기계학습 기반의 교량 위험성 모니터링 시스템 모델
Machine Learning-based Bridge Safety Monitoring System Model
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
교량에서의 신축이음 장치는 교량의 온도변화, 활하중, 지진, 콘크리트 재령에 따른 건조수축에 의한 상부고조의 이동과 회전거동을 원활하게 수용함으로써 내구성과 안전성을 보장하기 위한 장치이다. 그러나 다양한 원인으로 인해 신축이음 장치가 파손됨으로써 사고가 동반되고 있다. 일반적으로 교량은 정기적인 검사를 통하여 유지관리를 하고 있으나, 대부분 육안 검사에 의한 검사를 진행하고 있는 상황이다. 제안 모델은 교량의 교좌장치의 변위를 측정한 데이터를 수집하여 인공지능 기반의 교량 모니터링 시스템이다. 제안 모델의 실험을 위하여 모의교량과 모의 차량을 제작하였으며, 정상적인 상태와 비정상적인 상태의 데이터를 수집하고, 수집된 데이터를 정제하여 기계학습에 사용하였다. 모의 교량 데이터를 활용한 인공지능 모델의 성능 분석 및 선정은 모의 교량 테스트 데이터 20%, 학습 데이터 80%를 사용하여 XGBoost로 예측 성능 점수 평가를 진행하였고, 신축이음부의 단차 발생과 파손 발생 예측 정확도는 평균 96.2%의 결과가 도출되었다. 그러나 이러한 결과는 모의 교량을 대상으로 실험한 것으로, 실제 교량의 데이터를 이용한 결과가 아니라는 한계점을 가지고 있다. 실제 교량에서 발생하는 다양한 변수들의 데이터를 수집하여 적용한다면 신축이음 장치의 파손으로 인한 사고를 사전에 예방할 수 있을 것이다.
Abstract
Expansion joint devices in bridges are devices that ensure durability and safety by smoothly accommodating the movement and rotation of the upper deck due to temperature changes, live loads, earthquakes, and drying shrinkage due to the concrete age of the bridge. However, accidents occur as expansion joints are damaged due to various reasons. In general, bridges are maintained through regular inspections, but most inspections are conducted through visual inspection. The proposed model is an artificial intelligence-based bridge expansion joints monitoring system that collects data measuring the displacement of the bridge's suspension device. To test the proposed model, a simulated bridge and a simulated vehicle were manufactured, data in normal and abnormal states were collected, and the collected data was refined and used for machine learning. For the performance analysis and selection of the artificial intelligence model using simulated bridge data, 80% of the simulated bridge learning data and 20% of the test data were used to evaluate the predicted performance score using XGBoost. The average accuracy of predicting the occurrence of steps and damage in expansion joints was 96.2%. However, these results were tested on a simulated bridge, and have the limitation that the results are based on data from an actual bridge. It is expected that if data on various variables that occur in actual bridges are collected and applied, accidents due to damage to expansion joints can be prevented in advance.
참고문헌 (15)
- 나용수, "프리캐스트 콘크리트를 이용한 신축이음장치의 한강상 교량에서의 시공성 평가", 한국산학기술학회논문지, 23(3), pp.449~457 (2022), 한국산학기술학회[DOI]
- R. B. Malla, "Development and laboratory analysis of silicone foam sealant for bridge expansion joints", Journal of Bridge Engineering, 12(4), pp.438~448
- E. McCarthy, "Development of an experimentally validated analytical model for modular bridge expansion joint behavior", Journal of Bridge Engineering, 19(2), pp.235~244 (2014)
- J. S. Park, "Expansion joints risk prediction system based on IoT displacement device", Electronics, 12(12), pp.2713~ (2023)
- V. Gattulli, "Condition assessment by visual inspection for a bridge management system", Computer-Aided Civil and Infrastructure Engineering, 20(2), pp.95~107 (2005)
- C. Liu, "Bridge damage detection using spatiotemporal patterns extracted from dense sensor network", Measurement Science and Technology, 28(1), pp.1~21 (2016)
- Y. An, "Recent progress and future trends on damage identification methods for bridge structures", Structural Control and Health Monitoring, 26(10), pp.1~30 (2019)
- J. M. Brownjohn, "Structural health monitoring of civil infrastructure. Philosophical Transactions of the Royal Society A : Mathematical", Physical and Engineering Sciences, 365(1851), pp.589~622 (2007)
- F. Chen, "A feature selection approach for network intrusion detection based on tree-seed algorithm and k-nearest neighbor", 2018 IEEE 4th international symposium on wireless systems within the international conferences on intelligent data acquisition and advanced computing systems (IDAACS-SWS), pp.68~72 (2018)
- Alex Shenfield, "Intelligent intrusion detection systems using artificial neural networks", ICT Express, 4(2), pp.95~99 (2018), 한국통신학회
- J. Gu, "A novel approach to intrusion detection using SVM ensemble with feature augmentation", Computers & Security, 86, pp.53~62 (2019)
- S. Mukherjee, "Intrusion detection using naive Bayes classifier with feature reduction", Procedia Technology, 4, pp.119~128 (2012)
- X. Ma, "Study on a prediction of P2P network loan default based on the machine learning LightGBM and XGBoost algorithms according to different high dimensional data cleaning", Electron. Commerce Res. Appl, 31, pp.24~39 (2018)
- H. Jiang, "Network intrusion detection based on PSO-XGBoost model", IEEE Access, 8, pp.58392~58401 (2020)
- J. H. Mo, "Developing window behavior models for residential buildings using XGBoost algorithm", Energy and Buildings, 205, pp.1~23 (2019)