Thickness of excavation damaged zone estimation using four novel hybrid ensemble learning models: A case study of Xiangxi Gold Mine and Fankou Lead-zinc Mine in China

Abstract

Underground excavation can lead to stress redistribution and result in an excavation damaged zone (EDZ), which is an important factor affecting the excavation stability and support design. Accurately estimating the thickness of EDZ is essential to ensure the safety of the underground excavation. In this study, four novel hybrid ensemble learning models were developed by optimizing the extreme gradient boosting (XGBoost) and random forest (RF) algorithms through simulated annealing (SA) and Bayesian optimization (BO) approaches, namely SA-XGBoost, SA-RF, BO-XGBoost and BO-RF models. A total of 210 cases were collected from Xiangxi Gold Mine in Hunan Province and Fankou Lead-zinc Mine in Guangdong Province, China, including seven input indicators: embedding depth, drift span, uniaxial compressive strength of rock, rock mass rating, unit weight of rock, lateral pressure coefficient of roadway and unit consumption of blasting explosive. The performance of the proposed models was evaluated by the coefficient of determination, root mean squared error, mean absolute error and variance accounted for. The results indicated that the SA-XGBoost model performed best. The Shapley additive explanations method revealed that the embedding depth was the most important indicator. Moreover, the convergence curves suggested that the SA-XGBoost model can reduce the generalization error and avoid overfitting.

References

[1] Kwon S, Lee C S, Cho S J, et al.. An investigation of the excavation damaged zone at the KAERI underground research tunnel [J]. Tunnelling and Underground Space Technology, 2009, 24(1): 1-13

[2] Wu Z-J, Jiang Y-L, Liu Q-S, et al.. Investigation of the excavation damaged zone around deep TBM tunnel using a Voronoi-element based explicit numerical manifold method [J]. International Journal of Rock Mechanics and Mining Sciences, 2018, 112: 158-170

[3] Perras M A, Diederichs M S. Predicting excavation damage zone depths in brittle rocks [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2016, 8(1): 60-74

[4] Li S-J, Feng X-T, Li Z-H, et al.. Evolution of fractures in the excavation damaged zone of a deeply buried tunnel during TBM construction [J]. International Journal of Rock Mechanics and Mining Sciences, 2012, 55: 125-138

[5] Yang H-Y, Cao S-G, Li Y, et al.. Assessment of excavation broken zone around gateways under various geological conditions: A case study in Sichuan Province, China [J]. Minerals, 2016, 6(3): 72

[6] Balland C, Renaud V. High-resolution velocity field imaging around a borehole: Excavation-damaged zone characterization [J]. Geophysics, 2009, 74(6): E223-E232

[7] Chen S-W, Yang C-H, Wang G-B, et al.. Experimental study on on-site acoustic emission monitoring of excavation damage area of Beishan pit exploration facilities [J]. Rock and Soil Mechanics, 2017, 38(S2): 349-359

[8] Xu N-W, Dai F, Li B, et al.. Comprehensive evaluation of excavation-damaged zones in the deep underground Caverns of the Houziyan Hydropower Station, Southwest China [J]. Bulletin of Engineering Geology and the Environment, 2017, 76(1): 275-293

[9] Kruschwitz S, Yaramanci U. Detection and characterization of the disturbed rock zone in claystone with the complex resistivity method [J]. Journal of Applied Geophysics, 2004, 57(1): 63-79

[10] Shemyakin E I, Kurlenya M V, Oparin V N, et al.. Zonal disintegration of rocks around underground workings. IV. Practical applications [J]. Soviet Mining, 1989, 25(4): 297-302

[11] Serrano A, Olalla C, Reig I. Convergence of circular tunnels in elastoplastic rock masses with non-linear failure criteria and non-associated flow laws [J]. International Journal of Rock Mechanics and Mining Sciences, 2011, 48(6): 878-887

[12] Xu X-D, Zhou Y-J, Zhu C, et al.. An analytical model for the excavation damage zone in tunnel surrounding rock [J]. Minerals, 2022, 12(10): 1321

[13] Vazaios I, Vlachopoulos N, Diederichs M S. Assessing fracturing mechanisms and evolution of excavation damaged zone of tunnels in interlocked rock masses at high stresses using a finite-discrete element approach [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2019, 11(4): 701-722

[14] Kwon S, Min K B. Fracture transmissivity evolution around the geological repository of nuclear waste caused by the excavation damage zone, thermoshearing and glaciation [J]. International Journal of Rock Mechanics and Mining Sciences, 2021, 137: 104554

[15] Bian H-B, Zhang X-T, Shao J-F. A coupled elastoplastic and visco-plastic damage model for hard clay and its application for the underground gallery excavation [J]. Underground Space, 2017, 2(1): 60-72

[16] Zhao Y, Zhao G-Y, Zhou J, et al.. Failure mechanism analysis of rock in particle discrete element method simulation based on moment tensors [J]. Computers and Geotechnics, 2021, 136: 104215

[17] Zhao Y, Zhao G-Y, Xu L, et al.. Mechanical property evolution model of cemented tailings-rock backfill considering strengthening and weakening effects [J]. Construction and Building Materials, 2023, 377: 131081

[18] Chang S H, Lee C I, Lee Y K. An experimental damage model and its application to the evaluation of the excavation damage zone [J]. Rock Mechanics and Rock Engineering, 2007, 40(3): 245-285

[19] Saiang D. Stability analysis of the blast-induced damage zone by continuum and coupled continuum-discontinuum methods [J]. Engineering Geology, 2010, 116(1–2): 1-11

[20] Liu Y, Ye Y-C, Wang Q-H, et al.. Predicting the loose zone of roadway surrounding rock using wavelet relevance vector machine [J]. Applied Sciences, 2019, 9(10): 2064

[21] Zhou J, Li X B. Evaluating the thickness of broken rock zone for deep roadways using nonlinear SVMs and multiple linear regression model [J]. Procedia Engineering, 2011, 26: 972-981

[22] Chen Y-X, Yong W-X, Li C-Q, et al.. Predicting the thickness of an excavation damaged zone around the roadway using the DA-RF hybrid model [J]. Computer Modeling in Engineering & Sciences, 2023, 136(3): 2507-2526

[23] XIE Qiang, PENG Kang. Space-time distribution laws of tunnel excavation damaged zones (EDZs) in deep mines and EDZ prediction modeling by random forest regression [J]. Advances in Civil Engineering, 2019: 6505984. DOI: https://doi.org/10.1155/2019/6505984.

[24] Zhao G-Y, Wang M, Liang W-Z. A comparative study of SSA-BPNN, SSA-ENN, and SSA-SVR models for predicting the thickness of an excavation damaged zone around the roadway in rock [J]. Mathematics, 2022, 10(8): 1351

[25] Yu Z, Shi X-Z, Zhou J, et al.. Advanced prediction of roadway broken rock zone based on a novel hybrid soft computing model using Gaussian process and particle swarm optimization [J]. Applied Sciences, 2020, 10(17): 6031

[26] Liang W-Z, Sari Y A, Zhao G-Y, et al.. Probability estimates of short-term rockburst risk with ensemble classifiers [J]. Rock Mechanics and Rock Engineering, 2021, 54(4): 1799-1814

[27] Phoon K K, Zhang W-G. Future of machine learning in geotechnics [J]. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 2023, 17(1): 7-22

[28] Zhang W-G, Gu X, Tang L-B, et al.. Application of machine learning, deep learning and optimization algorithms in geoengineering and geoscience: Comprehensive review and future challenge [J]. Gondwana Research, 2022, 109: 1-17

[29] Zhang Z-L, Pan Q-J, Yang Z-H, et al.. Physics-informed deep learning method for predicting tunnelling-induced ground deformations [J]. Acta Geotechnica, 2023, 18(9): 4957-4972

[30] Zhang Z L, Pan Q J, Zhang W G, et al.. Prediction of surface settlements induced by shield tunnelling using physics-informed neural networks [J]. Engineering Mechanics, 2024, 41(4): 161-173 (in Chinese)

[31] Wang Y-Y, Sun S-F, Chen X-Q, et al.. Short-term load forecasting of industrial customers based on SVMD and XGBoost [J]. International Journal of Electrical Power & Energy Systems, 2021, 129: 106830

[32] Pang X-F, Luan C-F, Liu L, et al.. Data-driven random forest forecasting method of monthly electricity consumption [J]. Electrical Engineering, 2022, 104(4): 2045-2059

[33] Meng D-L, Xu J, Zhao J-J. Analysis and prediction of hand, foot and mouth disease incidence in China using random forest and XGBoost [J]. PLoS One, 2021, 16(12): e0261629

[34] Guo K-H, Shen C-C, Zhou X-K, et al.. Traffic data-empowered XGBoost-LSTM framework for infectious disease prediction [J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(1): 774-785

[35] Gu Z-Y, Cao M-C, Wang C-G, et al.. Research on mining maximum subsidence prediction based on genetic algorithm combined with XGBoost model [J]. Sustainability, 2022, 14(16): 10421

[36] Zhou X-Z, Zhao C, Bian X-C. Prediction of maximum ground surface settlement induced by shield tunneling using XGBoost algorithm with golden-sine seagull optimization [J]. Computers and Geotechnics, 2023, 154: 105156

[37] Bhagat S K, Tung T M, Yaseen Z M. Heavy metal contamination prediction using ensemble model: Case study of Bay sedimentation, Australia [J]. Journal of Hazardous Materials, 2021, 403: 123492

[38] Jia X-L, Fu T-T, Hu B-F, et al.. Identification of the potential risk areas for soil heavy metal pollution based on the source-sink theory [J]. Journal of Hazardous Materials, 2020, 393: 122424

[39] He X, Zhao K-Y, Chu X-W. AutoML: A survey of the state-of-the-art [J]. Knowledge-Based Systems, 2021, 212: 106622

[40] Abordán A, Szabó N P. Machine learning based approach for the interpretation of engineering geophysical sounding logs [J]. Acta Geodaetica et Geophysica, 2021, 56(4): 681-696

[41] Garcia Nieto P J, Garcia-Gonzalo E, Lasheras F S, et al.. Forecast of the higher heating value in biomass torrefaction by means of machine learning techniques [J]. Journal of Computational and Applied Mathematics, 2019, 357(C): 284-301

[42] Tsai C W, Hsia C H, Yang S-J, et al.. Optimizing hyperparameters of deep learning in predicting bus passengers based on simulated annealing [J]. Applied Soft Computing, 2020, 88: 106068

[43] Gel Y R, Miao W-W, Gastwirth J L. Robust directed tests of normality against heavy-tailed alternatives [J]. Computational Statistics & Data Analysis, 2007, 51(5): 2734-2746

[44] Desai A. Statistics: Not only how but also why and when [J]. Journal of Public Administration Research and Theory, 2014, 24(1): 253-263

[45] Chen T-Q, Guestrin C. XGBoost: A scalable tree boosting system [C]. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, San Francisco California USA, ACM 785-794

[46] Zhang W-G, Zhang R-H, Wu C-Z, et al.. State-of-the-art review of soft computing applications in underground excavations [J]. Geoscience Frontiers, 2020, 11(4): 1095-1106

[47] Bentéjac C, Csörgő A, Martínez-Muñoz G. A comparative analysis of gradient boosting algorithms [J]. Artificial Intelligence Review, 2021, 54(3): 1937-1967

[48] Breiman L. Random forests [J]. Machine Language, 2001, 45(1): 5-32

[49] Gislason P O, Benediktsson J A, Sveinsson J R. Random forests for land cover classification [J]. Pattern Recognition Letters, 2006, 27(4): 294-300

[50] Metropolis N, Rosenbluth A W, Rosenbluth M N, et al.. Equation of state calculations by fast computing machines [J]. The Journal of Chemical Physics, 1953, 21(6): 1087-1092

[51] Metropolis N, Ulam S. The Monte Carlo method [J]. Journal of the American Statistical Association, 1949, 44(247): 335-341

[52] Kirkpatrick S, Gelatt C D, Vecchi M P. Optimization by simulated annealing [J]. Science, 1983, 220(4598): 671-680

[53] Hasanpour R, Rostami J, Schmitt J, et al.. Prediction of TBM jamming risk in squeezing grounds using Bayesian and artificial neural networks [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2020, 12(1): 21-31

[54] Zhou J, Qiu Y-G, Zhu S-L, et al.. Estimation of the TBM advance rate under hard rock conditions using XGBoost and Bayesian optimization [J]. Underground Space, 2021, 6(5): 506-515

[55] Bergstra J, Bardenet R, Bengio Y, et al.. Algorithms for hyper-parameter optimization [C]. Proceedings of the 24th International Conference on Neural Information Processing Systems. December 12–15, 2011, 2011, Granada, Spain, ACM 2546-2554

[56] Yin X, Gao F, Wu J, et al.. Compressive strength prediction of sprayed concrete lining in tunnel engineering using hybrid machine learning techniques [J]. Underground Space, 2022, 7(5): 928-943

[57] Pedregosa F, Varoquaux G, Gramfort A, et al.. Scikit-learn: Machine learning in python [J]. Journal of Machine Learning Research, 2011, 12: 2825-2830

[58] Bergstra J, Yamins D, Cox D D. Making a science of model search: Hyperparameter optimization in hundreds of dimensions for vision architectures [C]. Proceedings of the 30th International Conference on International Conference on Machine Learning, 2013, Atlanta, GA, USA, ACM I-115-I-123

[59] Hong Z-X, Tao M, Liu L-L, et al.. An intelligent approach for predicting overbreak in underground blasting operation based on an optimized XGBoost model [J]. Engineering Applications of Artificial Intelligence, 2023, 126: 107097

[60] Lin W-X, Liu L-L, Zhao G-Y, et al.. Developing hybrid DMO-XGBoost and DMO-RF models for estimating the elastic modulus of rock [J]. Mathematics, 2023, 11(18): 3886

[61] Liu L-L, Zhao G-Y, Liang W-Z. Slope stability prediction using k-NN-based optimum-path forest approach [J]. Mathematics, 2023, 11(14): 3071

[62] Zhou Q-M, Chen D-L, Hu Z-Y, et al.. Decompositions of Taylor diagram and DISO performance criteria [J]. International Journal of Climatology, 2021, 41(12): 5726-5732

[63] Kim D, Antariksa G, Handayani M P, et al.. Explainable anomaly detection framework for maritime main engine sensor data [J]. Sensors, 2021, 21(15): 5200

[64] Fryer D, Strümke I, Nguyen H. Shapley values for feature selection: The good, the bad, and the axioms [J]. IEEE Access, 2021, 9: 144352-144360

[65] Li Z-Q. Extracting spatial effects from machine learning model using local interpretation method: An example of SHAP and XGBoost [J]. Computers, Environment and Urban Systems, 2022, 96: 101845

[66] Lu Y, Fan X-P, Zhang Y, et al.. Machine learning models using SHapley additive exPlanation for fire risk assessment mode and effects analysis of stadiums [J]. Sensors, 2023, 23(4): 2151

RIGHTS & PERMISSIONS

Central South University