基于GRU神经网络的地下洞室围岩变形预测研究

万晨; 王兴霞; 段杭; 郑龙; 黄建文

doi:10.20174/j.JUSE.2026.02.07

地下空间与工程学报 >

2026 , Vol. 22 >Issue 2: 448 - 458

DOI: https://doi.org/10.20174/j.JUSE.2026.02.07

理论与试验研究

基于GRU神经网络的地下洞室围岩变形预测研究

万晨 ,
王兴霞 ,
段杭 ,
郑龙 ,
黄建文

展开

1.湖北省水电工程智能视觉监测重点实验室,湖北宜昌 443002;
2.三峡大学计算机与信息学院,湖北宜昌 443002;
3.水电工程施工与管理湖北省重点实验室,湖北宜昌 443002;
4.中国三峡建工(集团)有限公司,成都 610095;
5.中国葛洲坝集团三峡建设工程有限公司,湖北宜昌 443000

万晨(2000—),男,武汉人,硕士生,主要从事神经网络技术及应用方面的研究。E-mail: ettalwanc@gmail.com

王兴霞(1980—),女,湖北十堰人,博士,副教授、硕士生导师,主要从事水电工程施工技术方面的研究。E-mail: xxwang@ctgu.edu.cn

收稿日期: 2025-01-18

网络出版日期: 2026-04-28

基金资助

国家自然科学基金(52009069, 51879147)

收起

Study on the Prediction of Underground Cavern Rock Deformation Based on GRU Neural Network

Wan Chen ,
Wang Xingxia ,
Duan Hang ,
Zheng Long ,
Huang Jianwen

Expand

1. Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, Yichang, Hubei 443002, P.R. China;
2. College of Computer and Information Technology, China Three Gorges University, Yichang, Hubei 443002, P.R. China;
3. Hubei Key Laboratory of Construction and Management in Hydropower Engineering, Yichang, Hubei 443002, P.R. China;
4. China Three Gorges Construction Engineering Corporation, Chengdu 610095, P.R. China;
5. China Gezhouba Group Three Gorges Construction Engineering Co., Ltd., Yichang, Hubei 443000, P.R. China

Received date: 2025-01-18

Online published: 2026-04-28

Fold

摘要

为了提高围岩变形预测精度,实时掌握变形状态,预防围岩变形破坏,保障施工安全,针对传统围岩变形预测方法训练效率低、收敛速度慢、泛化能力弱等问题,提出了一种基于GRU神经网络的地下洞室围岩变形时序预测方法,构建了相应的围岩变形预测框架流程。结合白鹤滩右岸地下厂房围岩变形监测数据进行预测,并将其与长短期记忆(LSTM)神经网络算法预测结果进行对比分析。结果表明:GRU神经网络模型能够较好地解决地下洞室围岩变形预测问题,具有结构简单、参数量相对较少、训练及收敛速度快、预测精度高等优势。与LSTM神经网络算法预测结果相比,模型训练时长降幅超过70%,预测误差降低幅度高达50%以上,累计最大变形的相对误差小于0.3%,绝对误差小于0.9 mm的概率高达95%,最大绝对误差仅为2.05 mm。

关键词： 地下洞室; 围岩; 变形预测; GRU神经网络

本文引用格式

万晨 , 王兴霞 , 段杭 , 郑龙 , 黄建文 . 基于GRU神经网络的地下洞室围岩变形预测研究[J]. 地下空间与工程学报, 2026 , 22(2) : 448 -458 . DOI: 10.20174/j.JUSE.2026.02.07

Abstract

In order to enhance the prediction accuracy of surrounding rock deformation, enable real-time monitoring of deformation status, prevent deformation failure, and ensure construction safety, a novel underground cavern surrounding rock deformation temporal prediction method based on GRU neural network is proposed to tackle the low training efficiency, slow convergence, and poor generalization of traditional methods, along with the establishment of a corresponding prediction framework. Utilizing monitoring data of surrounding rock deformation from the underground powerhouse on the right bank of the Baihetan Dam, predictions are made and subsequently compared and analyzed with the forecasting results generated by the Long Short-Term Memory (LSTM) neural network algorithm. The results indicate that the GRU neural network model effectively addresses the prediction challenges associated with underground cavern surrounding rock deformation, offering advantages such as simplified structure, relatively fewer parameters, rapid training and convergence rates, and high prediction accuracy. Compared to the predictions derived from the LSTM neural network algorithm, the GRU model demonstrates a reduction in training duration by over 70%, with a corresponding decrease in prediction error of more than 50%. The relative error for cumulative maximum deformation is less than 0.3%, the probability of absolute error less than 0.9 mm is as high as 95%, and the maximum absolute error is only 2.05 mm.

Key words： underground cavern; surrounding rock; deformation prediction; GRU neural network

参考文献

[1] 裴书锋, 王驭正, 江权, 等. 陡倾柱状节理岩体洞室松弛特征与破坏机制[J]. 地下空间与工程学报, 2024, 20(2): 597-605. (Pei Shufeng, Wang Yuzheng, Jiang Quan, et al. Relaxation characteristics and failure mechanism of steeply dipping columnar jointed rock caverns[J]. Chinese Journal of Underground Space and Engineering, 2024, 20(2): 597-605.(in Chinese))
[2] 熊自明, 卢浩, 王明洋, 等. 我国大型岩土工程施工安全风险管理研究进展[J]. 岩土力学, 2018, 39(10): 3703-3716. (Xiong Ziming, Lu Hao, Wang Mingyang, et al. Research progress on safety risk management for large scale geotechnical engineering construction in China[J]. Rock and Soil Mechanics, 2018, 39(10): 3703-3716. (in Chinese))
[3] 王涛, 孙文龙, 李磊. 基于回归分析与灰色理论的围岩变形组合预测[J]. 地下空间与工程学报, 2017, 13(增1): 48-51. (Wang Tao, Sun Wenlong, Li Lei. Study on surrounding rock deformation combined forecasting method based on regression analysis and grey theory [J]. Chinese Journal of Underground Space and Engineering, 2017, 13(Supp.1): 48-51.(in Chinese))
[4] 谢建雄, 鲁铁定. 一种优化高斯过程回归的隧道围岩变形预测方法[J]. 测绘科学, 2021, 46(4): 50-56. (Xie Jianxiong, Lu Tieding. A deformation prediction method for tunnel surrounding rock based on optimizing Gaussian process regression[J]. Science of Surveying and Mapping, 2021, 46(4): 50-56. (in Chinese))
[5] 杨林德, 颜建平, 王悦照, 等. 围岩变形的时效特征与预测的研究[J]. 岩石力学与工程学报, 2005(2): 212-216. (Yang Linde, Yan Jianping, Wang Yuezhao, et al. Study on time-dependent properties and deformation prediction of surrounding rock[J]. Chinese Journal of Rock Mechanics and Engineering, 2005(2): 212-216. (in Chinese))
[6] 付晓东, 盛谦, 张勇慧. 水电站地下洞室群分步开挖的非连续变形分析[J]. 岩土力学, 2013, 34(2): 568-574. (Fu Xiaodong, Sheng Qian, Zhang Yonghui. Stepwise excavation process of underground caverns of hydropower station using DDA[J]. Rock and Soil Mechanics, 2013, 34(2): 568-574. (in Chinese))
[7] Feng D L, Wu H N, Chen R P, et al. An analytical model to predict the radial deformation of surrounding rock during shaft construction via shaft boring Machine[J]. Tunnelling and Underground Space Technology Incorporating Trenchless Technology Research, 2023, 140: 105321.
[8] 周冠南, 孙玉永, 贾蓬. 基于遗传算法的BP神经网络在隧道围岩参数反演和变形预测中的应用[J]. 现代隧道技术, 2018, 55(1): 107-113. (Zhou Guannan, Sun Yuyong, Jia Peng. Application of genetic algorithm based BP neural network to parameter inversion of surrounding rock and deformation prediction[J]. Modern Tunnelling Technoligy, 2018, 55(1): 107-113. (in Chinese))
[9] 张炎, 朱珍德, 朱姝, 等. 龙滩水电站地下洞室围岩变形的智能化预测方法[J]. 水力发电学报, 2013, 32(2): 246-251. (Zhang Yan, Zhu Zhende, Zhu Shu, et al. Intelligent deformation predicting methods for underground chambers of Longtan hydropower station[J]. Journal of Hydroelectric Engineering, 2013, 32(2): 246-251. (in Chinese))
[10] Li S, Zhao H, Ru Z. Deformation prediction of tunnel surrounding rock mass using CPSO-SVM model[J]. Journal of Central South University, 2012, 19(11): 3311-3319.
[11] Ma C, Tan Y H, Li E B, et al. Allowable deformation prediction for surrounding rock of underground caverns based on support vector machine[J]. Periodica Polytechnica Civil Engineering, 2016, 60(3):361-369.
[12] Pan Y, Chen L, Wang J, et al. Research on deformation prediction of tunnel surrounding rock using the model combining firefly algorithm and nonlinear auto-regressive dynamic neural network[J]. Engineering with Computers, 2019, 37(2): 1-11.
[13] Ye X W, Ma S Y, Liu Z X, et al. LSTM-based deformation forecasting for additional stress estimation of existing tunnel structure induced by adjacent shield tunneling[J]. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research, 2024, 146: 105664.
[14] 吕擎峰, 李钰, 牛荣, 等. 基于深度学习的特殊岩土隧道围岩变形预测研究[J]. 应用基础与工程科学学报, 2023, 31(6): 1590-1600. (Lv Qingfeng, Li Yu, Niu Rong, et al. Research on deformation prediction of surrounding rock in special geotechnical tunnels based on deep learning[J]. Journal of Basic Science and Engineering, 2023, 31(6): 1590-1600. (in Chinese))
[15] 马莎, 肖明. 洞室围岩位移长期预报混沌-神经网络模型[J]. 地下空间与工程学报, 2011, 7(3): 564-569. (Ma Sha, Xiao Ming. The Chaotic-neural network model on long-term prediction of cavern rock mass displacement[J]. Underground Space and Engineering, 2011, 7(3): 564-569. (in Chinese))
[16] 向文俊,陈新,张榴梅.基于KPCA-IPSO-SVM模型的地下洞室围岩变形预测研究[J].水力发电,2023,49(1):58-62. (Xiang Wenjun, Chen Xin, Zhang Liumei. Prediction of underground cavern surrounding rock deformation based on KPCA-IPSO-SVM model [J]. Water Power, 2023, 49(1): 58-62. (in Chinese))
[17] Hou X, Ge F Z, Chen D B, et al. Temporal distribution-based prediction strategy for dynamic multi-objective optimization assisted by GRU neural network[J]. Information Sciences, 2023, 649: 119627.
[18] 徐冬梅, 王亚琴, 王文川. 基于VMD-GRU与非参数核密度估计的月径流区间预测方法及应用[J]. 水电能源科学, 2022, 40(6): 1-5. (Xu Dongmei, Wang Yaqin, Wang Wenchuan. Monthly runoff interval prediction method and application based on VMD-GRU nonparametric kernel density estimation[J]. Water Resource and Power, 2022, 40(6): 1-5. (in Chinese))
[19] 姚程文, 杨苹, 刘泽健. 基于CNN-GRU混合神经网络的负荷预测方法[J]. 电网技术, 2020, 44(9): 3416-3424. (Yao Chengwen, Yang Ping, Liu Zejian. Load forecasting method based on CNN-GRU hybrid neural network [J]. Power System Technology, 2020, 44(9): 3416-3424. (in Chinese))
[20] 王增平, 赵兵, 纪维佳, 等. 基于GRU-NN模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(5): 53-58. (Wang Zengping, Zhao Bing, Ji Weijia, et al. Short-term load forecasting method based on GRU-NN model[J]. Automation of Electric Power Systems, 2019, 43(5): 53-58. (in Chinese))
[21] 辛付宇,邢丽坤,刘笑.基于CNN-GRU神经网络的锂电池SOH估计与RUL预测[J].上海节能,2024(5):819-826. (Xin Fuyu, Xing Likun, Liu Xiao. SOH estimation and RUL prediction of lithium battery based on CNN-GRU neural networks[J]. Shanghai Energy Conservation, 2024(5): 819-826. (in Chinese))
[22] 寇敏,张萌萌,赵军学,等.融合天气特征的WOA-GRU高速公路短时交通流预测[J].公路,2024,69(4):193-199. (Kou Min, Zhang Mengmeng, Zhao Junxue, et al. Short-term traffic flow prediction on highways based on WOA-GRU with integrated weather features[J]. Highway, 2024, 69(4): 193-199. (in Chinese))
[23] Muhammad A, Kamaludin M Y, Benjamin W, et al. Traffic speed prediction using GARCH—GRU hybrid model[J]. IET Intelligent Transport Systems, 2023, 17(11): 2300-2312.
[24] 于德新,邱实,周户星,等.基于GRU-RNN模型的交叉口短时交通流预测研究[J].公路工程,2020,45(4):109-114. (Yu Dexin, Qiu Shi, Zhou Huxing, et al. Research on short-term traffic flow prediction at intersections based on GRU-RNN model[J]. Highway Engineering, 2020, 45(4): 109-114. (in Chinese))
[25] 黎煜昭, 刘启亮, 邓敏, 等. 基于物理约束GRU神经网络的河流水质预测模型[J]. 地球信息科学学报, 2023, 25(1): 102-114. (Li Yuzhao, Liu Qi-liang, Deng Min, et al. A physics-constrained GRU neural network for river water quality prediction[J]. Journal of Geo-information Science, 2023,25(1):102-114. (in Chinese))
[26] 张石磊,朱赵辉,段杭,等.高应力地下洞室顶拱深部变形及受力特性研究[J].地下空间与工程学报,2022,18(1):341-350. (Zhang Shilei, Zhu Zhaohui, Duan Hang, et al. Research on deep deformation characteristics of roof arch and loading variation of unboned pre-stressed cable at large underground cavern with high geostress [J]. Chinese Journal of Underground Space and Engineering, 2022, 18(1): 341-350. (in Chinese))
[27] 杨登科. 不同插值方法对GPS时间序列的影响分析[J]. 全球定位系统, 2019, 44(5): 66-69. (Yang Dengke. Influences of different interpolation methods on GPS time series[J]. GNSS World of China, 2019, 44(5): 66-69. (in Chinese))

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献