A surrogate model for the rapid prediction of rockburst risk based on numerical samples and random forest classifier

WANG Kezhong; XIE Tian; LI Mei; ZHANG Rujiu; HOU Shaokang; WANG Zhenzhou; LIU Yaoru

doi:10.16511/j.cnki.qhdxxb.2024.26.027

Journal of Tsinghua University(Science and Technology) >

2024 , Vol. 64 >Issue 7: 1203 - 1214

DOI: https://doi.org/10.16511/j.cnki.qhdxxb.2024.26.027

Research Article

A surrogate model for the rapid prediction of rockburst risk based on numerical samples and random forest classifier

WANG Kezhong ,
XIE Tian ,
LI Mei ,
ZHANG Rujiu ,
HOU Shaokang ,
WANG Zhenzhou ,
LIU Yaoru

Expand

1. School of Civil Engineering, Zhejiang University of Technology, Hangzhou 310014, China;
2. Yunnan Dianzhong Water Diversion Engineering Co., Ltd., Kunming 650000, China;
3. State Key Laboratory of Hydroscience and Engineering, Tsinghua University, Beijing 100084, China;
4. China Renewable Energy Engineering Institute, Power Construction Corporation of China, Beijing 100120, China;
5. State Grid Economic and Technological Research Institute Co., Ltd., Beijing 102200, China

Received date: 2023-10-31

Online published: 2024-06-25

Fold

Abstract

[Objective] This study aims to address the significant challenge of predicting rockburst risks during the excavation of deep tunnels using tunnel boring machine (TBM) tunnel boring machine and develop a rapid prediction model to provide the basis for rockburst prevention and control, enhancing the safety and efficiency of deep tunnel construction. The proposed model leverages numerical samples and random forest (RF) algorithms to overcome the limitations of existing methods, which often do not achieve real-time and rapid prediction or consider the underlying mechanisms and factors influencing rockbursts. [Methods] Considering the Xianglushan Tunnel within the Dianzhong Water Diversion Project, we introduced a model that utilizes geostress and rock constitutive parameters as inputs and the elastic strain energy density of the surrounding rock as output. Numerical simulations of tunneling using the TBM under various working conditions arewere conducted, and 611 numerical samples were crafted through an orthogonal experimental design. We employed RF as the underlying classifier, with hyperparameters optimized through 10-fold cross-validation to create an efficient prediction model. The accuracy and applicability of the model were confirmed by comparing several machine learning algorithms. [Results] We conducted a series of numerical simulations of excavation using the TBM, employing an elastoviscoplastic constitutive model with internal variables. These simulations disclosed the energy evolution within the rock mass throughout the excavation process. Energy concentration occurred during transient unloading and the time-dependent deformation of the surrounding rock, leading to two distinct peaks in strain energy density. The second peak indicative the final energy storage during the creep phase of the surrounding rock postexcavation and unloading. Notably, a higher value at the tunnel wall—under identical conditions—correlated with an elevated risk of strainburst. We verified the rationality of the input and output parameters by analyzing energy evolution and correlation. The predictive accuracy and computational efficiency of the model were enhanced following the optimization of the hyperparameters using a 10-fold cross-validation. The input parameters partially mirrored the factors influencing rockburst, while the output parameters measured the energy storage status of the surrounding rock before potential rockburst failure. The RF-based rockburst risk prediction proxy model exhibited commendable performance on the training and testing sets, achieving accuracies of 99.75% and 82.02%, respectively. The performance of the RF-based rockburst risk prediction proxy model was superior to that of four other machine learning models—decision tree, K-nearest neighbors, support vector machine, and logistic regression—achieving prediction accuracies of 82.02%, 76.40%, 79.77%, 75.28%, and 76.40% for all samples, respectively. This result indicateds the robust predictive capability and generalization performance of the RF-based rockburst risk prediction proxy model in assessing rockburst risk levels. [Conclusions] We offer a novel approach and framework for the rapid prediction of rockburst risks during the excavation phase of deep tunnels. The RF-based rockburst risk prediction proxy model is reportedly an effective tool for rockburst risk prediction, marking a significant advancement in rockburst risk management. We provide a research path and framework for the rapid prediction of rockburst risk during the excavation period of deep tunnels.

Key words： deep tunnel; rockburst risk; tunnel boring machine; numerical samples; random forest; elastic strain energy

Cite this article

WANG Kezhong , XIE Tian , LI Mei , ZHANG Rujiu , HOU Shaokang , WANG Zhenzhou , LIU Yaoru . A surrogate model for the rapid prediction of rockburst risk based on numerical samples and random forest classifier[J]. Journal of Tsinghua University(Science and Technology), 2024 , 64(7) : 1203 -1214 . DOI: 10.16511/j.cnki.qhdxxb.2024.26.027

References

[1] LIU Y R, HOU S K, LI C Y, et al. Study on support time in double-shield TBM tunnel based on self-compacting concrete backfilling material[J]. Tunnelling and Underground Space Technology, 2020, 96:103212.
[2] 刘晓丽,孙欢,董勤喜,等.深埋引水隧洞极硬岩TBM掘进及辅助破岩技术[J].清华大学学报(自然科学版), 2022, 62(8):1292-1301. LIU X L, SUN H, DONG Q X, et al. Extremely hard rock mass excavation using rock breakdown methods to assist TBM in a deep, long diversion tunnel in the Qinling Mountains[J]. Journal of Tsinghua University (Science and Technology), 2022, 62(8):1292-1301.(in Chinese)
[3] 杜立杰,洪开荣,王佳兴,等.深埋隧道TBM施工岩爆特征规律与防控技术[J].隧道建设(中英文), 2021, 41(1):1-15. DU L J, HONG K R, WANG J X, et al. Rockburst characteristics and prevention and control technologies for tunnel boring machine construction of deep-buried tunnels[J]. Tunnel Construction, 2021, 41(1):1-15.(in Chinese)
[4] 梁伟章,赵国彦.深部硬岩长短期岩爆风险评估研究综述[J].岩石力学与工程学报, 2022, 41(1):19-39. LIANG W Z, ZHAO G Y. A review of long-term and short-term rockburst risk evaluations in deep hard rock[J]. Chinese Journal of Rock Mechanics and Engineering, 2022, 41(1):19-39.(in Chinese)
[5] RUSSENES B F. Analysis of rock spalling for tunnels in steep valley sides[D]. Trondheim:Norwegian Institute of Technology, 1974.
[6] BARTON N, LIEN R, LUNDE J. Engineering classification of rock masses for the design of tunnel support[J]. Rock Mechanics and Rock Engineering, 1974, 6(4):189-236.
[7] TURCHANINOV I A, MARKOV G A, LOVCHIKOV A V. Conditions of changing of extra-hard rock into weak rock under the influence of tectonic stresses of massifs[C]//ISRM International Symposium. Tokyo, Japan:ISRM, 1981:133248625.
[8] HOEK E, BROWN E T. Practical estimates of rock mass strength[J]. International Journal of Rock Mechanics and Mining Sciences, 1997, 34(8):1165-1186.
[9] GONG F Q, WANG Y L, LUO S. Rockburst proneness criteria for rock materials:Review and new insights[J]. Journal of Central South University, 2020, 27(10):2793-2821.
[10] JIANG Q, FENG X T, XIANG T B, et al. Rockburst characteristics and numerical simulation based on a new energy index:A case study of a tunnel at 2500 m depth[J]. Bulletin of Engineering Geology and the Environment, 2010, 69(3):381-388.
[11] ZHANG C Q, ZHOU H, FENG X T. An index for estimating the stability of brittle surrounding rock mass:FAI and its engineering application[J]. Rock Mechanics and Rock Engineering, 2011, 44(4):401-414.
[12] 邱士利,冯夏庭,江权,等.深埋隧洞应变型岩爆倾向性评估的新数值指标研究[J].岩石力学与工程学报, 2014, 33(10):2007-2017. QIU S L, FENG X T, JIANG Q, et al. A novel numerical index for estimating strainburst vulnerability in deep tunnels[J]. Chinese Journal of Rock Mechanics and Engineering, 33(10):2007-2017.(in Chinese)
[13] ZHANG R J, LIU Y R, HOU S K. Evaluation of rockburst risk in deep tunnels considering structural planes based on energy dissipation rate criterion and numerical simulation[J]. Tunnelling and Underground Space Technology, 2023, 137:105128.
[14] 马天辉,唐春安,唐烈先,等.基于微震监测技术的岩爆预测机制研究[J].岩石力学与工程学报, 2016, 35(3):470-483. MA T H, TANG C A, TANG L X, et al. Mechanism of rock burst forcasting based on micro-seismic monitoring technology[J]. Chinese Journal of Rock Mechanics and Engineering, 2016, 35(3):470-483.(in Chinese)
[15] FENG G L, CHEN B R, XIAO Y X, et al. Microseismic characteristics of rockburst development in deep TBM tunnels with alternating soft-hard strata and application to rockburst warning:A case study of the Neelum-Jhelum hydropower project[J]. Tunnelling and Underground Space Technology, 2022, 122:104398.
[16] 李邵军,郑民总,邱士利,等.中国锦屏地下实验室开挖隧洞灾变特征与长期原位力学响应分析[J].清华大学学报(自然科学版), 2021, 61(8):842-852. LI S J, ZHENG M Z, QIU S L, et al. Characteristics of excavation disasters and long-term in-situ mechanical behavior of the tunnels in the China Jinping Underground Laboratory[J]. Journal of Tsinghua University (Science and Technology), 2021, 61(8):842-852.(in Chinese)
[17] 冯夏庭.地下峒室岩爆预报的自适应模式识别方法[J].东北大学学报, 1994, 15(5):471-475. FENG X T. Adaptive pattern recognition to predict rockbursts in underground openings[J]. Journal of Northeastern University, 1994, 15(5):471-475.(in Chinese)
[18] 王元汉,李卧东,李启光,等.岩爆预测的模糊数学综合评判方法[J].岩石力学与工程学报, 1998, 17(5):483-501. WANG Y H, LI W D, LI Q G, et al. Method of fuzzy comprehensive evaluations for rockburst prediction[J]. Chinese Journal of Rock Mechanics and Engineering, 1998, 17(5):483-501.(in Chinese)
[19] 宫凤强,李夕兵,张伟.基于Bayes判别分析方法的地下工程岩爆发生及烈度分级预测[J].岩土力学, 2010, 31(S1):370-377, 387. GONG F Q, LI X B, ZHANG W. Rockburst prediction of underground engineering based on Bayes discriminant analysis method[J]. Rock and Soil Mechanics, 2010, 31(S1):370-377, 387.(in Chinese)
[20] 徐飞,徐卫亚.岩爆预测的粒子群优化投影寻踪模型[J].岩土工程学报, 2010, 32(5):718-723. XU F, XU W Y. Projection pursuit model based on particle swarm optimization for rock burst prediction[J]. Chinese Journal of Geotechnical Engineering, 2010, 32(5):718-723.(in Chinese)
[21] 贾义鹏,吕庆,尚岳全.基于粒子群算法和广义回归神经网络的岩爆预测[J].岩石力学与工程学报, 2013, 32(2):343-348. JIA Y P, LV Q, SHANG Y Q. Rockburst prediction using particle swarm optimization algorithm and general regression neural network[J]. Chinese Journal of Rock Mechanics and Engineering, 2013, 32(2):343-348.(in Chinese)
[22] 王迎超,靖洪文,张强,等.基于正态云模型的深埋地下工程岩爆烈度分级预测研究[J].岩土力学, 2015, 36(4):1189-1194. WANG Y C, JING H W, ZHANG Q, et al. A normal cloud model-based study of grading prediction of rockburst intensity in deep underground engineering[J]. Rock and Soil Mechanics, 2015, 36(4):1189-1194.(in Chinese)
[23] 徐琛,刘晓丽,王恩志,等.基于组合权重-理想点法的应变型岩爆五因素预测分级[J].岩土工程学报, 2017, 39(12):2245-2252. XU C, LIU X L, WANG E Z, et al. Prediction and classification of strain mode rockburst based on five-factor criterion and combined weight-ideal point method[J]. Chinese Journal of Geotechnical Engineering, 2017, 39(12):2245-2252.(in Chinese)
[24] 汤志立,王雪,徐千军.基于过采样和客观赋权法的岩爆预测[J].清华大学学报(自然科学版), 2021, 61(6):543-555. TANG Z L, WANG X, XU Q J. Rockburst prediction based on oversampling and objective weighting method[J]. Journal of Tsinghua University (Science and Technology), 2021, 61(6):543-555.(in Chinese)
[25] LU J W, GONG Q M, YIN L J, et al. Study on the tunneling response of TBM in stressed granite rock mass in Yinhan Water Conveyance tunnel[J]. Tunnelling and Underground Space Technology, 2021, 118:104197.
[26] 王湘怡,周小雄,卢建炜,等.基于机器学习的TBM隧道掘进岩爆预测[J].施工技术(中英文), 2022, 51(20):1-7. WANG X Y, ZHOU X X, LU J W, et al. Rockburst prediction of TBM tunneling based on machine learning[J]. Construction Technology, 2022, 51(20):1-7.(in Chinese)
[27] 满轲,武立文,刘晓丽,等.基于CNN-LSTM模型的TBM隧道掘进参数及岩爆等级预测[J/OL].煤炭科学技术:1-19.(2023-10-26)[2023-10-28]. https://kns.cnki.net/kcms2/article/abstract?v=PkrNiO65NLlvH_l8Xa_2xQow2qwIZy01LmrkmhFRtqEkLAlzmoK5R0PMHJGAx5F_H85JydjWaGTfQ-i-WJFyeta_-JGY59F8MxNWYkgmv-yr3RgzK2IlPnOF8VK-ayLZChLO2ItcI1E=&uniplatform=NZKPT&language=CHS. MAN K, WU L W, LIU X L, et al. The prediction of TBM tunnel boring parameters and rockburst grade based on CNN-LSTM model[J/OL]. Coal Science and Technology:1-19.(2023-10-26)[2023-10-28]. https://kns.cnki.net/kcms2/article/abstract?v=PkrNiO65NLlvH_l8Xa_2xQow2qwIZy01LmrkmhFRtqEkLAlzmoK5R0PMHJGAx5F_H85JydjWaGTfQ-i-WJFyeta_-JGY59F8MxNWYkgmv-yr3RgzK2IlPnOF8VK-ayLZChLO2ItcI1E=&uniplatform=NZKPT&language=CHS. (in Chinese)
[28] 张泷.基于内变量热力学的流变模型及岩体结构长期稳定性研究[D].北京:清华大学, 2015. ZHANG L. Research on rheological model based on thermodynamics with internal state variables and long-term stability of rock mass structures[D]. Beijing:Tsinghua University, 2015.(in Chinese)
[29] LI C Y, HOU S K, LIU Y R, et al. Analysis on the crown convergence deformation of surrounding rock for double-shield TBM tunnel based on advance borehole monitoring and inversion analysis[J]. Tunnelling and Underground Space Technology, 2020, 103:103513.
[30] 侯少康,刘耀儒.双护盾TBM掘进数值仿真及护盾卡机控制因素影响分析[J].清华大学学报(自然科学版), 2021, 61(8):809-817. HOU S K, LIU Y R. Numerical simulations of double-shield TBM tunneling for analyzing shield jamming control factors[J]. Journal of Tsinghua University (Science and Technology), 2021, 61(8):809-817.(in Chinese)
[31] 宫凤强,闫景一,李夕兵.基于线性储能规律和剩余弹性能指数的岩爆倾向性判据[J].岩石力学与工程学报, 2018, 37(9):1993-2014. GONG F Q, YAN J Y, LI X B. A new criterion of rock burst proneness based on the linear energy storage law and the residual elastic energy index[J]. Chinese Journal of Rock Mechanics and Engineering, 2018, 37(9):1993-2014.(in Chinese)
[32] HOU S K, LIU Y R, ZHUANG W Y, et al. Prediction of shield jamming risk for double-shield TBM tunnels based on numerical samples and random forest classifier[J]. Acta Geotechnica, 2023, 18(1):495-517.

Options

Outlines

模态框（Modal）标题

Abstract

Cite this article

References

Visited