DOI QR코드

DOI QR Code

Damage Detection of Non-Ballasted Plate-Girder Railroad Bridge through Machine Learning Based on Static Strain Data

정적 변형률 데이터 기반 머신러닝에 의한 무도상 철도 판형교의 손상 탐지

  • 문태욱 (인하대학교 토목공학과) ;
  • 신수봉 (인하대학교 사회인프라공학과)
  • Received : 2020.11.10
  • Accepted : 2020.12.15
  • Published : 2020.12.31

Abstract

As the number of aging railway bridges in Korea increases, maintenance costs due to aging are increasing and continuous management is becoming more important. However, while the number of old facilities to be managed increases, there is a shortage of professional personnel capable of inspecting and diagnosing these old facilities. To solve these problems, this study presents an improved model that can detect Local damage to structures using machine learning techniques of AI technology. To construct a damage detection machine learning model, an analysis model of the bridge was set by referring to the design drawing of a non-ballasted plate-girder railroad bridge. Static strain data according to the damage scenario was extracted with the analysis model, and the Local damage index based on the reliability of the bridge was presented using statistical techniques. Damage was performed in a three-step process of identifying the damage existence, the damage location, and the damage severity. In the estimation of the damage severity, a linear regression model was additionally considered to detect random damage. Finally, the random damage location was estimated and verified using a machine learning-based damage detection classification learning model and a regression model.

국내의 노후 철도교량이 증가함에 따라 노후화로 인한 유지관리비가 점점 증가하고 있으며, 지속적인 관리가 더욱 더 중요해지고 있다. 하지만 관리해야하는 노후 시설물은 증가하지만, 노후 시설물을 점검 및 진단을 할 수 있는 전문 인력은 부족해지고 있다. 이러한 문제를 해결하기 위해 본 연구는 정적 변형률 응답 데이터를 적용하여 AI 기술의 머신러닝 기법으로 구조물의 국부적인 손상을 탐지하는 개선된 학습모델을 제시하고자 한다. 손상탐지 머신러닝 학습 모델을 구성하기 위해 우선 무도상 철도 판형교의 설계도면을 참고하여 교량의 해석모델을 설정하였으며, 설정된 해석모델로 손상시나리오에 따른 정적변형률 데이터를 추출하여 통계적 기법을 이용해 교량의 신뢰도 기반의 Local 손상 지수를 제시하였다. 손상 탐지는 손상 유무 탐지, 크기 탐지, 위치 탐지 3단계의 과정을 수행하여 손상 크기 탐지에서 선형 회귀 모델을 추가로 고려해 임의의 손상을 탐지하였으며, 최종적으로 손상 탐지 머신러닝 분류 학습 모델과 회귀 모델을 이용한 임의의 손상 위치를 추정 및 검증하였다.

Keywords

References

  1. Moon, T. U. (2020). Safety Evaluation of the Superstructure of Railway Non-ballasted Plate Girder Bridge in the Application of CWR, International Journal of Railway, 23(1), pp. 10-20.
  2. Ministry of Land Infrastructure and Transport (2017). Strategies to respond to the 4th industrial revolution, April, 21, 2017, pp. 1-16.
  3. Kim, D. G. (2018). Performance evaluation of SOC Structure using Deep Learning technology, Seoul: Korea Institute of Construction Technology, pp. 164-172.
  4. Gary Marcus. (2019). DeepMind's Losses and the Future of Artificial Intelligence, WIRED, Retrieved from https://www.wired.com/story/deepminds-losses-future-artificial-intelligence, Aug. 14. 2019.
  5. Gary Marcus. (2018). Deep Learning: A Critical Appraisal, arXIv:1801.00631 [cs], Jan. 2. 2018.
  6. Park, S. J. (2019). Reinforce Effect of Lower Bracing on Dynamic Behavior Characteristic in Railway Plate Girder Bridges, Master Thesis, Inha University, pp. 17-18.
  7. Ng, Andrew. (2013). Machine Learning and AI via Brain simulation, Standford University, Retrieved from http://ai.stanford.edu/-ang/slides/DeepLearning-Mar2013, Mar. 2013.
  8. Micro-Measurements, V. (2007). Strain gage selection: criteria, procedures, recommendations, Tech Note TN-505-6, pp. 4-6
  9. Lee, G. H. (2018). Damage Detection for RC Rahmen Bridge Based on Convolutional Neural Network, Master Thesis, Inha University, pp. 55-56.
  10. Kim, Y. H. (2017). Development and Application of Reliability-based Structural Health Monitoring Algorithm for Existing Bridges, Doctor Thesis, Inha University, pp. 58-59.