DOI QR코드

DOI QR Code

Prediction of replacement period of shield TBM disc cutter using SVM

SVM 기법을 이용한 쉴드 TBM 디스크 커터 교환 주기 예측

  • La, You-Sung (Dept. of Civil and Environmental Engineering, Dongguk University) ;
  • Kim, Myung-In (Dept. of Civil and Environmental Engineering, Dongguk University) ;
  • Kim, Bumjoo (Dept. of Civil and Environmental Engineering, Dongguk University)
  • 나유성 (동국대학교 건설환경공학과) ;
  • 김명인 (동국대학교 건설환경공학과) ;
  • 김범주 (동국대학교 건설환경공학과)
  • Received : 2019.06.13
  • Accepted : 2019.07.19
  • Published : 2019.09.30

Abstract

In this study, a machine learning method was proposed to use in predicting optimal replacement period of shield TBM (Tunnel Boring Machine) disc cutter. To do this, a large dataset of ground condition, disc cutter replacement records and TBM excavation-related data, collected from a shield TBM tunnel site in Korea, was built and they were used to construct a disc cutter replacement period prediction model using a machine learning algorithm, SVM (Support Vector Machine) and to assess the performance of the model. The results showed that the performance of RBF (Radial Basis Function) SVM is the best among a total of three SVM classification functions (80% accuracy and 10% error rate on average). When compared between ground types, the more disc cutter replacement data existed, the better prediction results were obtained. From this results, it is expected that machine learning methods become very popularly used in practice in near future as more data is accumulated and the machine learning models continue to be fine-tuned.

본 연구에서는 쉴드 TBM (Tunnel Boring Machine) 터널 디스크 커터의 적절한 교체 시기를 예측하기 위한 방법으로 머신러닝 기법을 사용한 방법을 제안하였으며, 이를 위해 국내 기 시공된 쉴드 TBM 현장의 데이터를 이용하여 다양한 머신러닝 알고리즘 중 SVM (Support Vector Machine)을 이용하여 예측 모델을 구축하고 그 성능을 평가하였다. 지반 조건별 디스크 커터의 마모와 높은 상관성을 갖는 TBM 기계 데이터와 디스크 커터 교체 이력을 분류하고, 이들을 SVM의 변수로 사용하여 3종류의 분류 함수를 적용하여 각각 학습을 한 후 예측을 수행한 결과, 각 지반 조건에 대해서 3종류의 SVM 분류 함수 중 전체적으로 RBF (Radial Basis Function) SVM의 예측성능이 가장 우수하며(평균적으로 80%의 정확도, 10% 오분류율), 지반 조건별로 구분 시 디스크 커터 교체 데이터의 수가 많을수록 예측 결과가 좋은 것으로 나타났다. 향후 많은 데이터를 축적하고 이를 모두 활용하여 학습모델을 지속적으로 발전시켜 나간다면 이와 같은 디스크 커터 교환주기를 예측하기 위한 머신러닝 기법의 실무 적용성이 매우 클 것으로 기대한다.

Keywords

References

  1. Bieniawski, Z.T., Celada, B., Galera, J.M., Tardaguila, I. (2009), "Prediction of cutter wear using RME", Proceedings of the ITA-AITES 2009 World Tunnel Congress and 35th ITA General Assembly, Budapest.
  2. Bruland, A. (2000), Hard rock tunnel boring: Background and discussion, Ph.D. Thesis, Norwegian University of Science and Technology, pp. 14-31.
  3. Burges, C.J.C. (1998), "A tutorial on support vector machines for pattern recognition", Data Mining and Knowledge Discovery, Vol. 2, No. 2, pp. 121-167. https://doi.org/10.1023/A:1009715923555
  4. Cortes, C., Vapnik, V. (1995), "Support-vector networks", Machine Learning, Vol. 20, No. 3, pp. 273-297. https://doi.org/10.1007/BF00994018
  5. Gehring, K. (1995), "Prognosis of advance rates and wear for underground mechanized excavations (in German)", Felsbau, Vol. 13, No. 6, pp. 439-448.
  6. Jeong, H.Y., Lee, S.D., Jeon, S.W. (2014), "Estimation of design parameters of TBM using punch penetration and Cerchar abrasiveness test", Journal of Korean Tunnelling and Underground Space Association, Vol. 16, No. 2, pp. 237-248. https://doi.org/10.9711/KTAJ.2014.16.2.237
  7. Kim, D.Y., Farrokh, E., Jung, J.H., Lee, J.W., Lee, S.H. (2017), "Development of a new test method for the prediction of TBM disc cutters life", Journal of Korean Tunnelling and Underground Space Association, Vol. 19, No. 3, pp. 475-488. https://doi.org/10.9711/KTAJ.2017.19.3.475
  8. Ko, T.Y., Yoon, H.J., Son, Y.J. (2014), "A comparative study on the TBM disc cutter wear prediction model", Journal of Korean Tunnelling and Underground Space Association, Vol. 16, No. 6, pp. 533-542. https://doi.org/10.9711/KTAJ.2014.16.6.533
  9. La, Y.S., Kim, M.I., Kim, B. (2019), "Development of penetration rate prediction model using shield TBM excavation data", Journal of Korean Tunnelling and Underground Space Association, Vol. 21, No. 4, pp. 519-534. https://doi.org/10.9711/KTAJ.2019.21.4.519
  10. Lee, J.S., Choi, I.Y., Kim, I.K., Hwang, S.H. (2018), "Tamping and renewal optimization of ballasted track using track measurement data and genetic algorithm", Journal of Transportation Engineering, Part A: systems, Vol. 144, No. 3, pp. 04017081-1:8. https://doi.org/10.1061/JTEPBS.0000120
  11. Rostami, J., Ozdemir, L. (1993), "A new model for performance prediction of hard rock TBMs", Proceedings of the Rapid Excavation and Tunneling Conference, Boston, pp. 793-809.
  12. Smola, A.J., Scholkopf, B. (2004), "A tutorial on support vector regression", Statistics and Computing, Vol. 14, No. 3, pp. 199-222. https://doi.org/10.1023/B:STCO.0000035301.49549.88
  13. Vapnik, V.N. (1995), The Nature of Statistical Learning Theory, Springer, New York, pp. 119-166.
  14. Yu, S.H. (2007), A study on rock cutting behavior by TBM disc cutter, Master's Thesis, Seoul National University, pp. 21-74.