Journal of Korean Tunnelling and Underground Space Association (한국터널지하공간학회 논문집)

Korean Tunnelling and Underground Space Association (한국터널지하공간학회)
권호 표지
DOI QR코드

DOI QR Code

Enhancing machine learning-based anomaly detection for TBM penetration rate with imbalanced data manipulation

불균형 데이터 처리를 통한 머신러닝 기반 TBM 굴진율 이상탐지 개선

  • Kibeom Kwon (Dept. of Civil, Environmental and Architectural Engineering, Korea University) ;
  • Byeonghyun Hwang (Dept. of Civil, Environmental and Architectural Engineering, Korea University) ;
  • Hyeontae Park (Dept. of Civil, Environmental and Architectural Engineering, Korea University) ;
  • Ju-Young Oh (Construction Technology Team, Samsung C&T) ;
  • Hangseok Choi (School of Civil, Environmental and Architectural Engineering, Korea University)
  • 권기범 (고려대학교 건축사회환경공학과) ;
  • 황병현 (고려대학교 건축사회환경공학과) ;
  • 박현태 (고려대학교 건축사회환경공학과) ;
  • 오주영 (삼성물산 건설부문 기반기술팀) ;
  • 최항석 (고려대학교 건축사회환경공학부)
  • Received : 2024.08.05
  • Accepted : 2024.08.26
  • Published : 2024.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Anomaly detection for the penetration rate of tunnel boring machines (TBMs) is crucial for effective risk management in TBM tunnel projects. However, previous machine learning models for predicting the penetration rate have struggled with imbalanced data between normal and abnormal penetration rates. This study aims to enhance the performance of machine learning-based anomaly detection for the penetration rate by utilizing a data augmentation technique to address this data imbalance. Initially, six input features were selected through correlation analysis. The lowest and highest 10% of the penetration rates were designated as abnormal classes, while the remaining penetration rates were categorized as a normal class. Two prediction models were developed, each trained on an original training set and an oversampled training set constructed using SMOTE (synthetic minority oversampling technique): an XGB (extreme gradient boosting) model and an XGB-SMOTE model. The prediction results showed that the XGB model performed poorly for the abnormal classes, despite performing well for the normal class. In contrast, the XGB-SMOTE model consistently exhibited superior performance across all classes. These findings can be attributed to the data augmentation for the abnormal penetration rates using SMOTE, which enhances the model's ability to learn patterns between geological and operational factors that contribute to abnormal penetration rates. Consequently, this study demonstrates the effectiveness of employing data augmentation to manage imbalanced data in anomaly detection for TBM penetration rates.

TBM (tunnel boring machine) 터널 프로젝트의 리스크 관리 측면에서 굴진율 예측은 중요하며, 이를 위한 머신러닝 기반 TBM 굴진율 예측 연구가 지속적으로 진행되어 왔다. 그러나, 기존 연구의 머신러닝 예측 모델은 정상 굴진율과 이상 굴진율 간의 불균형 데이터를 고려하는 데 한계가 있다. 본 연구에서는 데이터 증강 기법을 통해 불균형 데이터를 처리하여 머신러닝 기반 TBM 굴진율 이상탐지 성능을 개선하였다. 먼저, 상관관계 분석을 통해 유사 변수를 제거하여 6가지 입력특성을 선정하였다. 또한, 하위 10%와 상위 10%의 굴진율을 각각 이상 등급으로, 그 외 범위의 굴진율을 정상 등급으로 굴진율 등급을 구분하였다. 기존 학습 데이터와 SMOTE (synthetic minority oversampling technique)를 통해 증강된 학습 데이터를 각각 XGB (extreme gradient boosting)에 적용한 XGB 모델과 XGB-SMOTE 모델을 구축하였다. 굴진율 등급 예측 성능을 비교한 결과, XGB 모델은 정상 굴진율에 대한 예측 성능은 우수하나 이상 굴진율 예측 성능은 상대적으로 낮게 도출되었다. 반면, XGB-SMOTE 모델은 모든 굴진율 등급에서 일관되게 우수한 예측 성능을 보였다. 이는 SMOTE를 통한 이상 굴진율 데이터의 증강이 이상 굴진율을 유발하는 지반조건과 TBM 운영인자 간의 패턴 학습 수준을 향상시켰기 때문으로 판단된다. 결론적으로, 본 연구는 머신러닝 기반 TBM 굴진율 이상탐지 시 데이터 증강 기법을 활용한 불균형 데이터 처리가 효과적임을 보여준다.

Keywords

Acknowledgement

본 연구는 국토교통과학기술진흥원의 건설기술연구사업(No. RS-2022-00144188)의 지원으로 수행되었으며 이에 깊은 감사를 드립니다.

References

  1. Agrawal, A.K., Murthy, V.M.S.R., Chattopadhyaya, S., Raina, A.K. (2022), "Prediction of TBM disc cutter wear and penetration rate in tunneling through hard and abrasive rock using multi-layer shallow neural network and response surface methods", Rock Mechanics and Rock Engineering, Vol. 55, No. 6, pp. 3489-3506.
  2. Armaghani, D.J., Mohamad, E.T., Narayanasamy, M.S., Narita, N., Yagiz, S. (2017), "Development of hybrid intelligent models for predicting TBM penetration rate in hard rock condition", Tunnelling and Underground Space Technology, Vol. 63, pp. 29-43.
  3. Bae, S., Ham, S., Lee, I., Lee, G.P., Kim, D. (2022), "Deep learning based crack detection from tunnel cement concrete lining", Journal of Korean Tunnelling and Underground Space Association, Vol. 24, No. 6, pp. 583-598.
  4. Benardos, A.G., Kaliampakos, D.C. (2004), "Modelling TBM performance with artificial neural networks", Tunnelling and Underground Space Technology, Vol. 19, No. 6, pp. 597-605.
  5. Broere, W. (2016), "Urban underground space: solving the problems of today's cities", Tunnelling and Underground Space Technology, Vol. 55, pp. 245-248.
  6. Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P. (2002), "SMOTE: Synthetic minority oversampling technique", Journal of Artificial Intelligence Research, Vol. 16, pp. 321-357.
  7. Chen, J., Huang, H., Cohn, A.G., Zhang, D., Zhou, M. (2022), "Machine learning-based classification of rock discontinuity trace: SMOTE oversampling integrated with GBT ensemble learning", International Journal of Mining Science and Technology, Vol. 32, No. 2, pp. 309-322.
  8. Chen, T., Guestrin, C. (2016), "Xgboost: A scalable tree boosting system", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, USA, pp. 785-794.
  9. Farrokh, E. (2018), "Introducing hard rock TBMs' downtime analysis model with reference to past case histories' data", Journal of Mining and Environment, Vol. 9, No. 2, pp. 457-472.
  10. Frenzel, C., Kasling, H., Thuro, K. (2008), "Factors influencing disc cutter wear", Geomechanics and Tunnelling, Vol. 1, No. 1, pp. 55-60.
  11. Gao, B., Wang, R., Lin, C., Guo, X., Liu, B., Zhang, W. (2021), "TBM penetration rate prediction based on the long short-term memory neural network", Underground Space, Vol. 6, No. 6, pp. 718-731.
  12. Ghorbani, E., Yagiz, S. (2024), "Estimating the penetration rate of tunnel boring machines via gradient boosting algorithms", Engineering Applications of Artificial Intelligence, Vol. 136, 108985.
  13. Gong, Q.M., Zhao, J. (2009), "Development of a rock mass characteristics model for TBM penetration rate prediction", International Journal of Rock Mechanics and Mining Sciences, Vol. 46, No. 1, pp. 8-18.
  14. Grima, M.A., Bruines, P.A., Verhoef, P.N.W. (2000), "Modeling tunnel boring machine performance by neuro-fuzzy methods", Tunnelling and Underground Space Technology, Vol. 15, No. 3, pp. 259-269.
  15. Hyun, K.C., Min, S., Choi, H., Park, J., Lee, I.M. (2015), "Risk analysis using fault-tree analysis (FTA) and analytic hierarchy process (AHP) applicable to shield TBM tunnels", Tunnelling and Underground Space Technology, Vol. 49, pp. 121-129.
  16. Jamshidi, A. (2018), "Prediction of TBM penetration rate from brittleness indexes using multiple regression analysis", Modeling Earth Systems and Environment, Vol. 4, pp. 383-394.
  17. Kim, D., Kwon, K., Pham, K., Oh, J.Y., Choi, H. (2022), "Surface settlement prediction for urban tunneling using machine learning algorithms with Bayesian optimization", Automation in Construction, Vol. 140, 104331.
  18. Kwon, K., Choi, H., Hwang, C., Park, S., Hwang, B. (2024a), "Risk assessment for development of consecutive shield TBM technology", Journal of Korean Tunnelling and Underground Space Association, Vol. 26, No. 4, pp. 303-314.
  19. Kwon, K., Choi, H., Jung, J., Kim, D., Shin, Y.J. (2024b), "Prediction of abnormal TBM disc cutter wear in mixed ground condition using interpretable machine learning with data augmentation", Journal of Rock Mechanics and Geotechnical Engineering.
  20. Kwon, K., Choi, H., Oh, J.Y., Kim, D. (2022), "A study on EPB shield TBM face pressure prediction using machine learning algorithms", Journal of Korean Tunnelling and Underground Space Association, Vol. 24, No. 2, pp. 217-230.
  21. Kwon, K., Choi, H., Pham, K., Kim, S., Bae, A. (2024c), "Influence analysis of pavement distress on international roughness index using machine learning", KSCE Journal of Civil Engineering, pp. 1-12.
  22. Liu, Y., Yu, Z., Chen, C., Han, Y., Yu, B. (2020), "Prediction of protein crotonylation sites through LightGBM classifier based on SMOTE and elastic net", Analytical Biochemistry, Vol. 609, 113903.
  23. Liu, Z., Li, L., Fang, X., Qi, W., Shen, J., Zhou, H., Zhang, Y. (2021), "Hard-rock tunnel lithology prediction with TBM construction big data using a global-attention-mechanism-based LSTM network", Automation in Construction, Vol. 125, 103647.
  24. Mahdevari, S., Shahriar, K., Yagiz, S., Shirazi, M.A. (2014), "A support vector regression model for predicting tunnel boring machine penetration rates", International Journal of Rock Mechanics and Mining Sciences, Vol. 72, pp. 214-229.
  25. Moharrami, S., Bayat, A., AbouRizk, S. (2022), "Modeling microtunnel boring machine penetration rate using a mechanistic approach", Journal of Construction Engineering and Management, Vol. 148, No. 11, 04022128.
  26. Moon, J.S., Kim, H.K., An, J.W., Lee, J.G. (2020), "A study on performance-based evaluation system for NATM tunnels in use: development of evaluation model and validation", Journal of Korean Tunnelling and Underground Space Association, Vol. 22, No. 1, pp. 107-120.
  27. Pham, K., Kim, D., Le, C.V., Won, J. (2023), "Machine learning-based pedotransfer functions to predict soil water characteristics curves", Transportation Geotechnics, Vol. 42, 101052.
  28. Rezapour, M. (2021), "Sentiment classification of skewed shoppers' reviews using machine learning techniques, examining the textual features", Engineering Reports, Vol. 3, No. 1, e12280.
  29. Song, C., Peng, H., Xu, L., Zhao, T., Guo, Z., Chen, W. (2024), "Probabilistic evaluation of cultural soil heritage hazards in China from extremely imbalanced site investigation data using SMOTE-Gaussian process classification", Journal of Cultural Heritage, Vol. 67, pp. 121-133.
  30. Toth, A., Gong, Q., Zhao, J. (2013), "Case studies of TBM tunneling performance in rock-soil interface mixed ground", Tunnelling and Underground Space Technology, Vol. 38, pp. 140-150.
  31. Vergara, I.M., Saroglou, C. (2017), "Prediction of TBM performance in mixed-face ground conditions", Tunnelling and Underground Space Technology, Vol. 69, pp. 116-124.
  32. Xu, H., Zhou, J., Asteris, P.G., Armaghani, D.J., Tahir, M.M. (2019), "Supervised machine learning techniques to the prediction of tunnel boring machine penetration rate", Applied Sciences, Vol. 9, No. 18, 3715.
  33. Xue, Y.D., Luo, W., Chen, L., Dong, H.X., Shu, L.S., Zhao, L. (2023), "An intelligent method for TBM surrounding rock classification based on time series segmentation of rock-machine interaction data", Tunnelling and Underground Space Technology, Vol. 140, 105317.
  34. Yagiz, S., Gokceoglu, C., Sezer, E., Iplikci, S. (2009), "Application of two non-linear prediction tools to the estimation of tunnel boring machine performance", Engineering Applications of Artificial Intelligence, Vol. 22, No. 4-5, pp. 808-814.
  35. Yang, J., Yagiz, S., Liu, Y.J., Laouafa, F. (2022), "Comprehensive evaluation of machine learning algorithms applied to TBM performance prediction", Underground Space, Vol. 7, No. 1, pp. 37-49.
  36. Yu, L., Liu, H. (2004), "Efficient feature selection via analysis of relevance and redundancy", Journal of Machine Learning Research, Vol. 5, 1205-1224.
  37. Zhou, J., Qiu, Y., Armaghani, D.J., Zhang, W., Li, C., Zhu, S., Tarinejad, R. (2021), "Predicting TBM penetration rate in hard rock condition: a comparative study among six XGB-based metaheuristic techniques", Geoscience Frontiers, Vol. 12, No. 3, 101091.