基于CNN-LSTM复合模型的岩溶隧道突涌水风险等级预测——以五福隧道为例

何军元; 梁迪; 杨桂华; 何世永; 赖中玉; 张朝禹; 王维; 曾美婷; 王升

doi:10.3973/j.issn.2096-4498.2025.06.008

隧道建设 >

2025 , Vol. 45 >Issue 6: 1131 - 1141

DOI: https://doi.org/10.3973/j.issn.2096-4498.2025.06.008

研究与探索

基于CNN-LSTM复合模型的岩溶隧道突涌水风险等级预测——以五福隧道为例

何军元 ,
梁迪 ,
杨桂华 ,
何世永 ,
赖中玉 ,
张朝禹 ,
王维 ,
曾美婷 ,
王升

展开

（1. 重庆中环建设有限公司, 重庆 401120； 2. 重庆交通大学土木工程学院，重庆 400074）

何军元(1984—),男,四川广元人,2009年毕业于西南交通大学,铁道工程专业,本科,工程师,现从事项目管理工作。E-mail： 512797577@qq.com。*通信作者: 梁迪， E-mail： 1132454082@qq.com。

网络出版日期: 2025-06-20

收起

Prediction of Water Inrush Risk Level Based on a Convolutional Neural Network-Long Short-Term Memory Network Hybrid Model for Karst Tunnels: A Case Study of Wufu Tunnel

HE Junyuan ,
LIANG Di ,
YANG Guihua ,
HE Shiyong ,
LAI Zhongyu ,
ZHANG Chaoyu ,
WANG Wei ,
ZENG Meiting ,
WANG Sheng

Expand

(1. Chongqing Zhonghuan Construction Co., Ltd., Chongqing 401120, China; 2. College of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, China)

Online published: 2025-06-20

Fold

摘要

为准确预测西南地区岩溶隧道突涌水风险等级，避免因灾害引发工期延误、经济损失及人员伤亡等风险，基于国内外研究总结岩溶隧道突涌水影响因素，根据西南地区岩溶隧道水文地质特点，选取地层岩性、地形地貌、岩层产状、岩体质量指标RQD、岩层渗透系数、突涌水量、地下水位、年平均降水量、隧道埋深共9个致灾指标构建综合评价体系；提出一种基于卷积神经网络（CNN）和长短期记忆网络（LSTM）复合的深度学习模型；以西南地区多个岩溶隧道施工实例作为样本数据进行训练与验证，并与LSTM和CNN单独模型进行对比分析。结果表明： 1）预测模型对突涌水风险等级Ⅰ的预测准确率达到95%、等级Ⅱ的预测准确率为85%、等级Ⅲ的预测准确率为87%、等级Ⅳ的预测准确率为91%； 2）CNN-LSTM复合模型在预测准确性和泛化能力方面均优于传统方法和单一模型。

关键词： 岩溶隧道; 突涌水预测; 卷积神经网络; 长短期神经网络; 深度学习

本文引用格式

何军元 , 梁迪 , 杨桂华 , 何世永 , 赖中玉 , 张朝禹 , 王维 , 曾美婷 , 王升 . 基于CNN-LSTM复合模型的岩溶隧道突涌水风险等级预测——以五福隧道为例[J]. 隧道建设, 2025 , 45(6) : 1131 -1141 . DOI: 10.3973/j.issn.2096-4498.2025.06.008

Abstract

Accurate prediction of water inrush risk levels in karst tunnels in southwest China is essential to prevent casualties, construction delays, and economic losses caused by such disasters. Based on prior research, the factors influencing water inrush in karst tunnels are identified. A comprehensive evaluation system is constructed using nine indices: formation lithology, topography, rock formation, rock mass quality index, rock formation permeability coefficient, sudden water inflow, groundwater level, annual average precipitation, and tunnel burial depth. A hybrid deep learning model integrating a convolutional neural network (CNN) and a long short-term memory (LSTM) network is proposed. Several karst tunnels in southwest China are used for model training and validation, and the performance is compared with standalone CNN and LSTM models. The results show that: (1) The prediction accuracies of the hybrid model for water inrush risk levels Ⅰ, Ⅱ, Ⅲ, and Ⅳ reach 95%, 85%, 87%, and 91%, respectively. (2) The hybrid model outperforms traditional CNN and LSTM models in terms of both prediction accuracy and generalization capability.

Key words： karst tunnel; water inrush prediction; convolutional neural network; long short-term neural network; deep learning

Options

文章导航

〈

〉

模态框（Modal）标题

摘要

本文引用格式

Abstract