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

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

DOI QR Code

Development of a window-shifting ANN training method for a quantitative rock classification in unsampled rock zone

미시추 구간의 정량적 지반 등급 분류를 위한 윈도우-쉬프팅 인공 신경망 학습 기법의 개발

  • 신휴성 (한국건설기술연구원 지반연구실) ;
  • 권영철 (한국건설기술연구원 지반연구실)
  • Published : 2009.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

This study proposes a new methodology for quantitative rock classification in unsampled rock zone, which occupies the most of tunnel design area. This methodology is to train an ANN (artificial neural network) by using results from a drilling investigation combined with electric resistivity survey in sampled zone, and then apply the trained ANN to making a prediction of grade of rock classification in unsampled zone. The prediction is made at the center point of a shifting window by using a number of electric resistivity values within the window as input reference information. The ANN training in this study was carried out by the RPROP (Resilient backpropagation) training algorithm and Early-Stopping method for achieving a generalized training. The proposed methodology is then applied to generate a rock grade distribution on a real tunnel site where drilling investigation and resistivity survey were undertaken. The result from the ANN based prediction is compared with one from a conventional kriging method. In the comparison, the proposed ANN method shows a better agreement with the electric resistivity distribution obtained by field survey. And it is also seen that the proposed method produces a more realistic and more understandable rock grade distribution.

본 연구에서는 터널 설계구간의 대부분을 차지하는 미시추 구간의 지반 등급 분류를 정량적으로 수행할 수 있는 새로운 접근방법을 제안한다. 본 제안방법은 시추공에서 얻은 직접조사 결과와 시추구간의 전기 비저항 탐사결과를 이용해 인공 신경망을 학습시카고, 학습된 인공 신경망은 미시추 구간의 암반분류 등급을 추론하는데 적용된다. 지반등급 추론은 미시추 구간 영역에서 움직이는 격자형 창(window)의 중심점에서 이루어 지며 창내 귀속된 전기 비저항들은 추론을 위한 참고자료로 시용된다. 인공 신경망 학습은 최선 RPROP(Resilient backpropagation) 인공 신경망 학습 알고리즘과 early-stopping 기법을 이용하여 수행되었다. 본 연구에서는 실제 시추조사가 이루어진 터널현장에 제안기법을 적용하여 미시추 구간의 지반 등급을 추론하였으며, 전통적인 지구통계학적 크리깅(kriging) 기법에 의한 결과와도 상호 비교하였다. 결과적으로 본 연구를 통해 학습된 인공 신경망은 전통 크리깅 방법에 비해 매우 구체적이고 현실적인 예측결과를 제공하였다 또한, 인공 신경망 추론으로부터 얻어진 터널 종단 방향의 RMR과 Q-값의 분포에서는 전기 비저항 탐사로부터 추정된 취약지반 구간의 위치와 잘 일치하였으며, 두 값 상호간의 관계도 선행 연구 결과와 부합하였다.

Keywords

References

  1. 유광호 (1995a), “경계조건을 고려한 보통크리깅의 지하특성평가를 위한 응용”, 대한토목학회 논문집, 제 19권, 제3 호, pp. 645-652.
  2. 유광호 (1995b), “다분적 암반분류를 위한 정성적 자료의 지구통계학적 연구 - I. 이론", 한국지반공학회 논문집, 제 11 권 2호, pp. 71-77.
  3. 최종근 (2002), 공간정보 모델링, 구미서관, p. 286.
  4. Baecher, G. B. (1983), "Statistical analysis of rock mass fracturing", Journal of Mathematical Geology, Vol. 15, No. 2, pp. 329-347. https://doi.org/10.1007/BF01036074
  5. Bieniawski, Z. T. (1984), Rock mechanics design in mining and tunneling, Balkema, Boston, p. 271.
  6. Jacobs, R. A. (1988), “ Increased rates of convergence through learning rate adaptation", Neural Networks, Vol 1, No. 4, pp. 295-307. https://doi.org/10.1016/0893-6080(88)90003-2
  7. Journel, A. G. (1986), “ Geostatistic: Models and tools for the earth sciences", Mathematical Geology, Vol. 18, No. 1, pp. 119-140. https://doi.org/10.1007/BF00897658
  8. Ohtsu, H., Sakai, Y., Saegusa, H., Onoe, H. , Ijiri, Y. and Motoshima, T. (2006), “Risk Evaluation of Water Inrush During Shaft Excavation in Fractured Rock Masses 275", Proc. of ISRM International Symposium 2006, 4th Asian Rock Mechanics Symposium, CD-ROM
  9. Ozturk, C. A. and Nasuf, E. (2002), "Geostatistical assessment of rock zones for tunneling", Tunneling and Underground Space Technology, Vol. 17, pp. 275-285 https://doi.org/10.1016/S0886-7798(02)00023-8
  10. Pande, G. N. and Shin, H. S. (2004), “Artificial Intelligence v. Equations", Proceedings of Institute of Civil Engineers in Civil Engineering, Vol. 157, No. 1, pp. 39-42.
  11. Pao, Y. H. (1989), Adaptive Pattern Recognition and Neural Networks, Addison-Wesley Publishing Company, USA.
  12. Prechelt, L. (1998), “Automatic early stopping using cross validation: quantifying the criteria", Neural Networks, Vol. 11, pp. 761-767. https://doi.org/10.1016/S0893-6080(98)00010-0
  13. Riedmiller, M. (1994), “Advanced supervised learning in multi-Iayer perceptrons - from backpropagation to adaptive learning algorithms", International Journal of Computer Standards and Interfaces, Vol. 16, pp. 265-278
  14. Rumelhart, D. E., McClelland, J. L. (1986), Parallel distributed processing, Cambridge, MA: Exploitation in the MIT Press.
  15. Shin, H S. (2001), Neural network based material models for finite element analysis, Ph.D. thesis: C/Ph/250/01. Department of Civil Engineering, University of Wales Swansea.
  16. Shin, H. S., Lim, J. J., Chang, S. H and Bae, G. J. (1994), “Assessment of the major causes for tunnel collapses by using a neural network based sensitivity analysis", Proceeding of Annual Conference of Korea Institute of Civil Engineers, pp. 512-517.
  17. Skouras, K., Goutis, C. and Bramson, M. J. (1994), "Estimation in linear-modeIs using gradient descent with early stopping", Statistics and Computing, Vol. 4, No. 4, pp.271-278. https://doi.org/10.1007/BF00156750
  18. Soulie, M. (1984), Geostatistic applications in geotechnics, Advanced geostatistics in the mining industry, D. Reidel Publishing, Holland, pp. 703-730.
  19. Sturk, R., Olsson, L. and Johanson, J. (1996), “ Risk and decision analysis for large underground projects, as applied to the Stockholm ring road tunnels", Tunnelling and Underground Space Technology, Vol. 11, No. 2, pp.156-164
  20. Tollenaere, T. (1990), “ Fast adaptive Backpropagation with good scaling properties", Neural Networks, Vol. 3, No. 5, pp. 561-573. https://doi.org/10.1016/0893-6080(90)90006-7
  21. Webster, R. (1984), Elucidation and characterization of spatial variation in soil using regionalized variable theory, IEEE Trans. Sys., Man, Cybern, pp. 903-914.
  22. You, K H and Lee, J. S. (2006), “ Estimation of rock mass classes using the 3-dimensional multiple indicator kriging technique", Tunnelling and Underground Space Technology, Vol. 21, p. 229. https://doi.org/10.1016/j.tust.2005.12.009