Application of Artificial Intelligence in Bridge Health Monitoring

Yanzu Tang

doi:10.54097/4qpd9060

Authors

Yanzu Tang

DOI:

https://doi.org/10.54097/4qpd9060

Keywords:

Global damage detection, Correlation model, Faulty data detection and classification.

Abstract

The global challenge of aging bridge infrastructure, combined with the significant limitations of conventional manual inspection methods—including their labor-intensive nature, subjective outcomes, and inability to provide continuous assessment—has created an urgent need for advanced monitoring solutions. This context establishes Artificial Intelligence (AI) as a fundamentally important technological innovation for modern bridge engineering. This paper provides a comprehensive examination of AI applications in structural health monitoring, with specific focus on three critical domains: vibration-based global damage detection, data-driven correlation modeling for deflection analysis, and automated fault detection in monitoring systems. The research methodology involves detailed analysis of the operational mechanisms underlying each application area, investigating how machine learning algorithms process complex vibration data to identify damage patterns, how correlation models establish relationships between environmental factors and structural behavior, and how anomaly detection techniques ensure data reliability. The findings demonstrate that AI technologies substantially enhance monitoring capabilities across multiple dimensions. These systems enable automated processing of complex structural data, facilitate early detection of potential damage through pattern recognition, and provide predictive insights into structural behavior under varying operational conditions. The implementation of AI-driven approaches represents a paradigm shift from reactive to proactive infrastructure management, allowing for more informed decision-making regarding maintenance scheduling and resource allocation. However, several challenges require attention, particularly concerning the scalability of these systems across diverse bridge types and the computational demands of real-time data processing. This research concludes that AI technologies hold transformative potential for advancing bridge health monitoring practices. Future developments should focus on enhancing system interoperability, improving algorithmic efficiency, and developing standardized implementation frameworks. The continued integration of AI into structural monitoring represents a crucial advancement toward creating more resilient, sustainable, and safely managed transportation infrastructure systems, ultimately contributing to improved public safety and optimized lifecycle management of critical bridge assets.

Downloads

Download data is not yet available.

References

[1] M. Mansouri, O. Avci, H. Nounou, M. Nounou, Iterated square root unscented Kalman filter for nonlinear states and parameters estimation: three DOF damped system, J. Civil. Struct. Health Monit., 5 (2015), pp. 493-508

[2] M. Mansouri, O. Avci, H. Nounou, M. Nounou, A comparative assessment of nonlinear state estimation methods for structural health monitoring, in: Proceedings of IMAC XXXIII Conference and Exposition on Structural Dynamics, Orlando, FL, USA, 2015.

[3] O. Abdeljaber, O. Avci, S. Kiranyaz, M. Gabbouj, and D. J. Inman, Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks, Journal of sound and vibration, vol. 388, pp. 154–170, 2017.

[4] Martinez D, Malekjafarian A, OBrien E. Bridge health monitoring using deflection measurements under random traffic. Struct Control Health Monit. 2020; 27(9): e2593.

[5] Liu Y, Deng Y, Cai CS. Deflection monitoring and assessment for a suspension bridge using a connected pipe system: a case study in China. Struct Control Health Monit. 2015; 22(12): 1408-1425.

[6] Tang ZY, Chen ZC, Bao YQ, Li H. Convolutional neural network-based data anomaly detection method using multiple information for structural health monitoring. Struct Control Health Monit. 2019; 26(1): e2296.

[7] Ye XJ, Chen BC. Condition assessment of bridge structures based on a liquid level sensing system: theory, verification and application. Arab J Sci Eng. 2019; 44(5): 4405-4424.

[8] Zhou Y, Sun LM, Fu ZH, Jiang Z, Ren PJ. General formulas for estimating temperature-induced mid-span vertical displacement of cable-stay bridges. Eng Struct. 2020; 221:111012.

[9] Yue ZX, Ding YL, Zhao HW. Deep learning-based minute-scale digital prediction model of temperature-induced deflection of a cable-stayed bridge: case study. J Bridge Eng. 2021; 26(6):05021004.

[10] Y. Deng, H. Ju, W. Zhai, A. Li, and Y. Ding, Correlation model of deflection, vehicle load, and temperature for in‐service bridge using deep learning and structural health monitoring, Structural control and health monitoring, vol. 29, no. 12, Art. no. e3113, 2022.

[11] Ricardo D, Joe QS, Edgar Thomas F, McAvoy Thomas J. Identification of faulty sensors using principal component analysis. AIChE J. 1996; 42(10): 2797-2812.

[12] Maher A, Friswell Michael I. Sensor validation for structural systems with additive sensor faults. Struct Health Monit. 2004; 3(3): 265-275.

[13] Chun L, Lynch Jerome P, Mingyan L. Distributed model-based nonlinear sensor fault diagnosis in wireless sensor networks. Mech Syst Signal Process. 2016; 66: 470-484.

[14] Ka-Veng Y, Ortiz Gilberto A. Outlier detection and robust regression for correlated data. Comput Methods Appl Mech Eng.2017; 313: 632-646.

[15] X. Jian, H. Zhong, Y. Xia, and L. Sun, Faulty data detection and classification for bridge structural health monitoring via statistical and deep‐learning approach, Structural control and health monitoring, vol. 28, no. 11, Art. no. e2824, 2021.