Deep Learning and Signal Processing Approaches for Real-Time Damage Detection and Prognosis in Large-Scale Civil Infrastructure: A Comparative Study on a Cable-Stayed Bridge

Author(s): Leena Markus Huffmann

Affiliation: Evangelical Presbyterian University College,, Ghana

Page No: 71-76

Volume, Issue & Publishing Year: Volume 3, Issue 7, July 2026

Journal: International Journal of Advanced Engineering Application (IJAEA)

ISSN NO: 3048-6807

Abstract:
Structural Health Monitoring (SHM) of large-scale civil infrastructure such as cable-stayed bridges demands continuous acquisition and interpretation of multi-channel sensor data across heterogeneous modalities — accelerometers, fibre-optic strain gauges, acoustic emission transducers, and corrosion probes — generating data volumes that overwhelm traditional signal processing paradigms. This study presents a comparative evaluation of six machine learning architectures — Support Vector Machines (SVM), Random Forest (RF), Long Short-Term Memory networks (LSTM), a convolutional-LSTM hybrid (CNN-LSTM), Autoencoder with multilayer perceptron classifier (AE-MLP), and a Vision Transformer (ViT) adapted for multivariate time-series — applied to a 1.2 km cable-stayed bridge instrumented with 196 sensors over a 36-month monitoring period. Damage scenarios simulated include wire fatigue in hangers, bearing degradation, anchor bolt loosening, and deck delamination across four severity levels (L1–L4). The CNN-LSTM hybrid achieves the highest overall detection accuracy of 97.4% (F1 = 0.974) with a mean time-to-detection of 4.2 minutes for L3 damage events, outperforming the baseline SVM by 6.2 percentage points. Explainability analysis via Gradient-weighted Class Activation Mapping (Grad-CAM) identifies frequency bands 0.3–2.1 Hz and 8.4–12.6 Hz as primary discriminative features for hanger wire fatigue and bearing degradation respectively. A lifecycle-integrated cost model demonstrates that early AI-driven detection of L2 damage reduces maintenance intervention cost by 38% relative to periodic manual inspection schedules, with a net present value improvement of INR 4.2 crore over 30 years.

Keywords: structural health monitoring, deep learning, LSTM, CNN, cable-stayed bridge, damage detection, vibration-based SHM, explainable AI, Grad-CAM, sensor fusion

Reference:

• [1] Doebling, S. W., Farrar, C. R., Prime, M. B., & Shevitz, D. W. (1996). Damage identification and health monitoring of structural and mechanical systems from changes in their vibration characteristics: a literature review. Los Alamos National Laboratory Report LA-13070-MS.
• [2] Farrar, C. R., & Worden, K. (2012). Structural Health Monitoring: A Machine Learning Perspective. John Wiley & Sons.
• [3] Maeck, J., & De Roeck, G. (1999). Dynamic bending and torsion stiffness derivation from modal curvatures and torsion rates. Journal of Sound and Vibration, 225(1), 153–170.
• [4] Johnson, E. A., Lam, H. F., Katafygiotis, L. S., & Beck, J. L. (2004). Phase I IASC-ASCE structural health monitoring benchmark problem using simulated data. Journal of Engineering Mechanics, 130(1), 3–15.
• [5] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780.
• [6] Avci, O., Abdeljaber, O., Kiranyaz, S., Hussein, M., Gabbouj, M., & Inman, D. J. (2021). A review of vibration-based damage detection in civil structures. Mechanical Systems and Signal Processing, 147, 107077.
• [7] Zhang, W., Peng, G., Li, C., Chen, Y., & Zhang, Z. (2017). A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals. Sensors, 17(2), 425.
• [8] Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., ... & Houlsby, N. (2021). An image is worth 16x16 words: Transformers for image recognition at scale. ICLR 2021.
• [9] Pathirage, C. S. N., Li, J., Li, L., Hao, H., Liu, W., & Ni, P. (2018). Structural damage identification based on autoencoder neural networks and deep learning. Engineering Structures, 172, 13–28.
• [10] Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. ICCV 2017, 618–626.
• [11] Sun, L., Shang, Z., Xia, Y., Bhowmick, S., & Nagarajaiah, S. (2020). Review of bridge structural health monitoring aided by big data and artificial intelligence. Journal of Structural Engineering, 146(11), 04020280.
• [12] Sohn, H., Farrar, C. R., Hemez, F. M., Shunk, D. D., Stinemates, D. W., Nadler, B. R., & Czarnecki, J. J. (2003). A review of structural health monitoring literature: 1996–2001. Los Alamos National Laboratory Report LA-13976-MS.