Tunnel rockburst prediction based on LLE-DBSCAN-SMOTE data processing

doi:10.16265/j.cnki.issn1003-3033.2024.12.1917

Abstract

Abstract:

To address issues of correlation prediction indicators, outliers, and data imbalance in original data in rockburst prediction, a rockburst prediction method based on LLE-DBSCAN-SMOTE for data processing was proposed. Firstly, the maximum tangential stress of surrounding rock $σ θ$ , uniaxial compressive strength of rock $σ c$ , uniaxial tensile strength of rock $σ t$ , elastic strain energy index $W e t$ , brittle coefficient $σ c / σ t$ , stress coefficient $σ θ / σ c$ , and stress concentration value β characterizing the stress gradient of surrounding rock were selected to construct a rockburst prediction indicator system. Secondly, the LLE algorithm was used for data dimensionality reduction to eliminate the cross-correlation effect between indicators, and the DBSCAN algorithm was introduced to remove outliers. Then, the SMOTE technology was introduced for data balancing. Finally, three types of rockburst prediction models were proposed using Decision Tree (DT), Random Forest (RF), and Gradient Boosting Decision Tree (GBDT) algorithms. The prediction accuracy of the data training models before and after processing was compared and analyzed. Moreover, engineering verification was performed through the measurement in the diversion tunnel of Jiangbian Hydropower Station. The results show that the prediction accuracy of the three types of algorithm models which reduce the prediction index from the 7 dimensions of the original data to the 4 dimensions and adopt the graded outlier processing is the highest among the similar models. The rockburst prediction of the Jiangbian Hydropower Station demonstrates that the proposed model significantly improves prediction accuracy compared to similar models using original rockburst data.

Key words: local linear embedding (LLE), density-based spatial clustering of applications with noise (DBSCAN), synthetic minority over-sampling technique (SMOTE), data processing, rockburst prediction

CLC Number:

X936

FAN Chengqiang, XIA Yuanyou, ZHANG Hongwei, HUANG Jian. Tunnel rockburst prediction based on LLE-DBSCAN-SMOTE data processing[J]. China Safety Science Journal, 2024, 34(12): 140-148.

Figures/Tables 11

Fig.1

Table 1

Rockburst case dataset

序号	$σ θ / M P a$	$σ c / M P a$	$σ t / M P a$	$σ θ / σ c$	$σ c / σ t$	$W e t$	$β / (M P a · m - 1)$	岩爆烈度	工程名称
1	18.80	178.00	5.70	0.11	31.23	7.40	1.04	1	龙羊峡水电站地下洞室
2	11.00	115.00	5.00	0.10	23.00	5.70	0.26	1	李家峡水电站地下洞室
3	34.00	150.00	5.40	0.23	27.78	7.80	0.70	1	鲁布革水电站地下隧洞
4	13.90	124.00	4.20	0.11	29.52	2.00	0.19	1	括苍山隧道
︙	︙	︙	︙	︙	︙	︙	︙	︙	︙
307	65.20	114.38	5.79	0.57	19.74	0.62	1.27	4	西康铁路秦岭隧道
308	85.50	144.91	7.89	0.59	18.35	0.81	1.31	4	新建川藏铁路某隧道
309	78.10	144.62	7.26	0.54	19.92	0.71	2.17	4	秦岭终南山公路隧道

Table 1

Table 2

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Table 3

Index parameters of diversion tunnel of Jiangbian Hydropower Station

里程	$σ θ / M P a$	$σ c / M P a$	$σ t / M P a$	$σ θ / σ c$	$σ c / σ t$	W_et	$β / (M P a · m - 1)$	岩爆烈度
0 + 250	33.15	106.94	5.84	0.31	18.31	2.15	0.24	2
0 + 550	91.43	157.63	11.96	0.58	13.17	6.27	1.32	3
0 + 900	51.50	132.05	6.33	0.39	20.86	4.63	0.96	2
2 + 230	23.39	106.32	2.92	0.22	36.41	1.75	0.70	1
3 + 320	12.96	117.81	3.21	0.11	36.70	2.49	0.09	1
3 + 710	89.52	146.75	7.54	0.61	19.46	4.70	3.11	3
7 + 330	121.09	159.33	11.29	0.76	14.11	11.62	1.94	4
7 + 625	121.09	135.67	9.02	0.45	15.04	11.2	0.11	4

Table 3

Table 4

References 25

[1]	冯夏庭, 肖亚勋, 丰光亮. 岩爆孕育过程研究[J]. 岩石力学与工程学报, 2019, 38(4): 650-673.
	FENG Xiating, XIAO Yaxun, FENG Guangliang. Study on the development process of rockbursts[J]. Chinese Journal of Rock Mechanics and Engineering, 2019, 38 (4): 650-673.
[2]	江飞飞, 周辉, 刘畅, 等. 地下金属矿山岩爆研究进展及预测与防治[J]. 岩石力学与工程学报, 2019, 38(5): 956-972.
	JIANG Feifei, ZHOU Hui, LIU Chang, et al. Research progress, prediction and prevention of rock bursts in underground metal mines[J]. Chinese Journal of Rock Mechanics and Engineering, 2019, 38 (5): 956-972.
[3]	ZHANG Junfei, WANG Yuhang, SUN Yuantian, et al. Strength of ensemble learning in multiclass classification of rockburst intensity[J]. International Journal for Numerical and Analytical Methods in Geomechanics, 2020, 44(13): 1833-1853. doi: 10.1002/nag.3111
[4]	CHEN Guangyao, WANG Mingwu, YAN Jiahui, et al. A connection cloud model coupled with improved conflict evidence fusion method for prediction of rockburst intensity[J]. IEEE Access, 2021,9: 113 535-113 549.
[5]	WOJTECKI A, IWASZENKO S, APEL D B, et al. An attempt to use machine learning algorithms to estimate the rockburst hazard in underground excavations of hard coal mine[J]. Energies, 2021, 14(21): 6920-6928.
[6]	谭文侃, 胡南燕, 叶义成, 等. 基于四大集成学习的岩爆烈度分级预测[J]. 岩石力学与工程学报, 2022, 41(增2): 3250-3259.
	TAN Wenkan, HU Nanyan, YE Yicheng, et al. Rockburst intensity classification prediction based on four ensemble learning[J]. Chinese Journal of Rock Mechanics and Engineering, 2022, 41 (S2): 3250-3259.
[7]	夏元友, 张宏伟, 吝曼卿, 等. 基于数据预处理技术并考虑围岩应力梯度影响的隧洞岩爆预测[J]. 岩土工程学报, 2023, 45(10): 1 987-1 994.
	XIA Yuanyou, ZHANG Hongwei, LIN Manqing, et al. Prediction of tunnel rockbursts based on data preprocessing technology considering influences of stress gradient of surrounding rock[J]. Chinese Journal of Geotechnical Engineering, 2023, 45(10): 1 987-1 994.
[8]	YIN Xin, LIU Quansheng, PAN Yucong, et al. Strength of stacking technique of ensemble learning in rockburst prediction with imbalanced data: comparison of eight single and ensemble models[J]. Natural Resources Research, 2021, 30(2): 1795-1815. doi: 10.1007/s11053-020-09787-0
[9]	GONG Fengqiang, WANG Yunliang, LUO Song. Rockburst proneness criteria for rock materials: review and new insights[J]. Journal of Central South University, 2020, 27(10): 2793-2821.
[10]	满轲, 武立文, 刘晓丽, 等. 基于灰色关联分析和SSA-RF模型的岩爆等级预测[J]. 金属矿山, 2023, 34(5): 202-212.
	MAN Ke, WU Liwen, LIU Xiaoli, et al. Rockburst grade prediction based on grey correlation analysis and SSA-RF model[J]. Metal Mines, 2023, 34(5): 202-212.
[11]	GUO Deping, CHEN Hemao, TANG Libin, et al. Assessment of rockburst risk using multivariate adaptive regression splines and deep forest model[J]. Acta Geotechnica, 2022, 17(4): 1183-1205.
[12]	李明亮, 李克钢, 秦庆词, 等. 岩爆烈度等级预测的机器学习算法模型探讨及选择[J]. 岩石力学与工程学报, 2021, 40(增1): 2806-2816.
	LI Mingliang, LI Kegang, QIN Qingci, et al. Discussion and selection of machine learning algorithm model for rockburst intensity grade prediction[J]. Chinese Journal of Rock Mechanics and Engineering, 2021, 40 (S1): 2806-2816.
[13]	侯克鹏, 孙华芬, 包广拓. 改进的MVO-GRNN神经网络岩爆预测模型研究[J]. 安全与环境学报, 2023, 41(3):1-12.
	HOU Kepeng, SUN Huafen, BAO Guangtuo. Research on improved MVO-GRNN neural network rockburst prediction model[J]. Journal of Safety and Environment, 2023, 41(3):1-12.
[14]	谢学斌, 李德玄, 孔令燕, 等. 基于CRITIC-XGB算法的岩爆倾向等级预测模型[J]. 岩石力学与工程学报, 2020, 39(10): 1975-1982.
	XIE Xuebin, LI Dexuan, KONG Lingyan, et al. Rockburst propensity prediction model based on CRITIC-XGB algorithm[J]. Chinese Journal of Rock Mechanics and Engineering, 2020, 39 (10): 1975-1982.
[15]	陈则黄, 李克钢, 李明亮, 等. 基于PCA-SOFM模型的岩爆烈度等级预测[J]. 地下空间与工程学报, 2022, 18(增2): 934-942.
	CHEN Zehuang, LI Kegang, LI Mingliang, et al. Prediction of rock burst intensity level based on PCA-SOFM model[J]. Chinese Journal of Underground Space and Engineering, 2022, 18 (S2): 934-942.
[16]	杨小彬, 裴艳宇, 程虹铭, 等. 基于SOFM神经网络模型的岩爆烈度等级预测方法[J]. 岩石力学与工程学报, 2021, 40(增1): 2708-2715.
	YANG Xiaobin, PEI Yanyu, CHENG Hongming, et al. Prediction method of rockburst intensity grade based on SOFM neural network model[J]. Chinese Journal of Rock Mechanics and Engineering, 2021, 40 (S1): 2708-2715.
[17]	张传庆, 卢景景, 陈珺, 等. 岩爆倾向性指标及其相互关系探讨[J]. 岩土力学, 2017, 38(5): 1397-1404.
	ZHANG Chuanqing, LU Jingjing, CHEN Jun, et al. Discussion on rock burst proneness indexes and their relation[J]. Rock and Soil Mechanics, 2017, 38 (5): 1397-1404.
[18]	夏元友, 刘昌昊, 刘夕奇, 等. 均布与梯度应力加载路径下岩爆破坏特征试验[J]. 中国安全科学学报, 2020, 30(5): 149-155. doi: 10.16265/j.cnki.issn1003-3033.2020.05.023
	XIA Yuanyou, LIU Changhao, LIU Xiqi, et al. Experimental study on rockburst characteristics under uniform and gradient stress loading paths[J]. China Safety Science Journal, 2020, 30 (5): 149-155. doi: 10.16265/j.cnki.issn1003-3033.2020.05.023
[19]	宋英华, 庞昭胜, 李墨潇, 等. 基于反赋权与MBCT-SR多维云模型算法岩爆预测研究[J]. 中国安全生产科学技术, 2022, 18(3): 40-46.
	SONG Yinghua, PANG Zhaosheng, LI Moxiao, et al. Research on rock burst prediction based on anti weighting and MBCT-SR multidimensional cloud model algorithm[J]. Journal of Safety Science and Technology, 2022, 18 (3): 40-46.
[20]	黄建, 夏元友, 吝曼卿. 基于改进组合赋权的岩爆多维云模型预测研究[J]. 中国安全科学学报, 2019, 29(7): 26-32. doi: 10.16265/j.cnki.issn1003-3033.2019.07.005
	HUANG Jian, XIA Yuanyou, LIN Manqing. Study on prediction of rock burst by multi-dimensional cloud model based on improved combined weight[J]. China Safety Science Journal, 2019, 29 (7): 26-32. doi: 10.16265/j.cnki.issn1003-3033.2019.07.005
[21]	胡建华, 黄鹏莅, 周坦, 等. 岩爆倾向性的改进有限云评价模型与工程应用[J]. 中国安全科学学报, 2022, 32(2): 90-98. doi: 10.16265/j.cnki.issn1003-3033.2022.02.013
	HU Jianhua, HUANG Pengli, ZHOU Tan, et al. Improved finite cloud evaluation model for rock burst tendency and its engineering application[J]. China Safety Science Journal, 2022, 32 (2): 90-98. doi: 10.16265/j.cnki.issn1003-3033.2022.02.013
[22]	白明洲, 王连俊, 许兆义. 岩爆危险性预测的神经网络模型及应用研究[J]. 中国安全科学学报, 2002, 12(4):65-69.
	BAI Mingzhou, WANG Lianjun, XU Zhaoyi. Neural network model and application research for predicting rock burst risk[J]. China Safety Science Journal, 2002, 12 (4): 65-69.
[23]	何佳其, 吝曼卿, 刘夕奇, 等. 引入梯度应力的岩爆预测方法[J]. 岩土工程学报, 2020, 42(11):2098-2105.
	HE Jiaqi, LIN Manqing, LIU Xiqi, et al. New method for introducing gradient stress into rock-burst prediction[J]. Chinese Journal of Geotechnical Engineering, 2020, 42(11):2098-2105.
[24]	ZHANG Hongwei, XIA Yuanyou, LIN Manqin, et al. A three-step rockburst prediction model based on data preprocessing combined with clustering and classification algorithms[J]. Bulletin of Engineering Geology and the Environment, 2024, 83:1-19.
[25]	王艳芳. 基于微震监测数据的岩爆智能预警方法研究[D]. 沈阳: 东北大学, 2019.
	WANG Yanfang. Research on intelligent early warning method for rock burst based on microseismic monitoring data[D]. Shenyang: Northeastern University, 2019.

n	最小重构误差值
2	1.53×10^-16
3	8.13×10^-17
4	4.86×10^-17
5	5.15×10^-17
6	9.09×10^-16