PSI - Issue 84

Valentina Giglioni et al. / Procedia Structural Integrity 84 (2026) 481–488

488

numerical models and monitored systems. By adopting an archetypal finite element model as a labeled source domain and a more refined model as a proxy target domain (real in service bridges), a OvR sensitivity analysis was performed to quantify the relevance of many damage-sensitive features across several simulated health-state scenarios. The resulting sensitivity heatmap highlighted a strong scenario-dependent variability in feature importance, demonstrating that different damage mechanisms are governed by distinct subsets of modal and rotational features. These findings (i) confirm that a single global feature set is inadequate for robust damage classification and effective transfer learning in complex bridge systems and (ii) naturally supports the implementation of multiple parallel DANNs, each tailored to a specific damage scenario and trained using only the most informative features for that class. This strategy is expected to improve domain alignment, reduce negative transfer effects, and enhance the interpretability of the learned representations. Future work will focus on the full implementation and validation of the proposed damage-tailored DANN framework, including the definition of statistical metrics to identify the most probable damage scenario in the target structure. Acknowledgements This work was supported by the Italian Ministry of University and Research (MUR) through the funded project of national interest “TIMING – Time evolution laws for IMproving the structural reliability evaluation of existING post tensioned concrete deck bridges” (Protocol No. P20223Y947). The authors would like to acknowledge the FABRE– Veneto Strade agreement for the support provided to this research activity. References Sconocchia, G. G., Mariani, F., Ierimonti, L., Meoni, A., Venanzi, I., Ubertini, F., 2024. Insights into structural assessment and long-term effects of a post-tensioned multispan concrete box girder bridge with vertically prestressed internal joints. Structures, 63, 106396. Figueiredo, E., Brownjohn, J., 2022. Three decades of statistical pattern recognition paradigm for SHM of bridges.Structural Health Monitoring, 21(6), 3018-3054. Svendsen, B. T., Øiseth, O., Frøseth, G. T., & Rønnquist, A., 2023. A hybrid structural health monitoring approach for damage detection in steel bridges under simulated environmental conditions using numerical and experimental data. Structural Health Monitoring, 22(1), 540-561. Sclafani, L., Stagi, L., Tronci, E. M., Betti, R., Milana, S., 2025. Real-time unsupervised structural damage detection using cepstral features and principal component analysis. Structural Health Monitoring, 14759217251355946. Giglioni, V., Venanzi, I., Baia, A. E., Poggioni, V., Milani, A., Ubertini, F., 2022. Deep autoencoders for unsupervised damage detection with application to the Z24 benchmark bridge. In European workshop on structural health monitoring, 1048-1057. Sun, Z., Sun, M., Siringoringo, D. M., Dong, Y., Lei, X., 2023. Predicting bridge longitudinal displacement from monitored operational loads with hierarchical CNN for condition assessment. Mechanical Systems and Signal Processing, 200, 110623. Cianci, E., Civera, M., De Biagi, V., Chiaia, B., 2026. Physics-informed machine learning for the structural health monitoring and early warning of a long highway viaduct with displacement transducers. Mechanical Systems and Signal Processing, 242, 113659. Pan, Q., Bao, Y., Li, H., 2023. Transfer learning-based data anomaly detection for structural health monitoring. Structural Health Monitoring, 22(5), 3077-3091. Giglioni, V., Eva, A. E., Worden, K., Ubertini, F., Venanzi, I., 2025. Assessing similarity requirements for effective transfer learning across a network of rigid frame bridges. Reliability Engineering & System Safety, 111994. Wang, J., Zhang, Z., Liu, Z., Han, B., Bao, H., Ji, S., 2023. Digital twin aided adversarial transfer learning method for domain adaptation fault diagnosis. Reliability Engineering & System Safety, 234, 109152. Jin, Y., Song, X., Yang, Y., Hei, X., Feng, N., Yang, X., 2025. An improved multi-channel and multi-scale domain adversarial neural network for fault diagnosis of the rolling bearing. Control Engineering Practice, 154, 106120. Singhal, P., Walambe, R., Ramanna, S., Kotecha, K. 2023. Domain adaptation: challenges, methods, datasets, and applications. IEEE access, 11, 6973-7020. Computers and Structures, 2018. Inc. SAP2000: Integrated Software for Structural Analysis and Design. Berkeley, CA, USA. Dassault Systèmes Simulia Corp. 2020. Abaqus Analysis User’s Guide. Providence, RI, USA.