岩溶区隧道突泥涌水灾害发育特征分析及处治方案研究与效果评价

何为; 张世林; 徐恺奇; 赵格立

doi:10.11932/karst2026y013

岩溶区隧道突泥涌水灾害发育特征分析及处治方案研究与效果评价

doi: 10.11932/karst2026y013

何为^{1, 2,},
张世林³,
徐恺奇⁴,
赵格立⁵

1.
长安大学, 陕西西安 710064
2.
湖北交投巴楚建设管理有限公司, 湖北恩施 444300
3.
中国地质科学院探矿工艺研究所, 四川成都 611734
4.
中咨公路养护检测技术有限公司, 北京 100089
5.
雅砻江流域水电开发有限公司, 四川成都 610051

基金项目: 国家自然科学基金项目（52079091）

详细信息

作者简介:
何为（1985-），男，在读博士研究生，高级工程师，主要从事隧道工程、岩土工程等研究工作。E-mail：hewei202508@163.com

中图分类号: U453.6
计量
- 文章访问数: 28
- HTML浏览量: 10
- PDF下载量: 17
- 被引次数: 0
出版历程
- 收稿日期: 2025-09-23
- 录用日期: 2026-04-08
- 修回日期: 2026-03-26
- 刊出日期: 2026-04-01

Analysis of development characteristics, treatment schemes, and effectiveness evaluation of mud and water inrush in tunnels within karst area

HE Wei^{1, 2
,},
ZHANG Shilin³,
XU Kaiqi⁴,
ZHAO Geli⁵

1.
Chang'an University, Xi'an, Shaanxi 710064 , China
2.
Hubei Jiaotong Bachu Construction Management Co., LTD., Enshi, Hubei 444300, China
3.
Institute of Exploration Technology, Chinese Academy of Geological Sciences, Chengdu, Sichuan 611734, China
4.
Beijing Zhongjiao Engineering Testing Technology Co., Ltd., Beijing 100089 , China
5.
Yalong River Hydropower Development Co., Ltd., Chengdu, Sichuan 610051 , China

摘要

摘要: 为确保岩溶区隧道突泥涌水灾害段的安全施工，先开展隧道突泥涌水灾害的发育特征及成因分析，结合工程实际，根据其评价结果，开展了突泥涌水灾害的处治方案设计；再利用麻雀搜索算法、支持向量机联合构建隧道变形预测模型，充分掌握突泥涌水灾害处治后的隧道变形规律，以评价灾害处治措施的合理性。结果表明：隧道左洞ZK24+405m处突泥涌水灾害发育特征较为显著，提出通过超前预支护、溶腔分段加固、衬砌加强及防排水加强等处治方案来确保灾害段的安全施工；经灾害防治后的变形成果分析，其变形值将会趋于收敛，且始终在设计要求的变形控制范围内，验证了各类处治措施的有效性，为类似工程积累了经验。
- 岩溶 /
- 隧道 /
- 发育特征 /
- 变形预测 /
- 处治方案设计
Abstract: To ensure the safe construction of tunnel sections prone to mud and water inrush disasters in karst areas, taking the Naqing Tunnel of the Luowang Expressway in Qiannan Buyi and Miao Autonomous Prefecture, Guizhou Province as a case study, the developmental characteristics and genesis of such disasters were first analyzed. Based on the evaluation results and considering practical engineering conditions, treatment solutions were designed from aspects such as advanced pre-support, karst cavities, lining, and waterproofing-drainage measures. Finally, a deformation prediction model for tunnels was jointly constructed using the Sparrow Search Algorithm and Support Vector Machine, namely the Sparrow Search Algorithm-Support Vector Machine model (SSA-SVM). This model was then applied to predict the deformation at sections ZK24+405 m and ZK24+400 m, with the prediction results used to understand the deformation patterns of the tunnel after disaster treatment and evaluate the rationality of the mitigation measures.The analysis results indicate that at the left tunnel section ZK24+405 m, mud ejection occurred during drilling at the left arch waist of the upper step on August 29, with a spraying distance exceeding 30 meters. The mud pressure in the borehole pushed the drill rod out, triggering a mud and water inrush disaster. By September 6, the inflow gradually decreased and stabilized, with on-site measurements showing a maximum flow rate of approximately 6,651.33 m³·d⁻¹, indicating significant developmental characteristics of the disaster at this section. Measures such as advanced pre-support, segmented reinforcement of karst cavities, enhanced lining, and improved waterproofing-drainage were proposed to ensure safe construction in the disaster-prone section. Specifically, advanced pre-support involved grouted small guide pipes, with their length extended to 6 m. For the karst cavity from ZK24+405 m to ZK24+390 m, a "concrete backfill" measure was adopted; for the cavity from ZK24+390 m to ZK24+375 m, a "bentonite pre-support + protective arch + concrete backfill" approach was used; and for the cavity from ZK24+375 m to ZK24+348 m, a "concrete (shotcrete) backfill" measure was implemented. Lining reinforcement primarily involved "water diversion + pressure-resistant lining", while drainage enhancement primarily utilized a double-layer EVA waterproof membrane semi-encapsulation and C30 concrete self-waterproof structure, supplemented by additional drainage holes and densified circumferential drainage systems. In the prediction results for the ZK24+405 m section, the relative error for arch settlement ranged from 1.36% to 2.08%, and the relative error for horizontal convergence was between 1.63% and 1.80%; In the deformation prediction results of the ZK24+400 m cross-section, the relative errors for vault settlement were 1.78% and 2.14%, while the relative errors for horizontal convergence were 1.73% and 1.91%, respectively. Therefore, both cross-sections exhibited high prediction accuracy in the two types of monitoring items, with convergence trends observed in both categories. Except for the general convergence trend in vault settlement at the ZK24+405 m cross-section, all other deformation trends showed strong convergence. This confirms that tunnel deformation consistently remained within the design-required control limits, validating the effectiveness of various treatment measures.The study conducted through this paper has accumulated experience in the prevention and control of mud and water inrush disasters in karst tunnel areas, and the research findings possess certain practical value.
- karst /
- tunnel /
- developmental characteristics /
- deformation prediction /
- treatment scheme design

HTML

图 1 隧道突泥涌水的现场情况

Figure 1. Photo showing the on-site situation of sudden tunnel water and mud inrush

下载: 全尺寸图片幻灯片

图 2 ZK24+405~ZK24+390 m的溶腔处治示意图

Figure 2. Schematic diagram of karst cavity treatment at ZK24+405~ZK24+390 m

下载: 全尺寸图片幻灯片

图 3 ZK24+398 m处排水横洞设计示意图

Figure 3. Design schematic diagram of drainage cross hole at ZK24+398 m

下载: 全尺寸图片幻灯片

图 4 ZK24+390~ZK24+375 m的溶腔处治示意图

Figure 4. Schematic diagram of karst cavity treatment at ZK24+390~ZK24+375 m

下载: 全尺寸图片幻灯片

图 5 ZK24+375~ZK24+348 m的溶腔处治示意图

Figure 5. Schematic diagram of karst cavity treatment at ZK24+375~ZK24+348 m

下载: 全尺寸图片幻灯片

图 6 拱顶沉降变形曲线

Figure 6. Vault settlement deformation curve

下载: 全尺寸图片幻灯片

图 7 水平收敛变形曲线

Figure 7. Horizontal convergence deformation curve

下载: 全尺寸图片幻灯片

表 1 隧道变形预测结果

Table 1. Results of tunnel deformation prediction

监测周期/d	ZK24+405						ZK24+400
	拱顶沉降			水平收敛			拱顶沉降			水平收敛
	沉降值/ mm	预测值/ mm	相对误差/%	沉降值/ mm	预测值/ mm	相对误差/%	沉降值/ mm	预测值/ mm	相对误差/%	沉降值/ mm	预测值/ mm	相对误差/%
29	75.21	74.19	1.36	44.36	43.58	1.76	63.33	62.14	1.88	40.32	39.55	1.91
30	75.96	74.63	1.75	45.53	44.71	1.80	64.09	62.72	2.14	40.96	40.19	1.88
31	76.35	74.92	1.87	46.02	45.27	1.63	64.98	63.75	1.89	41.35	40.56	1.91
32	77.03	75.43	2.08	46.59	45.82	1.65	65.64	64.31	2.03	41.78	41.01	1.84
33	77.61	76.16	1.87	47.05	46.21	1.79	66.40	65.22	1.78	42.21	41.48	1.73
34		76.59			46.68			65.91			41.97
35		77.06			47.01			66.58			42.43
36		77.48			47.26			67.01			42.71
37		77.95			47.52			67.39			42.89

下载: 导出CSV

表 2 CNN、RNN和LSTM的优化参数

Table 2. Optimization parameters for CNN, RNN, and LSTM

模型类型	模型参数
CNN	结构层数7层，批大小80，核大小为3，单元数64个，丢失率为0.45，学习率0.002，最大迭代次数700次，激活函数为ReLu，优化器为Adam，损失函数为MSE
RNN	结构层数4层，批大小为64，单元数48个，丢失率为0.50，最大迭代次数700次，激活函数为ReLu，优化器为Adam，损失函数为MSE
LSTM	层数3层，批大小为64，单元数48，丢失率0.40，最大迭代次数700次，激活函数为ReLu，优化器为Adam，损失函数为MSE

下载: 导出CSV

表 3 不同模型的相对误差均值/ %

Table 3. Mean relative errors of different models/%

预测模型	ZK24+405		ZK24+400
预测模型	拱顶沉降	水平收敛	拱顶沉降	水平收敛
SSA-SVM	1.79	1.73	1.94	1.85
CNN	3.11	2.95	3.05	3.16
RNN	2.85	2.90	2.66	2.81
LSTM	2.97	2.94	3.02	3.12

下载: 导出CSV

参考文献(17)

[1]	李开富, 彭倩坤, 王伟, 黄斌, 庹洪涛. 岩溶病害预报分析与评价 [J]. 公路, 2025, 70 (6): 479-484. Li Kaifu, Peng Qiankun, Wang Wei, Huang Bin, Tuo Hongtao. Analysis and evaluation of karst disease prediction[J]. Highway,2025, 70 (6): 479-484.
[2]	唐科行, 万川, 谭钢, 陈子龙. 广邻路华蓥山特长隧道岩溶地质问题与对策[J]. 中国岩溶, 2019, 38(4): 480-487. doi: 10.11932/karst20190402 Tang Kehang, Wan Chuan, Tan Gang, Chen Zilong. Karst geological problems and countermeasures of the Huayingshan super long tunnel on the Guang’an-Linshui highway[J]. Carsologica Sinica, 2019, 38(4): 480-487. doi: 10.11932/karst20190402
[3]	蒋英礼, 张彦龙, 王景梅. 基于未确知测度-SPA的岩溶隧道水害危险性评价[J]. 人民长江, 2021, 52(5): 78-85. doi: 10.16232/j.cnki.1001-4179.2021.05.013 Jiang Yingli, Zhang Yanlong, Wang Jingmei. Risk assessment on karst tunnels water disaster based on unascertained measuremen-set pair analysis theory[J]. Yangtze River, 2021, 52(5): 78-85. doi: 10.16232/j.cnki.1001-4179.2021.05.013
[4]	资西阳. 断层破碎带隧道突涌水灾害风险等级评价研究[J]. 公路, 2021, 66(12): 410-416. Zi Xiyang. Research on the risk level evaluation of sudden water inrush disasters in tunnels with fault fracture zones[J]. Highway, 2021, 66(12): 410-416.
[5]	杜义祥, 朱宏伟, 李明骏. 基于AHP-理想点法的隧道突泥涌水风险评价[J]. 公路交通科技, 2022, 39(3): 100-107. doi: 10.3969/j.issn.1002-0268.2022.03.013 Du Yixiang, Zhu Hongwei, Li Mingjun. Evaluation of water and mud inrush risk in tunnel based on AHP-ideal point method[J]. Journal of Highway and Transportation Research and Development, 2022, 39(3): 100-107. doi: 10.3969/j.issn.1002-0268.2022.03.013
[6]	王建, 王丽君, 马建新, 宋书志, 黄鹏. 岩溶地区公路隧道穿越大型溶洞处治措施分析[J]. 地下空间与工程学报, 2020, 16(S2): 964-970,998. Wang Jian, Wang Lijun, Ma Jianxin, Song Shuzhi, Huang Peng. Analysis on treatment measures for highway tunnel passing through karst caves[J]. Chinese Journal of Underground Space and Engineering, 2020, 16(S2): 964-970,998.
[7]	邹育麟, 刘静, 汪波, 陈子全, 谢作栋, 古浩, 王楷越. 四川沿江高速公路隧道涌水突泥灾害成因分析及防治对策研究[J]. 现代隧道技术, 2025, 62(3): 259-269. Zou Yulin, Liu Jing, Wang Bo, Chen Ziquan, Xie Zuodong, Gu Hao, Wang Kaiyue. Study on the causes and prevention measures of water and mud inrush disasters in Sichuan Yanjiang Expressway Tunnels[J]. Modern Tunnelling Technology, 2025, 62(3): 259-269.
[8]	张广泽, 冯君, 崔建宏. 高速铁路隧道岩溶灾害及防治关键技术研究[J]. 铁道标准设计, 2018, 62(11): 108-113. doi: 10.13238/j.issn.1004-2954.201803160003 Zhang Guangze, Feng Jun. Cui Jianhong. Study on karst disasters and key prevention technologies in tunnel engineering of high-speed railway[J]. Railway Standard Design, 2018, 62(11): 108-113. doi: 10.13238/j.issn.1004-2954.201803160003
[9]	张建. 基于蜣螂优化算法-双向长短时记忆网络的隧道软弱围岩变形预测[J]. 地球科学与环境学报, 2025, 47(4): 634-645. doi: 10.19814/j.jese.2024.12051 Zhang Jian. Deformation prediction of weak surounding rock of the tunnel based on DBO-BiLSTM[J]. Journal of Earth Sciences and Environment, 2025, 47(4): 634-645. doi: 10.19814/j.jese.2024.12051
[10]	杨林, 江鸿, 陈培帅, 袁青. 浅埋大断面公路隧道塌方处治措施及效果分析[J]. 现代隧道技术, 2020, 57(6): 207-213. doi: 10.13807/j.cnki.mtt.2020.06.028 Yang Lin, Jiang Hong, Chen Peishuai, Yuan Qing. Treatment measures for collapses of the shallow-buried highway tunnel with large cross-section and its effects analysis[J]. Modern Tunnelling Technology, 2020, 57(6): 207-213. doi: 10.13807/j.cnki.mtt.2020.06.028
[11]	秦鹏飞, 赵亚敏, 宋梦, 王文菁. 隧道突泥涌水致灾机理及防控技术研究[J]. 中国岩溶, 2025, 44(2): 359-369. doi: 10.11932/karst20250213 Qin Pengfei, Zhao Yamin, Song Meng, Wang Wenjing. Research on the mechanism and prevention technologies of mud and water inrush disasters in tunnels[J]. Carsologica Sinica, 2025, 44(2): 359-369. doi: 10.11932/karst20250213
[12]	郭宝德, 张毅, 王新文, 陈志高, 董志明, 蒙彦. 基于决策树模型的岩溶隧道涌水可能性快速判断: 以桂河高速泗顶特长隧道为例[J]. 中国岩溶, 2024, 43(5): 1156-1165. doi: 10.11932/karst20240512 Guo Baode, Zhang Yi, Wang Xinwen, Chen Zhigao, Dong Zhiming, Meng Yan. Rapid assessment of water and mud inrush likelihood in karst tunnels based on decision tree model: A case in Siding tunnel of Guihe expressway[J]. Carsologica Sinica, 2024, 43(5): 1156-1165. doi: 10.11932/karst20240512
[13]	王楷翔, 王晓涛, 宫亮, 黄丹. 考虑时空效应的岩溶隧道围岩稳定性分析[J]. 中国岩溶, 2024, 43(3): 717-726. doi: 10.11932/karst20240311 Wang Kaixiang, Wang Xiaotao, Gong Liang, Huang Dan. Stability analysis of karst tunnel surrounding rock considering spatiotemporal effects[J]. Carsologica Sinica, 2024, 43(3): 717-726. doi: 10.11932/karst20240311
[14]	万飞, 张念, 牛萌萌, 卢佳乐, 王耀耀. 基于RF-AHP模型的富水隧道突涌水风险预测[J]. 人民长江, 2025, 56(7): 148-155. doi: 10.16232/j.cnki.1001-4179.2025.07.019 Wan Fei, Zhang Nian, Niu Mengmeng, Lu Jiale, Wang Yaoyao. Prediction model for water inrush risk in water- rich tunnels based on RF-AHP model[J]. Yangtze River, 2025, 56(7): 148-155. doi: 10.16232/j.cnki.1001-4179.2025.07.019
[15]	毛永强, 张子平, 袁青, 熊齐欢, 陈世豪. 隧道穿越富水破碎带涌水突泥风险评估模型研究[J]. 人民长江, 2022, 53(12): 88-95. doi: 10.16232/j.cnki.1001-4179.2022.12.014 Mao Yongqiang, Zhang Ziping, Yuan Qing, Xiong Qihuan, Chen Shihao. Risk assessment model of water-mud bursting for tunnel crossing water rich fracture zone[J]. Yangtze River, 2022, 53(12): 88-95. doi: 10.16232/j.cnki.1001-4179.2022.12.014
[16]	胡涛涛, 钱兴邦, 涂鹏, 陈建勋, 康志斌. 基于优化支持向量机的高地应力炭质板岩隧道大变形灾变预测[J]. 兰州大学学报(自然科学版), 2022, 58(5): 607-615. doi: 10.13885/j.issn.0455-2059.2022.05.005 Hu Taotao, Qian Xingbang, Tu Peng, Chen Jianxun, Kang Zhibin. High stress carbonaceous shale tunnel based on optimized support vector machine prediction of major deformation disasters[J]. Journal of Lanzhou University(Natural Sciences), 2022, 58(5): 607-615. doi: 10.13885/j.issn.0455-2059.2022.05.005
[17]	赵淑敏. 基于递进复合式组合预测模型的软岩隧道大变形预测研究[J]. 隧道建设(中英文), 2018, 38(7): 1131-1137. Zhao Shumin. Study of large deformation prediction of soft rock tunnel based on progressive composite combined prediction model[J]. Tunnel Construction, 2018, 38(7): 1131-1137.