터널과지하공간 (Tunnel and Underground Space)

  • 제33권6호
  • /
  • Pages.594-609
  • /
  • 2023
  • /
  • 1225-1275(pISSN)
  • /
  • 2287-1748(eISSN)
한국암반공학회 (Korean Society for Rock Mechanics and Rock Engineering)
권호 표지
DOI QR코드

DOI QR Code

커터수명지수 예측을 위한 다중선형회귀분석과 트리 기반 머신러닝 기법 적용

Application of Multiple Linear Regression Analysis and Tree-Based Machine Learning Techniques for Cutter Life Index(CLI) Prediction

  • 홍주표 (강원대학교 강원대학교 에너지.인프라 융합학과) ;
  • 고태영 (강원대학교 에너지자원・산업공학부)
  • Ju-Pyo Hong (Department of Integrated Energy and Infra System, Kangwon National University) ;
  • Tae Young Ko (Department of Energy and Resources Engineering, Kangwon National University)
  • 투고 : 2023.12.11
  • 심사 : 2023.12.18
  • 발행 : 2023.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

TBM 공법은 굴착면 안정성 확보 및 주변환경에 비치는 영향을 최소화하기 때문에 도심지나 하·해저터널 등에서 적용 사례가 증가하는 추세이다. 디스크 커터의 수명을 예측하는 대표적인 모델 중 NTNU모델은 커터수명지수(Cutter Life Index, CLI)를 주요 매개 변수로 활용하지만 복잡한 시험절차와 시험장비의 희귀성으로 측정에 어려움이 있다. 본 연구에서는 다중선형회귀분석과 트리 기반의 머신러닝 기법으로 암석물성을 활용하여 CLI를 예측하였다. 문헌 조사를 통해 암석의 일축압축강도, 압열인장강도, 등 가석영함량과 세르샤 마모지수 등을 포함한 데이터베이스를 구축하였고 파생변수를 계산하여 추가하였다. 다중선형회귀분석은 통계적 유의성과 다중공선성을 고려하여 입력 변수를 선정하였고 머신러닝 예측 모델은 변수 중요도를 기반으로 입력 변수를 선정하였다. 학습용과 검증용 데이터를 8:2로 나누어 모델 간 예측 성능을 비교한 결과 XGBoost가 최적의 모델로 선정되었다. 본 연구에서 도출된 다중선형회귀모델과 XGBoost모델을 선행 연구와 예측 성능을 비교하여 타당성을 확인하였다.

TBM (Tunnel Boring Machine) method is gaining popularity in urban and underwater tunneling projects due to its ability to ensure excavation face stability and minimize environmental impact. Among the prominent models for predicting disc cutter life, the NTNU model uses the Cutter Life Index(CLI) as a key parameter, but the complexity of testing procedures and rarity of equipment make measurement challenging. In this study, CLI was predicted using multiple linear regression analysis and tree-based machine learning techniques, utilizing rock properties. Through literature review, a database including rock uniaxial compressive strength, Brazilian tensile strength, equivalent quartz content, and Cerchar abrasivity index was built, and derived variables were added. The multiple linear regression analysis selected input variables based on statistical significance and multicollinearity, while the machine learning prediction model chose variables based on their importance. Dividing the data into 80% for training and 20% for testing, a comparative analysis of the predictive performance was conducted, and XGBoost was identified as the optimal model. The validity of the multiple linear regression and XGBoost models derived in this study was confirmed by comparing their predictive performance with prior research.

키워드

과제정보

이 논문은 2023년도 정부 (산업통상자원부)의 재원으로 해외자원개발협회의 지원(2021060003, 스마트 마이닝 전문 인력 양성)과 2023년도 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임(No.NRF-2022R1F1A1063228)

참고문헌

  1. Aligholi, S., Lashkaripour, G.R., Ghafoori, M., and Azali, S.T., 2017, Evaluating the relationships between NTNU/SINTEF drillability indices with index properties and petrographic data of hard igneous rocks, Rock Mech. Rock Eng., 50, 2929-2953.  https://doi.org/10.1007/s00603-017-1289-9
  2. Altindag, R., 2003, Correlation of specific energy with rock brittleness concepts on rock cutting, J. South. Afr. Inst. Min. Metall., 103(3), 163-171. 
  3. Altindag, R., 2010, Assessment of some brittleness indexes in rock-drilling efficiency, Rock Mech. Rock Eng., 43, 361-370.  https://doi.org/10.1007/s00603-009-0057-x
  4. Bruland, A., 2000, Hard rock tunnel boring: vol. 8 drillability test methods, Ph. D. Thesis, Norwegian University of Science and Technology. 
  5. Chang, S.H., Choi, S.W., Lee, G.P., and Bae, G.J., 2011, Statistical analysis of NTNU test results to predict rock TBM performance, J. of Korean Tunn Undergr Sp. Assoc., 13(3), 243-260. 
  6. Chen, T. and Guestrin, C., 2016, Xgboost: A scalable tree boosting system, Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., 785-794. 
  7. Cutler, A., Cutler, D.R., and Stevens, J. R., 2012, Random forests, Ensemble machine learning: Methods and applications, 157-175. 
  8. Dahl, F., Bruland, A., Jakobsen, P.D., Nilsen, B., and Grov E., 2012, Classifications of properties influencing the drillability of rocks, Tunn. Undergr. Space Technol., 28, 150-158.  https://doi.org/10.1016/j.tust.2011.10.006
  9. Drucker, H., 1997, Improving regressors using boosting techniques, In icml, 97, 107-115. 
  10. Eide, L.N.R., 2014, TBM tunnelling at the Stillwater Mine, MS Thesis, NTNU. 
  11. Gehring, K., 1995, Leistungs-und Verschleissprognosen im maschinellen Tunnelbau, Felsbau, 13(6), 439-448. 
  12. Geurts, P., Ernst, D., and Wehenkel, L., 2006, Extremely randomized trees, Machine learning, 63, 3-42.  https://doi.org/10.1007/s10994-006-6226-1
  13. Hucka, V. and Das, B., 1974, Brittleness determination of rocks by different methods, Int. j. rock mech. min. sci. geomech. abstr., 11(10), 389-392.  https://doi.org/10.1016/0148-9062(74)91109-7
  14. Ko, T.Y., Kim, T.K., Son, Y., and Jeon, S., 2016, Effect of geomechanical properties on Cerchar Abrasivity Index (CAI) and its application to TBM tunnelling, Tunn. Undergr. Space Technol., 57, 99-111.  https://doi.org/10.1016/j.tust.2016.02.006
  15. Kuhn, M. and Johnson, K., 2013, Applied predictive modeling, New York: Springer, 26, 13. 
  16. Macias, F.J. Dahl, F., and Bruland, A., 2016, New rock abrasivity test method for tool life assessments on hard rock tunnel boring: the rolling indentation abrasion test (RIAT), Rock Mech. Rock Eng., 49(5), 1679-1693.  https://doi.org/10.1007/s00603-015-0854-3
  17. Macias, F.J., 2016, Hard rock tunnel boring: performance predictions and cutter life assessments, Ph. D. Thesis, Norwegian University of Science and Technology. 
  18. Majeed, Y. and Abu Bakar, M.Z., 2019, Effects of variation in the particle size of the rock abrasion powder and standard rotational speed on the NTNU/SINTEF abrasion value steel test, Bull. Eng. Geol. Environ., 78, 1537-1554.  https://doi.org/10.1007/s10064-017-1211-4
  19. Majeed, Y., Abu Bakar, M.Z., and Butt, I.A., 2020, Abrasivity evaluation for wear prediction of button drill bits using geotechnical rock properties, Bull. Eng. Geol. Environ., 79, 767-787.  https://doi.org/10.1007/s10064-019-01587-y
  20. Massalov, T., Yagiz, S., and Adoko, A.C., 2022, Application of soft computing techniques to estimate cutter life index using mechanical properties of rocks, Appl. Sci., 12(3), 1446. 
  21. Meng, F., Wong, L.N.Y., and Zhou, H., 2021, Rock brittleness indices and their applications to different fields of rock engineering: A review, J. Rock Mech. Geotech. Eng., 13(1), 221-247.  https://doi.org/10.1016/j.jrmge.2020.06.008
  22. Moon, K. and Yang, S.B., 2020, Cohesion and internal friction angle estimated from Brazilian tensile strength and unconfined compressive strength of volcanic rocks in Jeju Island, Journal of the Korean Geotechnical Society, 36(2), 17-28.  https://doi.org/10.7843/KGS.2020.36.2.17
  23. Natekin, A. and Knoll, A., 2013, Gradient boosting machines, a tutorial, Frontiers in neurorobotics, 7, 21. 
  24. Ozfirat, M.K., Yenice, H., Simsir, F., and Yarali, O., 2016, A new approach to rock brittleness and its usability at prediction of drillability, J. Afr. Earth Sci., 119, 94-101.  https://doi.org/10.1016/j.jafrearsci.2016.03.017
  25. Piratheepan, J., Gnanendran, C.T., and Arulrajah, A., 2012, Determination of c and φ from IDT and unconfined compression testing and numerical analysis, J. Mater. Civ. Eng., 24(9), 1153-1164.  https://doi.org/10.1061/(ASCE)MT.1943-5533.0000493
  26. Plinninger, R.J., 2002, Klassifizierung und Prognose von Werkzeugverschleiss bei konventionellen Gebirgslosungsverfahren im Festgestein. 
  27. Rostami, J. and Ozdemir, L., 1993, A new model for performance prediction of hard rock TBMs, Proc. Rapid Excav. Tunneling Conf., 793-809.