To improve the accuracy of water inrush and mud outburst disaster prediction during tunnel construction through karst caves, a method of first forecasting the unfavorable geological conditions of karst caves and subsequently predicting water inrush and mud outburst geological disasters was proposed in this paper. First, a karst cave advanced prediction model based on the polynomial naive Bayes algorithm was established to predict the scale grade of karst caves. Secondly, taking the scale prediction results of karst caves as the input condition for water inrush and mud outburst, a support vector machine (SVM) grade prediction model for karst cave water inrush and mud outburst based on the genetic algorithm (GA) was established. The random seed was regarded as an optimizable hyperparameter, and the overall training effect of machine learning models could be improved by adjusting the random seed during data partitioning; the prediction imputation methods, namely the K-nearest neighbor algorithm and SMOTE algorithm, were adopted to generate synthetic samples to fill in missing sample values and solve the problem of sample imbalance. By comparing the GA-SVM model with three models (SVM, GS-SVM, and IPOS-SVM), the results indicate that the overall prediction accuracy of the GA-SVM model is the highest. Model testing demonstrates that the accuracy of the comprehensive advanced prediction model for karst caves is 81.25%, and the accuracy of the grade prediction model for water inrush and mud outburst is 91.67%. Applications in multiple tunnels indicate good effects, and the research results provide a new method for the prediction of water inrush and mud outburst in tunnel karst caves.
为了量化分析岩溶溶洞等级与预测指标之间的相关性,采用最大信息系数理论(Maximal Information Coefficient,MIC),进行相关性系数分析,结果如图2所示。由图2可见:两两预报指标之间的相关性系数基本小于0.6。根据Fan等[18]研究,当两个指标之间的相关性系数绝对值小于0.6时,则认为两个指标相互独立,可以同时保留在模型中。
LIShucai, XUEYiguo, ZHANGQingsong, et al. Key technology study on comprehensive prediction and early-warning of geological hazards during tunnel construction in high-risk karst areas[J]. Chinese Journal of Rock Mechanics and Engineering, 2008, 27(7): 1297-1307.
ZHANGQingsong, LIShucai, HANHongwei, et al. Study on risk evaluation and water inrush disaster preventing technology during construction of karst tunnels[J]. Journal of Shandong University (Engineering Science),2009,39(3):106-110.
LILiping, LIShucai, CHENJun, et al. Research on the tunnel construction permit mechanism based on karst water burst risk evaluation and its application [J]. Chinese Journal of Rock Mechanics and Engineering, 2011, 30(07): 1345-1355.
BAIMingzhou, XUZhaoyi, WANGMeng, et al. Study on the technology of advance forecasting underground water and mud in rush during tunnel construction in karst area[J]. Journal of Highway and Transportation Research and Development, 2005, 22(S1): 123-126.
HANLin, LIHongtao, SUNYafei, et al. Classification method and engineering application of risk of water inrush and mud in tunnel[J]. Water Resources and Hydropower Engineering, 2023, 54(sup 1): 197-202.
MAOBangyan. Modern deep karst formation mechanism and evaluation of control action on cross-mountain tunnel[D]. Chengdu: Chengdu University of Technology, 2008.
ZHUZhen, WANGXuchun, YUANYongcai, et al. Risk assessment of water inrush in karst tunnels based on weighted average method[J]. Highway Engineering, 2015, 40(6): 51-54.
XUZhong, DENGHui, LUPing, et al. Preliminary study on risk evaluation system of water inrush and mud burst in karst section of tunnel[J]. Water Resources and Hydropower Engineering, 2018(3): 127-134.
WENGQineng, WUBingqi, XIANGShuai, et al. Model study of risk assessment about water bursting and mud gushing in tunnel[J]. Journal of Chongqing Jiaotong University (Natural Sciences), 2012, 31(5): 944-947,965.
[19]
杨艳娜. 西南山区岩溶隧道涌突水灾害危险性评价系统研究[D]. 成都: 成都理工大学, 2009.
[20]
YANGYanna. Research of karst tunnel water bursting hazard risk assessment system in the southwest mountainous area[D]. Chengdu: Chengdu University of Technology, 2009.
[21]
LIS C, WUJ. A multi-factor comprehensive risk assessment method of karst tunnels and its engineering application[J]. Bulletin of Engineering Geology and the Environment, 2019, 78(3): 1761-1776.
[22]
LIZ Y, WANGY C, OLGUNC G, et al. Risk assessment of water inrush caused by karst cave in tunnels based on reliability and GA-BP neural network[J]. Geomatics, Natural Hazards and Risk, 2020, 11(1): 1212-1232.
[23]
刘志刚, 凌宏亿, 俞文生. 隧道隧洞超前地质预报[M]. 北京: 人民交通出版社, 2011.
[24]
LIUZhigang, LINGHongyi, YUWensheng. Advanced geological prediction of tunnel[M]. Beijing: China Communications Press, 2011.
[25]
刘阳飞. 超特长隧道综合超前地质预报工法研究与应用[D]. 成都: 成都理工大学, 2016.
[26]
LIUYangfei. Comprehensive advanced geological prediction method for super long tunnel and its application[D]. Chengdu: Chengdu University of Technology, 2016.
[27]
周双龙. 基于CNN的隧道TSP超前地质预报智能解译初探[D]. 成都:成都理工大学, 2022.
[28]
ZHOUShuanglong. Preliminary study on intelligent interpretation of tunnel TSP advanced geological forecasting based on CNN[D]. Chengdu:Chengdu University of Technology, 2022.
[29]
李晨博. 岩溶隧道探地雷达超前地质预报的智能解译研究[D]. 成都: 成都理工大学, 2021.
[30]
LIChenbo. Research on intelligent interpretation of advanced geological forecast by ground penetrating radar in karst tunnel[D]. Chengdu: Chengdu University of Technology, 2021.
HUJuyi, WANGLiancheng. Research on prewarning system for unfavorable geologies in tunnel construction based on comprehensive advanced geological prediction[J]. Technology of Highway and Transport, 2009, 25(6): 116-119,124.
[33]
FANX M, ROSSITERD G, VAN WESTENC J, et al. Empirical prediction of coseismic landslide dam formation[J]. Earth Surface Processes and Landforms, 2014, 39(14): 1913-1926.
WUGao, LIMing, ZHOUDaoxiang, et al. Text classification based on deep ensemble naive Bayes model[J]. Journal of University of Jinan (Science and Technology), 2020, 34(5): 436-442.
WANGLu. Prediction of water inflow in southwest karst crossing-mountain tunnel based on genetic algorithm and support vector machine[D]. Chengdu: Chengdu University of Technology, 2019.
[38]
王凯平. 弱胶结软岩隧道突水灾变演化特征与预测方法研究[D]. 济南: 济南大学, 2022.
[39]
WANGKaiping. Study on evolution characteristics and prediction method of water inrush disaster in weakly consolidated soft rock tunnel[D]. Jinan: University of Jinan, 2022.
FUZhongliang, CHENXiaoqing, RENWei, et al. Random K-nearest neighbor algorithm with learning process[J]. Journal of Jilin University (Engineering and Technology Edition), 2024, 54(1): 209-220.
WANGZhen, TONGZhiyong, PENGMeihua, et al. Research on network intrusion detection based on improved SMOTE algorithm[J]. Journal of Natural Science of Heilongjiang University, 2024, 41(4): 470-476.