All Issue

2025 Vol.26, Issue 4
Development of Mining Ground Subsidence Prediction Model Using Group-wise Parallel Structure

폐광산 지반침하 예측을 위한 인공신경망 모델링 및 정보군별 분석

Minkoan Kim1*
김 민관1*
1Postdoctoral Researcher, Korea Construction Standards Center, Korea Institute of Civil Engineering and Building Technology
*Corresponding Author
1 April 2025. pp. 5-14
PDF XML Citation
Abstract

This study proposes a multi-layer artificial neural network (MSA: Mines Subsidence Artificial neural network) model to effectively predict ground subsidence in abandoned mines. To account for the differing characteristics of coal and non-coal mines, the input variables were grouped into Rock, Mining, and Subsidence datasets, and prediction accuracy was verified for each group. Analysis of 247 subsidence cases showed that the proposed model, which adopts a two-hidden-layer structure (30-45 nodes), achieved an average inference error rate of 19% and a high correlation coefficient (R2 up to 0.87), indicating its potential for ground stability assessment in abandoned mines. By mine type, coal mines exhibited slightly higher prediction errors than non-coal mines, presumably due to the complex geology and irregular subsidence mechanisms within coal seams. Among the information groups, the Rock group recorded the lowest error rate (8-9%), whereas the Subsidence group showed the highest (17-19%). Furthermore, the relatively low correlation (R2=0.45) for the Subsidence group suggests a need for more detailed classification of that parameter and additional data collection.

Keywords
Ground subsidence
Artificial Neural Network
Abandoned mine
Resilient Propagation Algorithm
Ground stability

본 연구에서는 폐광산 지반침하를 효율적으로 예측하기 위하여 다층 인공신경망(MSA, Mines Subsidence Artificial neural network) 모델을 제안하고, 석탄광과 일반광의 구분에 따른 특성 차이를 반영하기 위해 암반(Rock), 채광(Mining), 침하(Subsidence) 정보군으로 나누어 예측 정확도를 검증하였다. 총 247개 침하 사례 분석 결과, 제안된 모델은 2개의 은닉층(30-45노드) 구조에서 평균 추론 오차율 19%와 높은 상관계수(R2 최대 0.87)를 보여 폐광산 지반안정성 평가에 대한 적용 가능성을 확인하였다. 광종별로는 석탄광이 일반광보다 예측 오차가 다소 크게 나타났는데, 이는 석탄층의 복잡한 지질구조 및 불규칙한 침하 메커니즘에 기인하는 것으로 추정된다. 또한 정보군별로는 암반 정보군에서 가장 낮은 오차율(8-9%), 침하 정보군에서 가장 높은 오차율(17-19%)을 보였으며, 침하정보(R2=0.45)에서 상대적으로 낮은 상관성을 보여 해당 인자의 세분화 및 추가 데이터 확보가 필요한 것으로 판단되었다.

키워드
지반침하
인공신경망
폐광산
복원 전파 알고리즘
지반안정성
References
1

Ahn, J. B. and Yoon, W. J. (2003), Development of 3D visualization program for comparison of core log data and tomography data in subsidence area, Journal of the Korean Society of Exploration Geophysicists, Vol. 6, No. 4, pp. 195~198 (In Korean).

2

Ahn, S. C. and Kim, K. D. (2007), Application of probabilistic methods and sensitivity analysis to mine subsidence risk assessment, Proceedings of the Korean Remote Sensing Society, 2007 Spring Conference, pp. 212~220 (In Korean).

3

Bell, F. G. (2007), Engineering Geology, 2nd Edition, Butterworth-Heinemann, Oxford, pp. 156~198.

4

Bishop, C. M. (2006), Pattern recognition and machine learning, Springer, New York, pp. 225~290.

5

Brady, B. H. G. and Brown, E. T. (1985), Rock mechanics: for underground mining, Kluwer Academic Publishers, Dordrecht, p. 527.

6

Caruana, R., Lawrence, S. and Giles, C. L. (2001), Overfitting in neural networks: backpropagation, conjugate gradient, and early stopping, Proceedings of Neural Information Processing Systems (NIPS), Vol. 13, pp. 402~408.

7

Choi, J. G., Kim, K. D., Lee, S. R., Kim, I. S. and Won, J. S. (2007), Prediction of ground subsidence hazard in abandoned coal mine areas using GIS and probabilistic model, Journal of Resource and Environmental Geology, Vol. 40, No. 3, pp. 295~306 (In Korean).

8

Choi, J. K., Kim, K. D. and Lee, S. R. (2008), Risk assessment of ground subsidence in the Sam-cheok coalfield region, Proceedings of the Korean Geological Society, 2008 Autumn Conference, pp. 321~330 (In Korean).

9

Choi, S. O., Jeon, Y. S., Park, E. S., Jeong, Y. B. and Chun, D. S. (2005), Analysis of subsidence mechanism and development of evaluation program, Journal of Korean Society for Rock Mechanics, Vol. 15, No. 3, pp. 195~212 (In Korean).

10

Chun, B. S., Joo, Y. J., Oh, S. J. and Park, J. H. (1997), Earth reinforcement of underground mining cavern, Proceedings of the Korean Society of Civil Engineers Conference (In Korean).

11

Goel, S. C. and Page, C. H. (1982), An empirical method for predicting the probability of chimney cave occurrence over a mining area, International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts, Vol. 19, No. 6, pp. 325~337.

10.1016/0148-9062(82)91367-5
12

Gowara, K. and Ahn, J. B. (2003), Study on geological structures affecting ground subsidence in mining areas, Journal of Korean Society for Rock Mechanics, Vol. 23, No. 4, pp. 215~229 (In Korean).

13

Han, G. C., Park, E. S. and Choi, J. Y. (2007), Geotechnical characteristics of mine subsidence areas, Korean Society for Geotechnical Engineering, Vol. 22, No. 1, pp. 101~115 (In Korean).

14

Hastie, T., Tibshirani, R. and Friedman, J. (2009), The elements of statistical learning, Springer, New York, pp. 222~241.

10.1007/978-0-387-84858-7
15

Ji, J. B., Kim, J. W., Moon, S. H. and Choi, S. W. (2000), Study on the depth influence of mining voids on ground stability, Korean Society for Rock Mechanics, Vol. 16, No. 2, pp. 75~89 (In Korean).

16

Jung, Y. B., Kim, T. H. and Song, W. K. (2008), Development of mine subsidence risk evaluation methods through analysis of influencing factors, Journal of Korean Earth System Engineering, Vol. 47, No. 3, pp. 324~338 (In Korean).

17

Kim, B. C. (1999), Investigation on the influence of underground water flow on ground subsidence, Journal of Korean Hydrogeology Society, Vol. 13, No. 2, pp. 115~130 (In Korean).

18

Kim, C. Y. (1998), Prediction of ground behavior due to urban tunnel excavation and structural damage assessment, Ph.D. thesis, Hanyang University, Seoul, South Korea (In Korean).

19

Kim, D. H., Choi, C. R., Kim, D. H., Lee, D. H., Lee, B. S. and Je, H. C. (2007), A case study on reinforcement of ground and foundation against subsidence in abandoned mining area, Tunnel and Underground Space, Vol. 17, No. 4, pp. 255~265.

20

Kim, D. K. (1968), Ground subsidence caused by water discharge and piping phenomenon at Eunseong coal mine, Journal of the Korean Institute of Resources Engineering, Vol. 5, No. 1, pp. 26~33 (In Korean).

21

Kim, J. W., Jeon, S. W. and Seo, Y. H. (2004), Experimental study on the effect of underground mined out voids on the stability of surface structures, Journal of Korean Society for Rock Mechanics: Tunnel and Underground Space, Vol. 14, No. 1, pp. 43~53 (In Korean).

22

Kim, J. W., Song, W. K. and Jang, K. T. (2007), Experimental study on the effect of underground mine tunnels on ground subsidence, Journal of the Korean Society of Mineral and Energy Resources Engineers, Vol. 44, No. 3, pp. 208~217 (In Korean).

23

Kim, J. Y. (2006), Analysis of subsidence hazard factors in abandoned mine areas, Ph.D. thesis, Hanyang University, Seoul, South Korea (In Korean).

24

Kim, Y. K., Park, H. J., Lee, S. J. and Jung, C. H. (2017), The Development of Subsidence Hazard Index for an Abandoned Mine Using Artificial Neural Network and Interaction Matrix, Journal of the Korean Society of Mineral and Energy Resources Engineers, Vol. 54, No. 5, pp. 480~490 (In Korean).

10.12972/ksmer.2017.54.5.480
25

Kwon, K. S. and Park, Y. J. (2001), Stability assessment of building foundation over abandoned mines, Tunnel and Underground Space, Vol. 11, No. 2, pp. 174~181.

26

Kwon, K. S., Chun, B. S. and Kim, B. H. (1994), Mechanical properties of rock mass and its relationship with ground subsidence, Korean Geotechnical Journal, Vol. 10, No. 2, pp. 45~58 (In Korean).

27

Lee, D. G., Cho, Y. D., Jung, Y. B. and Song, W. K. (2008), Analysis of domestic ground subsidence characteristics through case study, Journal of the Korea Institute of Geoscience and Mineral Resources, Vol. 12, No. 1, pp. 1~10 (In Korean).

28

Lee, H. J., Kim, D. H. and Ryu, D. W. (2007), Evaluation of rock mass swelling characteristics in subsidence prone areas, Korean Rock Mechanics Journal, Vol. 21, No. 5, pp. 375~390 (In Korean).

29

Ministry of Trade, Industry, and Energy (MOTIE). (2022), Basic plan for mine reclamation prevention, Korea Mine Reclamation Corporation, Seoul, pp. 1~85 (In Korean).

30

Moon, S. H., Choi, S. W. and Kim, J. Y. (2004), Numerical analysis of subsidence behavior in mining areas, Journal of Rock Engineering, Vol. 18, No. 3, pp. 278~290 (In Korean).

31

National Coal Board (NCB). (1975), Subsidence Engineers' Handbook, Mining Department, London, p. 111.

32

Park, E. S., Chun, D. S., Jung, Y. B., Shin, H. S. and Choi, S. W. (2007), Estimation of key variables related to subsidence and determination of weights by numerical analysis, Proceedings of the Korean Society for Rock Mechanics Conference, Korean Society for Rock Mechanics and Rock Engineering (In Korean).

33

Peng, S. S. (1992), Surface subsidence engineering, Society for Mining, Metallurgy and Exploration, Littleton, Colorado, USA, pp. 161~195.

34

Piggott, R. J. and Eynon, P. (1977), Ground movements arising from the presence of shallow abandoned mine workings, Proceedings of Conference on Large Ground Movements and Structures, Cardiff, pp. 749~780.

35

Prechelt, L. (1998), Automatic early stopping using cross validation: quantifying the criteria, Neural Networks, Vol. 11, pp. 761~767.

10.1016/S0893-6080(98)00010-012662814
36

Riedmiller, M. and Braun, H. (1993), A direct adaptive method for faster back-propagation learning: the RPROP algorithm, Proceedings of the IEEE International Conference on Neural Networks, San Francisco, pp. 586~591.

37

Rumelhart, D. E. and McClelland, J. L., PDP Research Group. (1986), Parallel distributed processing: explorations in the microstructure of cognition, MIT Press, Cambridge, pp. 318~362.

10.7551/mitpress/5236.001.0001
38

Ryu, D. W., Shin, J. H., Song, W. K., Kim, T. G. and Park, J. Y. (2007), Subsidence hazard assessment method using diffusion equation and its application, Journal of the Korean Society for Rock Mechanics: Tunnel and Underground Space, Vol. 17, No. 5, pp. 372~380 (In Korean).

39

Shin, H. S. (2001), Neural network based constitutive models for finite element analysis, Ph.D. thesis, University of Wales Swansea, Swansea, p. 250.

40

Shin, H. S., Lee, S. S., Kim, K. Y., Bae, G. J. (2013), Quantitative evaluation of collapse hazard levels of tunnel faces by interlinked consideration of face mapping, design and construction data: focused on adaptive weights, Journal of Korean Tunnelling and Underground Space Association, Vol. 15, No. 5, pp. 505~522 (In Korean).

10.9711/KTAJ.2013.15.5.505
41

Seo, J. W., Choi, Y. S., Park, H. D., Kwon, H. H., Yoon, S. H. and Go, W. R.(2010), Application of frequency ratio model and analytic hierarchy process for regional ground subsidence hazard assessment in abandoned mine areas, Journal of the Korean Society of Mineral and Energy Resources Engineers, Vol. 47, No. 5, pp. 690~704 (In Korean).

42

Song, J. H., Chae, H. Y. and Chun, B. S. (2012), Field applicability of tunnel construction using artificial neural networks, Journal of the Korean Geo-Environmental Society, Vol. 13, No. 8, pp. 25~33 (In Korean).

43

Yang, I., Lee, S. (2017), Current status and major achievements of technology development for prevention of mine subsidence and ground settlement, Tunnel and Underground Space, Vol. 27, No. 6, pp. 357~365 (In Korean).

Information
  • Publisher :Korean Geo-Environmental Society
  • Publisher(Ko) :한국지반환경공학회
  • Journal Title :Journal of the Korean Geo-Environmental Society
  • Journal Title(Ko) :한국지반환경공학회 논문집
  • Volume : 26
  • No :4
  • Pages :5-14
  • Received Date : 2025-02-10
  • Revised Date : 2025-02-11
  • Accepted Date : 2025-03-07
  • DOI :https://doi.org/10.14481/jkges.2025.26.4.5
Journal Informaiton Journal of the Korean Geo-Environmental Society Journal of the Korean Geo-Environmental Society
Online Submission
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • orcid
Journal Informaiton Journal Informaiton - close