터널과지하공간 (Tunnel and Underground Space)

  • 제30권3호
  • /
  • Pages.214-225
  • /
  • 2020
  • /
  • 1225-1275(pISSN)
  • /
  • 2287-1748(eISSN)
한국암반공학회 (Korean Society for Rock Mechanics and Rock Engineering)
권호 표지
DOI QR코드

DOI QR Code

쉴드 TBM 기계 데이터 및 머신러닝 기법을 이용한 암석의 일축압축강도 예측

Prediction of Uniaxial Compressive Strength of Rock using Shield TBM Machine Data and Machine Learning Technique

  • 투고 : 2020.06.08
  • 심사 : 2020.06.22
  • 발행 : 2020.06.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

쉴드 TBM(Tunnel Boring Machine) 터널 굴착 시 암반의 상태는 굴진 성능을 결정하는 중요한 요소 중 하나이다. 암석 강도는 지반조사 시 실내시험을 통해 얻을 수 있으나, 전체 TBM 굴진 구간에 대해 모두 알 수 없다. TBM 굴진 시 최적 Operation Parameter를 적용하기 위해서는 굴진 속도에 영향을 미치는 암석 강도를 파악하는 것이 매우 중요하다. 이에 본 연구에서는 TBM 굴착 중 생성되는 기계 데이터와 머신러닝(Machine Learning) 기법을 활용하여 암석 강도를 예측하고자 한다. 암석 강도를 예측하기 위해 여러 머신러닝 기법을 사용하여 비교하였고, 가장 예측 성능이 좋은 스태킹 모델을 최종 모델로 선택하였다. 암반 구간 Slurry 쉴드 TBM 굴진 사례에서 지반조사 및 시공 중 조사한 암석 강도와 강도를 획득한 위치에서의 TBM 굴착 데이터를 사용하였다. TBM 굴착 데이터는 Training과 Test용으로 8:2로 분할하였으며, 변수 선택(feature selection), 표준화(scaling), 이상치(outlier) 제거 등 전처리 과정을 수행하였다. 하이퍼파라미터 튜닝까지 마친 후, 스태킹 모델에 대해 평균 제곱근 오차(Root Mean Square Error, RMSE)와 결정 계수(R2)로 모델을 평가한 결과 각각 5.556과 0.943로 나타났으며, TBM 굴착 데이터로 암석 강도를 예측하는 모델로 유용할 것으로 판단된다.

Uniaxial compressive strength (UCS) of rock is one of the important factors to determine the advance speed during shield TBM tunnel excavation. UCS can be obtained through the Geotechnical Data Report (GDR), and it is difficult to measure UCS for all tunneling alignment. Therefore, the purpose of this study is to predict UCS by utilizing TBM machine driving data and machine learning technique. Several machine learning techniques were compared to predict UCS, and it was confirmed the stacking model has the most successful prediction performance. TBM machine data and UCS used in the analysis were obtained from the excavation of rock strata with slurry shield TBMs. The data were divided into 8:2 for training and test and pre-processed including feature selection, scaling, and outlier removal. After completing the hyper-parameter tuning, the stacking model was evaluated with the root-mean-square error (RMSE) and the determination coefficient (R2), and it was found to be 5.556 and 0.943, respectively. Based on the results, the sacking models are considered useful in predicting rock strength with TBM excavation data.

키워드

참고문헌

  1. Barton, N.R., 1999, TBM Performance Estimation in Rock using Q (TBM), Tunnels and Tunnelling International, 31(9), 30-34.
  2. Bruland, A., 1998, Hard Rock Tunnel Boring Vol. 3 - Advance Rate and Cutter Wear, Ph.D. Thesis, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.
  3. Bieniawski, Z.T., Celada, B., Galera J.M., and Alvares M., 2006, Rock Mass Excavability (RME) Index: A New Way to Select the Optimum Tunnel Construction Method. In Proceedings of the ITA World Tunnelling Congress, Seoul, PITA02-254.
  4. Cortes, C. and Vapnik, V., 1995, Support Vector Networks, Machine Learning, 20, 273-297. https://doi.org/10.1007/BF00994018
  5. Gehring, K. 1995. Prognosis of Advance Rates and Wear for Underground Mechanized Excavations, Felsbau, 13(6), 439-448 (in German).
  6. Hamidi, J.K. and Bejari, H., 2013, Rock Mass Classification Systems: Are They Applicable to Prediction of TBM Performance?, Conference: 3rd International Symposium and Exhibition on Underground Excavations for Transportation, Istanbul, Turkey.
  7. Lee, H.L., Song, K.I., and Cho, G.C., 2016, Analysis on Prediction Models of TBM Performance: A Review, Journal of Korean Tunnelling and Underground Space Association, 18(2), 245-256. https://doi.org/10.9711/KTAJ.2016.18.2.245
  8. Macias, F.j., 2016, Hard Rock Tunnel Boring: Performance Predictions and Cutter Life Assessments, Ph.D. Thesis, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.
  9. Rostami, J., 1997, Development of a Force Estimation Model for Rock Fragmentation with Disc Cutters through Theoretical Modeling and Physical Measurement of Crushed Zone Pressure, Ph.D. Thesis, Colorado School of Mines, Golden, Colorado, USA.
  10. Rostami, J. and Ozdemir L., 1993, A New Model for Performance Prediction of Hard Rock TBMs, In Proceedings of Rapid Excavation and Tunneling Conference, Boston, 50, 793-809.
  11. Vapnik, V., Golowich, S., and Smola, A., 1996, Support Method for Function Approximation Regression Estimation and Signal Processing, In Proceedings of the 9th International Conference on Neural Information Processing Systems (NIPS' 96), Cambridge, 281-287.