DOI QR코드

DOI QR Code

Study on Q-value prediction ahead of tunnel excavation face using recurrent neural network

순환인공신경망을 활용한 터널굴착면 전방 Q값 예측에 관한 연구

  • Hong, Chang-Ho (Div. of Radioactive Waste Disposal Research, KAERI) ;
  • Kim, Jin (Dept. of Civil and Environmental Engineering, KAIST) ;
  • Ryu, Hee-Hwan (Structural & Seismic Tech. Group, KEPCO Research Institute) ;
  • Cho, Gye-Chun (Dept. of Civil and Environmental Engineering, KAIST)
  • 홍창호 (한국원자력연구원 방사성폐기물처분연구부) ;
  • 김진 (한국과학기술원 건설및환경공학과) ;
  • 류희환 (한국전력공사 전력연구원 구조내진그룹) ;
  • 조계춘 (한국과학기술원 건설환경공학과)
  • Received : 2020.04.14
  • Accepted : 2020.05.07
  • Published : 2020.05.31

Abstract

Exact rock classification helps suitable support patterns to be installed. Face mapping is usually conducted to classify the rock mass using RMR (Rock Mass Ration) or Q values. There have been several attempts to predict the grade of rock mass using mechanical data of jumbo drills or probe drills and photographs of excavation surfaces by using deep learning. However, they took long time, or had a limitation that it is impossible to grasp the rock grade in ahead of the tunnel surface. In this study, a method to predict the Q value ahead of excavation surface is developed using recurrent neural network (RNN) technique and it is compared with the Q values from face mapping for verification. Among Q values from over 4,600 tunnel faces, 70% of data was used for learning, and the rests were used for verification. Repeated learnings were performed in different number of learning and number of previous excavation surfaces utilized for learning. The coincidence between the predicted and actual Q values was compared with the root mean square error (RMSE). RMSE value from 600 times repeated learning with 2 prior excavation faces gives a lowest values. The results from this study can vary with the input data sets, the results can help to understand how the past ground conditions affect the future ground conditions and to predict the Q value ahead of the tunnel excavation face.

터널 굴착 시 정확한 암반 분류는 적합한 지보패턴을 설치하는 데 도움을 준다. 암반의 분류를 위해 주로 RMR (Rock Mass Ration)과 Q값을 산정하여 수행되며, 페이스 매핑(face mapping)을 바탕으로 산정된다. 점보드릴 및 프로브드릴의 기계 데이터을 활용하거나 딥러닝을 활용한 굴착면 사진 분석 등의 방법이 암반등급 분류를 예측하기 위해 사용되고 있으나, 분석 시 오랜 시간이 소요되거나, 굴착면 전방의 암반등급을 파악할 수 없다는 점에서 한계를 갖는다. 본 연구에서는 순환인공신경망(Recurrent neural network, RNN)을 활용하여 굴착면 전방의 Q값을 예측하는 방법을 개발하였고 페이스 매핑으로부터 획득한 Q값과 비교/검증하였다. 4,600여개의 굴착면 데이터 중 70%를 학습에 활용하였고, 나머지 30%는 검증에 사용하였다. 학습의 횟수와 학습에 활용한 이전굴착면의 개수를 변경하여 학습을 수행하였다. 예측된 Q값과 실제 Q값의 유사도는 RMSE (root mean square error)를 기준으로 비교하였다. 현재 굴착면과 바로 직전의 굴착면의 Q값을 활용하여 600회 학습하여 예측한 Q값의 RMSE값이 가장 작은 것을 확인하였다. 본 연구의 결과는 학습에 사용한 데이터 값 등이 변화하는 경우 변화할 수 있으나 터널에서의 이전 지반상태가 앞으로의 지반상태에 영향을 미치는 시스템을 이해하고, 이를 통해 터널 굴착면 전방의 Q값의 예측이 가능할 것으로 판단된다.

Keywords

References

  1. Barton, N., Lien, R., Lunde, J. (1974), "Engineering classification of rock masses for the design of tunnel support", Rock Mechanics, Vol. 6, No. 4, pp. 189-236. https://doi.org/10.1007/BF01239496
  2. Bieniawski, Z.T. (1973), "Engineering classification of jointed rock masses", The Civil Engineer in South Africa, Vol. 15, No. 12, pp. 335-343.
  3. Hochreiter, S., Schmidhuber, J. (1997), "Long short-term memory", Neural Computation, Vol. 9, No. 8, pp. 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  4. Jung, J.H., Kim, B.K., Chung, H.Y., Kim, H.M., Lee, I.M. (2019), "A ground condition prediction ahead of tunnel face utilizing time series analysis of shield TBM data in soil tunnel", Journal of Korean Tunnelling and Underground Space Association, Vol. 21, No. 2, pp. 227-242. https://doi.org/10.9711/KTAJ.2019.21.2.227
  5. Kim, H.M., Lee, I.M., Hong, C.H. (2019a), "Effect of RMR and rock type on tunnel drilling speed", Journal of Korean Tunnelling and Underground Space Association, Vol. 21, No. 4, pp. 561-571. https://doi.org/10.9711/KTAJ.2019.21.4.561
  6. Kim, H.Y., Cho, L.H., Kim, K.S. (2019b), "Rock classification prediction in tunnel excavation using CNN", Journal of the Korean Geotechnical Society, Vol. 35, No. 9, pp. 37-45. https://doi.org/10.7843/KGS.2019.35.9.37
  7. Kim, K.S., Kim, J.H., Jeong, L.C., Lee, I.M., Cho, G.C. (2015), "Development for prediction system of TBM tunnel face ahead using probe drilling equipment and drilled hole imaging equipment", Journal of Korean Tunnelling and Underground Space Association, Vol. 17, No. 3, pp. 393-401. https://doi.org/10.9711/KTAJ.2015.17.3.393
  8. Ryu, H.H., Oh, T.M., Cho, G.C., Kim, K.Y., Lee, K.R., Lee, D.S. (2014), "Probabilistic relationship between Q-value and electrical Resistivity", KSCE Journal of Civil Engineering, Vol. 18, No. 3, pp. 780-786. https://doi.org/10.1007/s12205-014-0339-z
  9. Zhang, S., Bamakan, S.M.H., Qu, Q., Li, S. (2018), "Learning for personalized medicine: A comprehensive review from a deep learning perspective", IEEE Reviews in Biomedical Engineering, Vol. 12, pp. 194-208. https://doi.org/10.1109/rbme.2018.2864254
  10. Zhao, H., Chen, Z., Jiang, H., Jing, W., Sun, L., Feng, M. (2019), "Evaluation of three deep learning models for early crop classification using sentinel-1A imagery time series-A case study in Zhanjiang, China", Remote Sensing, Vol. 11, No. 22, pp. 2673. https://doi.org/10.3390/rs11222673