耦合多源勘察信息的水工岩体完整性智能评价方法

李明超; 李明泽; 韩帅; 张佳文; 赵文超

doi:10.11779/CJGE20220718

耦合多源勘察信息的水工岩体完整性智能评价方法 English Version

李明超^1,,
李明泽¹,
韩帅^2, ,,
张佳文¹,
赵文超^{1, 3}

1.
天津大学水利工程仿真与安全国家重点实验室, 天津 300350
2.
香港理工大学建筑与房地产学系, 香港 999077
3.
中水北方勘测设计研究有限责任公司, 天津 300222

基金项目:

国家自然科学面上基金项目 51879185

天津市杰出青年科学基金项目 17JCJQJC44000

详细信息

作者简介:
李明超(1979—)，男，博士，教授，博士生导师，主要从事水利工程仿真与智能分析方面的研究工作。E-mail: lmc@tju.edu.cn

通讯作者:
韩帅, E-mail: drjasonhan@163.com

中图分类号: TU451
计量
- 文章访问数: 0
- HTML全文浏览量: 0
- PDF下载量: 0
出版历程
- 收稿日期: 2022-06-05
- 网络出版日期: 2023-02-26

Intelligent evaluation method for integrity of hydraulic rock mass by coupling multi-source survey information

1.
State Key Laboratory of Hydraulic Engineering Simulation and Safety, Tianjin University, Tianjin 300350, China
2.
Architecture and Real Estate Learning, The Hong Kong Polytechnic University, Hongkong 999077, China
3.
Bei Fang Investigation, Design & Research Corporation Limited, Tianjin 300222, China

摘要

摘要: 岩体完整性是评价水工岩体质量等级的重要参数，传统方法往往采用的单一指标难以全面反映结构面、地下水、卸荷等地质条件对评判结果的综合影响。提出了一种耦合多源勘察信息的水工岩体完整性智能评价方法。首先利用合成少数类过采样算法对勘察信息数据进行均衡处理，改善数据集结构；进而采用随机森林算法分别对原始岩体完整性数据和预处理后的数据进行预测，结合实际工程数据进行有效性及适用性验证，并针对影响岩体完整性的不同因素对预测结果进行了讨论分析。结果表明，所提出的方法对数据集的均衡处理可有效改善少数类岩体样本完整性的评价准确率，通过耦合并挖掘不同完整性指标中的深层信息，实现岩体完整性的智能评价，为进一步辅助岩体质量评价提供新的方法。
- 岩体完整性 /
- 岩体质量评价 /
- 多源勘察信息 /
- 随机森林 /
- 合成少数类过采样
Abstract: The integrity of rock mass is an important parameter in evaluating the quality grade of hydraulic rock mass. The traditional methods often adopt a single index that cannot fully reflect the comprehensive influences of geological conditions such as structural plane, groundwater, and unloading on the evaluation results. An intelligent evaluation method for the integrity of hydraulic rock mass is proposed by coupling with multi-source survey information. Firstly, the synthetic minority oversampling (SMOTE) algorithm is used to balance the survey information data to improve the data set structure. Then the random forest algorithm is used to predict the original rock mass integrity data and the pre-processed data, respectively. Based on the data of actual projects, the validity and applicability are verified, and the predicted results are discussed and analyzed according to different factors affecting the integrity of rock mass. The results show that the proposed method can effectively improve the evaluation accuracy of the integrity of a few rock samples by balancing the data sets. By coupling and mining the deep information of different integrity indexes, the intelligent evaluation of rock mass integrity can be realized, which provides a new method for further assisting the evaluation of rock mass quality.
- rock mass integrity /
- evaluation of rock mass quality /
- multi-source survey information /
- random forest /
- SMOTE

HTML全文

图 1 方法总体结构

Figure 1. Overall framework

下载: 全尺寸图片幻灯片

图 2 SMOTE-RF模型混淆矩阵

Figure 2. Confusion matrix of SMOTE-RF model

下载: 全尺寸图片幻灯片

图 3 RF模型混淆矩阵

Figure 3. Confusion matrix of RF model

下载: 全尺寸图片幻灯片

图 4 PD01局部展开图

Figure 4. PD01 partial expansion diagram

下载: 全尺寸图片幻灯片

图 5 SMOTE-RF模型混淆矩阵

Figure 5. Confusion matrix of SMOTE-RF model

下载: 全尺寸图片幻灯片

图 6 RF模型混淆矩阵

Figure 6. Confusion matrix of RF model

下载: 全尺寸图片幻灯片

图 7 PD102岩脉照片

Figure 7. PD102 rock vein photograph

下载: 全尺寸图片幻灯片

图 8 各方法评价情况

Figure 8. Evaluation of various methods

下载: 全尺寸图片幻灯片

表 1 岩体完整性评价标准

Table 1 Evaluation standards for rock mass integrity

完整程度	完整	较完整	完整性差	较破碎	破碎
RQD/%	90~100	75~90	50~75	25~50	0~25
J_v/ (条·m^-3)	0~3	3~10	10~20	20~35	＞35
K_v	0.75~1	0.55~0.75	0.35~0.55	0.15~0.35	0~0.15

下载: 导出CSV

表 2 Kappa系数一致等级表

Table 2 Scales of Kappa coefficient

系数	0~0.2	0.2~0.4	0.4~0.6	0.6~0.8	0.8~1.0
一致等级	极低	一般	中等	高度	几乎完全一致

下载: 导出CSV

表 3 工程岩体多源勘察数据

Table 3 Multi-source survey data of engineering rock mass

序号	平硐	桩号	RQD/%	J_v/(条·m^-3)	K_v	应力分带	岩体结构类型	岩体嵌合程度	完整性
1	PD01	220—250	99	6.0	0.85	集中带	块状结构	紧密	完整(1)
2	PD04	130—175	63	12.9	0.37	过渡带	次块状结构	较紧密	完整性差(3)
3	PD05	146.7—188	55	7.3	0.87	平稳带	镶嵌碎裂结构	较紧密	较完整(2)
4	PD07	70—160	80	5.5	0.88	过渡带	中厚层状结构	较紧密	较完整(2)
5	PD12	37—75	29	21.0	0.17	弱卸荷带	层状破碎结构	松弛	较破碎(4)
6	PD30	17—65	51	16.8	0.53	释放带	破碎结构	较松驰	完整性差(3)

下载: 导出CSV

表 4 两种模型输出结果与岩体完整性对应关系

Table 4 Correspondence between output results of two models and rock mass integrity

岩体完整性	两种机器学习模型测试准确率/%
岩体完整性	RF模型	SMOTE-RF模型
完整	33.3	66.7
较完整	92.3	92.3
完整性差	88.2	100.0
较破碎	66.7	100.0
总准确率	83.3	94.4

下载: 导出CSV

表 5 工程岩体分段评价结果

Table 5 Sectional evaluation results of engineering rock mass

序号	平硐	桩号	野外勘探岩体完整性评价	SMOTE-RF模型岩体完整性评价
1	PD01	250—301.5	完整(1)	较完整(2)
2	PD46	190—215	较完整(2)	完整性差(3)

下载: 导出CSV

表 6 裂隙发育特征

Table 6 Characteristics of fracture development

组别	规则裂隙产状	裂隙线平均间距统计（5 m×2 m）	规则裂隙延伸长度	张开度/mm	粗糙度	充填物类型	主要结构面类型	地下水
1	N40-60°E/SE∠65-80°	一般 > 1，个别0.1~0.5	一般1~3	0~2	平糙	少许钙膜	节理裂隙	干燥
2	N30-50°E/NW∠25-35°	薄—中层	> 2	0~2	起滑	绿泥石膜	节理裂隙	湿润
3	N5-20°W/SW∠50-70°	零星	1~3	0~5	起糙	不连续泥膜、岩屑	结构面	湿润
4	N30-50°W/NE∠25-30°	零星	> 1	0~2	起糙	不连续泥膜、岩屑	结构面	湿润

下载: 导出CSV

表 7 该岩体分段三组构造带基本情况与特征

Table 7 Basic situations and characteristics of three groups of structural zones in rock mass

编号	构造类型	产状	起伏情况	充填物	力学性质	其他特征
f46-1	切层错动带	N25E/SE∠80	起伏	灰黄色压碎岩	压性	上游壁与顶拱接触处见一长25 cm，宽 6 cm，深60 cm的溶洞
f46-3	切层错动带	N75E/NW∠70	起伏	黑色压碎岩	压性	含土白色方解石粉沫，局部软化夹泥；上游壁见宽40~80 cm影响带，逆断层
f46-4	切层错动带	N20E/SE∠25	微起伏	灰黑色角砾岩及压碎岩	压扭性	局部软化，泥化；下盘见宽1.2 m破碎岩，破碎结构，上盘湿润

下载: 导出CSV

表 8 两种模型输出结果与岩体完整性对应关系

Table 8 Correspondence between output results of two models and rock mass integrity

岩体完整性	两种机器学习模型测试准确率/%
岩体完整性	RF模型	SMOTE-RF模型
完整	66.7	83.3
较完整	100.0	100.0
完整性差	62.5	87.5
总准确率	79.2	91.7

下载: 导出CSV

表 9 工程岩体分段评价结果

Table 9 Sectional evaluation results of engineering rock mass

序号	平硐	桩号	野外勘探岩体完整性评价	SMOTE-RF模型岩体完整性评价
1	PD102	64—85	完整(1)	较完整(2)
2	PD107	25—38	完整性差(3)	较完整(2)

下载: 导出CSV

表 10 工程岩体岩脉特征描述表

Table 10 Description of dike characteristics of engineering rock mass

编号	桩号	产状	宽度/cm	颜色	特征	完整性	蚀变程度
Ym102~5	左72顶72.8右75	N60°E/NW∠80°	30~60	浅黄色	发育断层f102~5及宽 2~4 cm的石英脉，石英脉局部宽9 cm	完整性差	蚀变不明显
Ym102~6	左78.3~79.4顶78.5~79.1右81~81.6	N60°E/NW∠60°	60~110	灰白—浅黄色	宽2~3 cm的石英脉，断续延伸	较完整	上盘局部蚀变3~5 cm

下载: 导出CSV

参考文献(30)

[1]	刘飞跃, 刘一汉, 杨天鸿, 等. 基于岩芯图像深度学习的矿山岩体质量精细化评价[J]. 岩土工程学报, 2021, 43(5): 968-974. doi: 10.11779/CJGE202105023 LIU Feiyue, LIU Yihan, YANG Tianhong, et al. Meticulous evaluation of rock mass quality in mine engineering based on machine learning of core photos[J]. Chinese Journal of Geotechnical Engineering, 2021, 43(5): 968-974. (in Chinese) doi: 10.11779/CJGE202105023
[2]	梁靖, 崔圣华, 裴向军, 等. 考虑岩体构造损伤的强震大型滑坡启动成因[J]. 岩土工程学报, 2021, 43(6): 1039-1049. doi: 10.11779/CJGE202106007 LIANG Jing, CUI Shenghua, PEI Xiangjun, et al. Initiation mechanism of earthquake-induced large landslides considering structural damage[J]. Chinese Journal of Geotechnical Engineering, 2021, 43(6): 1039-1049. (in Chinese) doi: 10.11779/CJGE202106007
[3]	NIKAFSHAN R H, JALALI Z, JALALIFAR H. Prediction of rock mass rating system based on continuous functions using Chaos–ANFIS model[J]. International Journal of Rock Mechanics and Mining Sciences, 2015, 73: 1-9. doi: 10.1016/j.ijrmms.2014.10.004
[4]	GHOLAMI R, RASOULI V, ALIMORADI A. Improved RMR rock mass classification using artificial intelligence algorithms[J]. Rock Mechanics and Rock Engineering, 2013, 46(5): 1199-1209. doi: 10.1007/s00603-012-0338-7
[5]	ZHANG Y Z, JI H G, LI W G, et al. Research on rapid evaluation of rock mass quality based on ultrasonic borehole imaging technology and fractal method[J]. Advances in Materials Science and Engineering, 2021, 2021: 1-9.
[6]	水利水电工程地质勘察规范: GB 50487—2008[S]. 北京: 中国计划出版社, 2009. Code for Engineering Geological Invcstigation of Water Resources and Hydropower: GB 50487—2008[S]. Beijing: China Planning Press, 2009. (in Chinese)
[7]	BARTON N, LIEN R, LUNDE J. Engineering classification of rock masses for the design of tunnel support[J]. Rock Mechanics, 1974, 6(4): 189-236. doi: 10.1007/BF01239496
[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. doi: 10.1016/S1365-1609(97)80069-X
[9]	许宏发, 陈锋, 王斌, 等. 岩体分级BQ与RMR的关系及其力学参数估计[J]. 岩土工程学报, 2014, 36(1): 195-198. doi: 10.11779/CJGE201401021 XU Hongfa, CHEN Feng, WANG Bin, et al. Relationship between RMR and BQ for rock mass classification and estimation of its mechanical parameters[J]. Chinese Journal of Geotechnical Engineering, 2014, 36(1): 195-198. (in Chinese) doi: 10.11779/CJGE201401021
[10]	李明超, 史博文, 韩帅, 等. 基于对穿声波波速的岩体完整性多尺度评价新指标与分析方法[J]. 岩石力学与工程学报, 2020, 39(10): 2060-2068. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX202010010.htm LI Mingchao, SHI Bowen, HAN Shuai, et al. New index and analysis method for multi-scale rock mass integrity assessment based on P-wave velocity[J]. Chinese Journal of Rock Mechanics and Engineering, 2020, 39(10): 2060-2068. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX202010010.htm
[11]	曹瑞琅, 王玉杰, 赵宇飞, 等. 基于钻进过程指数定量评价岩体完整性原位试验研究[J]. 岩土工程学报, 2021, 43(4): 679-687. doi: 10.11779/CJGE202104010 CAO Ruilang, WANG Yujie, ZHAO Yufei, et al. In-situ tests on quantitative evaluation of rock mass integrity based on drilling process index[J]. Chinese Journal of Geotechnical Engineering, 2021, 43(4): 679-687. (in Chinese) doi: 10.11779/CJGE202104010
[12]	李清波, 杜朋召. 基于边缘阈值分割的钻孔图像RQD自动分析方法研究[J]. 岩土工程学报, 2020, 42(11): 2153-2160. doi: 10.11779/CJGE202011022 LI Qingbo, DU Pengzhao. Automatic RQD analysis method based on information recognition of borehole images[J]. Chinese Journal of Geotechnical Engineering, 2020, 42(11): 2153-2160. (in Chinese) doi: 10.11779/CJGE202011022
[13]	GUO H S, FENG X T, LI S J, et al. Evaluation of the integrity of deep rock masses using results of digital borehole televiewers[J]. Rock Mechanics and Rock Engineering, 2017, 50(6): 1371-1382.
[14]	殷明伦, 张晋勋, 江玉生, 等. 岩体体积节理数表征岩体完整系数的结构面类别修正研究[J]. 岩土力学, 2021, 42(4): 1133-1140. https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX202104027.htm YIN Minglun, ZHANG Jinxun, JIANG Yusheng, et al. Study of correction of the structural plane category based on the rock mass integrity coefficient characterized by the volumetric joint count[J]. Rock and Soil Mechanics, 2021, 42(4): 1133-1140. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX202104027.htm
[15]	SONG Y H, XUE H S. Correlations between rock mass intactness index (K_v) and other rock mass classification indices (RMR₈₉ system and GSI)[J]. Bulletin of Engineering Geology and the Environment, 2021, 80(10): 7807-7816.
[16]	PELLS P J, BIENIAWSKI Z T, HENCHER S R, et al. Rock quality designation (RQD): time to rest in peace[J]. Canadian Geotechnical Journal, 2017, 54(6): 825-834.
[17]	DEERE D U. Technical description of rock cores for engineering purpose[J]. Rock Mechanics and Engineering Geology, 1964, 1(1): 17-22.
[18]	张文, 陈剑平, 牛岑岑, 等. 基于三维裂隙网络RQD的确定及最佳测线数量的研究[J]. 岩土工程学报, 2013, 35(2): 321-327. http://www.cgejournal.com/cn/article/id/14979 ZHANG Wen, CHEN Jianping, NIU Cencen, et al. Determination of RQD and number of optimum scanlines based on three-dimensional fracture network[J]. Chinese Journal of Geotechnical Engineering, 2013, 35(2): 321-327. (in Chinese) http://www.cgejournal.com/cn/article/id/14979
[19]	薛秋池, 赵其华, 何云松. 岩体结构面网络模拟的改进与应用[J]. 岩土工程学报, 2016, 38(7): 1351-1356. doi: 10.11779/CJGE201607025 XUE Qiuchi, ZHAO Qihua, HE Yunsong. Improvement and application of network simulation of rock mass discontinuities[J]. Chinese Journal of Geotechnical Engineering, 2016, 38(7): 1351-1356. (in Chinese) doi: 10.11779/CJGE201607025
[20]	ZHANG L Y. Determination and applications of rock quality designation (RQD)[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2016, 8(3): 389-397.
[21]	FALLS S D, YOUNG R P. Acoustic emission and ultrasonic-velocity methods used to characterise the excavation disturbance associated with deep tunnels in hard rock[J]. Tectonophysics, 1998, 289(1/2/3): 1-15.
[22]	段世委, 许仙娥. 岩体完整性系数确定及应用中的几个问题探讨[J]. 工程地质学报, 2013, 21(4): 548-553. https://www.cnki.com.cn/Article/CJFDTOTAL-GCDZ201304012.htm DUAN Shiwei, XU Xiane. Discussion of problems in calculation and application of rock mass integrity coefficient[J]. Journal of Engineering Geology, 2013, 21(4): 548-553. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GCDZ201304012.htm
[23]	陈祥, 孙进忠, 谭朝爽, 等. 岩块波速–应力关系及其卸荷效应[J]. 岩土工程学报, 2010, 32(5): 757-761. http://www.cgejournal.com/cn/article/id/13396 CHEN Xiang, SUN Jinzhong, TAN Chaoshuang, et al. Relation between P-wave velocity and stress of rock samples and their unloading effect[J]. Chinese Journal of Geotechnical Engineering, 2010, 32(5): 757-761. (in Chinese) http://www.cgejournal.com/cn/article/id/13396
[24]	水力发电工程地质勘察规范: GB/T 50287—2016[S]. 北京: 中国计划出版社, 2017. Code for Hydropower Engineering Geological Investigation: GB/T 50287—2016[S]. Beijing: China Planning Press, 2017. (in Chinese)
[25]	工程岩体分级标准: GB/T 50218—2014[S]. 北京: 中国计划出版社, 2015. Standard for Engineering Classification of Rock Mass: GB/T 50218—2014[S]. Beijing: China Planning Press, 2015. (in Chinese)
[26]	中小型水利水电工程地质勘察规范: SL 55—1993[S]. 北京: 中国水利电力出版社, 1994. Specification of Engineering Geological Investigation for Medium-Small Water Resources and Hydropower Development: SL 55—1993[S]. Beijing: China Water & Power Press, 1994. (in Chinese)
[27]	韩振华, 张路青, 袁广祥. 基于BQ系统的阿拉善巴彦诺日公NRG01号钻孔岩体质量评价[J]. 工程地质学报, 2019, 27(6): 1208-1215. https://www.cnki.com.cn/Article/CJFDTOTAL-GCDZ201906002.htm HAN Zhenhua, ZHANG Luqing, YUAN Guangxiang. Rock mass quality assessment of borehole NRG01 in bayannuorigong Alxa based on bq-system[J]. Journal of Engineering Geology, 2019, 27(6): 1208-1215. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GCDZ201906002.htm
[28]	CHAWLA N V, BOWYER K W, HALL L O, et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16: 321-357.
[29]	BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.
[30]	陈旭, 俞缙, 李宏, 等. 不同岩性及含水率的岩石声波传播规律试验研究[J]. 岩土力学, 2013, 34(9): 2527-2533. https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX201309015.htm CHEN Xu, YU Jin, LI Hong, et al. Experimental study of propagation characteristics of acoustic wave in rocks with different lithologies and water contents[J]. Rock and Soil Mechanics, 2013, 34(9): 2527-2533. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX201309015.htm