Quasi-Static Vehicle Load Estimation on Bridge Using Explainable Artificial Intelligence-Based Flexural Rigidity Matrix

Man-Sung Kang; Dong-Woo Seo; Minjun Kong; Hojin Kim; Jaehwan Kim; Yun-Kyu An

doi:10.21729/ksds.2025.18.3.1

All Issue

2025 Vol.18, Issue 3 Next Page

Original Article

Quasi-Static Vehicle Load Estimation on Bridge Using Explainable Artificial Intelligence-Based Flexural Rigidity Matrix 설명가능 인공지능 기반 Flexural Rigidity Matrix를 이용한 교량상 의사정적 차량 하중 추정

30 September 2025. pp. 1-9

PDF XML

Abstract

This study proposes a quasi-static vehicle load estimation technique for bridge using an explainable artificial intelligence (XAI)-based flexural rigidity matrix (FRM), aiming to enhance the safety and sustainability of road and bridge infrastructure. The proposed XAI network consists of a physics-informed neural network (PINN), which incorporates governing equations and boundary conditions, and a deflection profiling network designed to generate training data that reflects bridge-specific characteristics. Point-wise responses obtained from spatially distributed displacement sensors are integrated within the network to reconstruct continuous bridge deflection profiles. Real-time sensor measurements are subsequently input into the XAI framework to generate updated deflection profiles and calibrate the FRM of the bridge. Based on the calibrated FRM, vehicle loads can be estimated by exploiting the linear relationship between load and deflection. The proposed methodology was experimentally validated on the Seorak Bridge, a single span arch bridge located in Sokcho, Gangwon-do, Korea, and the results demonstrated an estimation error of approximately 4%, confirming the reliability and practical applicability of the approach.

Keywords

Explainable artificial intelligence

Bridge deflection profile

Flexural rigidity

Inclinometer

Vehicle load estimation

본 연구는 안전한 도로 및 교량 인프라의 환경 유지를 위해 교량 구조물의 변위 계측 시스템을 통한 설명가능 인공지능(XAI: Explainable Artificial Intelligence) 기반 Flexural rigidity matrix(FRM)을 이용한 교량상 의사정적 차량 하중 추정 기술을 제안한다. XAI 네트워크는 지배방정식과 경계조건을 기반으로 하는 물리 기반의 Physics-informed Neural Network(PINN) 및 처짐 프로파일링 네트워크로 구성되고, 교량의 특성에 맞추어 학습 데이터를 생성한다. 이후 교량상 공간적으로 분포된 변위 계측 센서의 지점별(Point-wise) 응답을 융합하여 연속적인 교량 형상 프로파일을 생성할 수 있도록 네트워크를 구성하고 학습한다. 다음으로 교량에서 실제 계측되는 변위 센서 응답을 XAI 네트워크에 입력하여 실시간 교량 처짐 프로파일을 생성하고, 이를 기반으로 교량의 FRM을 교정한다. 교정된 FRM을 기반으로 실시간 계측되는 처짐 프로파일을 통해 하중과 처짐의 선형 관계에 따라 교량 통행 차량의 하중 추정이 가능하다. 제안기술은 강원도 속초에 위치한 설악대교에서 실험적으로 검증하였으며, 실험 결과 4% 내외의 오차를 가지고 하중을 추정하였다.

키워드

설명가능 인공지능

교량 처짐 프로파일

휨 강성

경사계

차량 하중 추정

References

Hussain, S., S. M. Hussain, Y. Xin, and Z. C. Wang. (2025). Vehicle Load Identification Based on Bridge Response Using Deep Convolutional Neural Network. Journal of Asian Architecture and Building Engineering. 24(3): 1235-1252.

10.1080/13467581.2024.2328634

Kim, S. W., D. W. Yun, D. U. Park, S. J. Chang, and J. B. Park. (2021). Estimation of Live Load Distribution Factor for a PSC I Girder Bridge in an Ambient Vibration Test. Applied Sciences. 11(22): 11010.

10.3390/app112211010

Li, D., M. Liu, L. Yang, H. Wei, and J. Guo. (2024). A Non-Contact Identification Method of Overweight Vehicles Based on Computer Vision and Deep Learning. Computer-Aided Civil and Infrastructure Engineering. 39(22): 3452-3476.

10.1111/mice.13299

Li, S., Z. Liang, and P. Guo. (2023). A FBG Pull-Wire Vertical Displacement Sensor for Health Monitoring of Medium-Small Span Bridges. Measurement. 225: 114042.

10.2139/ssrn.4281791

Liu, M., D. M. Frangopol, and S. Kim. (2009). Bridge Safety Evaluation Based on Monitored Live Load Effects. Journal of Bridge Engineering. 14(4).

10.1061/(ASCE)1084-0702(2009)14:4(257)

Paris, J. C., S. I. R. Amorim, and M. J. C. Minhoto. (2013). Impact of Traffic Overload on Road Pavement Performance. Journal of Transportation Engineering. 139(9).

10.1061/(ASCE)TE.1943-5436.0000571

Park, Sangki, Dong-Woo Seo, Ki-Tae Park, Hojin Kim, Wook Kim, Minjoon Kong, and Tackgon Lee. (2024). Development of Digital Sensor Based Real-Time Monitoring System for Aged Bridges. Journal of Korean Society of Disaster and Security. 17(4): 1-11.

10.21729/KSDS.2024.17.4.1

Ventura, R., B. Barabino, D. Vetturi, and G. Maternini. (2023). Bridge’s Vehicular Loads Characterization Through Weigh-in-Motion Systems. The Case Study of Brescia. European Transport. 90.

10.48295/ET.2023.90.6

Wang, H., T. Nagayama, T. Kawakatsu, and A. Takasu. (2023). A Data-Driven Approach for Bridge Weigh-in-Motion from Impact Acceleration Responses at Bridge Joints. Structural Control and Health Monitoring. 2287978.

10.1155/2023/2287978

Yu, Y., C. Cai, and L. Deng. (2016). State-of-the-Art Review on Bridge Weigh-in-Motion Technology. Advances in Structural Engineering. 19(9).

10.1177/1369433216655922

Zhang, C., W. Zhang, G. Ying, L. Ying, J. Hu, and W. Chen. (2024). A Deep Learning Method for Heavy Vehicle Load Identification Using Structural Dynamic Response. Computer & Structures. 297: 107341.

10.1016/j.compstruc.2024.107341

Korean References Translated from the English

박상기, 서동우, 박기태, 김호진, 김욱, 공민준, 이탁곤 (2024). 디지털 센서 기반 노후교량 상시 모니터링 시스템 개발. 한국방재안전학회 논문집. 17(4): 1-11.

10.21729/KSDS.2024.17.4.1

Information

Publisher :Korean Society of Disaster and Security
Publisher(Ko) :한국방재안전학회
Journal Title :Journal of Korean Society of Disaster and Security
Journal Title(Ko) :한국방재안전학회 논문집
Volume : 18
No :3
Pages :1-9
Received Date : 2025-09-01
Revised Date : 2025-09-23
Accepted Date : 2025-09-23
DOI :https://doi.org/10.21729/ksds.2025.18.3.1

Journal of Korean Society of Disaster and Security ISSN:2466-1147(Print) 2508-285X(Online) 한국방재안전학회 논문집

All Issue

Korean References Translated from the English