, Italy

1. Introduction

Infrastructure in seismic regions faces heightened risk due to the combined e ff ects of structural deterioration and repeated dynamic loading from earthquakes. In such contexts, the challenge of prioritising interventions across vast infrastructure networks is further exacerbated by limited economic resources for maintenance and retrofitting. Struc tural Health Monitoring (SHM) has been identified as a potentially e ff ective solution, with the ability to leverage

∗ Corresponding author. Tel.: + 39-327-549-0524. E-mail address: francesco.mariani@dottorandi.unipg.it

2452-3216 © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of XX ANIDIS Conference organizers 10.1016/j.prostr.2025.12.112