To further enhance the prevention and control level of sudden and sudden water inrush disasters in tunnels with rich water formations, based on a clear understanding of the influencing factors of water inflow, employs the PSO-RF algorithm to conduct training set predictions for the water inflow at the tunnel faces of two tunnels in the West Chongqing Water Resources Allocation Project.And based onearly prediction results,propose intelligent drainage technology for tunnels in rich water formations to achieve early prevention and control of sudden and sudden water disasters in tunnels, further improving the intelligence level of tunnel engineering construction.The research findings reveal that:Compared to conventional single prediction models such as RF, BP neural networks, and LSTM, the PSO-RF model demonstrates a significant improvement in prediction accuracy, exceeding 37.14%. The R2 values for the predicted water inflow at the tunnel entrance and exit faces are 0.971 1 and 0.952 8, respectively, while the MSE values are 4.324 4 and 8.143 1, indicating that the model has good feasibility for predicting water inflow in tunnels. Furthermore, by integrating the prediction results of water inflow at the tunnel faces, intelligent control of water pressure can be achieved, thereby effectively ensuring the safety of construction and operation of tunnels crossing water-rich strata, and to a great extent, preserving the ecological environment of the tunnel site. The research results can provide technical support for the construction safety of tunnels crossing water rich strata.
Ma Jian
,
Zhang Tianhua
,
Huang Jianjun
,
Xu Jinghua
,
Chen Zheng
. Research on Water Inflow Prediction and Intelligent Drainage Technology for Tunnels in Rich Water Strata[J]. Chinese Journal of Underground Space and Engineering, 2026
, 22(S1)
: 451
-458
.
DOI: 10.20174/j.JUSE.2026.S1.49
[1] 朱合华,凌加鑫,朱梦琦,等. 钻爆法隧道智能建造:最新技术与未来展望[J].现代隧道技术,2024,61(2):18-27.
[2] 陶志刚,赵易,丰于翔,等. 双向地震作用下跨断层破碎带隧道地震响应动力学试验研究[J].中南大学学报(自然科学版),2025,56(2):637-650.
[3] 崔靖奇,吴顺川,程海勇,等.滇中引水软岩隧洞围岩位移时序预测[J].清华大学学报(自然科学版),2024,64(7):1215-1225.
[4] 朱宝强,王述红,张泽,等.基于时间序列与DEGWO-SVR模型的隧道变形预测方法[J].浙江大学学报(工学版),2021,55(12):2275-2285.
[5] Tao R, Peng R, Wang H, et al. Temperature and humidity prediction of mountain highway tunnel entrance road surface based on improved Bi-LSTM neural network[J]. Evolving Systems, 2024,15:691-702.
[6] Zhang P, Chen R P, Wu H N. Real-time analysis and regulation of EPB shield steering using Random Forest[J].Automation in Construction, 2019, 106:102860.
[7] 叶肖伟,张小龙,陈延博,等.基于粒子群优化-随机森林(PSO-RF)算法的盾构隧道施工期管片最大上浮量预测(英文)[J].Journal of Zhejiang University-Science A(Applied Physics & Engineering),2024,25(1):1-18.
[8] 张平,任松,吴斐,等.多源数据融合的深埋隧道岩爆预测方法[J].东南大学学报(自然科学版),2024,54(3):707-716.
[9] Henrik B, Jonatan G, Brian M, et al. Tunnel fire safety management and systems thinking:Adapting engineering practice through regulations and education[J]. Fire Safety Journal, 2024,146:104140.
[10] 张延杰,浦仕江,周辉,等.滇中引水工程安全建设与高效运行关键技术研究若干进展——地下工程[J].岩石力学与工程学报,2024,43(2):333-357.
[11] 李术才,王康,李利平,等.岩溶隧道突水灾害形成机理及发展趋势[J].力学学报,2017,49(1):22-30.
[12] Dohyun P. Numerical investigation on the effect of water leakage on the ground surface settlement and tunnel stability[J]. Tunnelling and Underground Space Technology, 2024, 146:105656.
[13] 刘鹏,唐思勤,陈征.跨积水煤层段小断面隧洞注浆防渗影响行为研究[J].地下空间与工程学报,2023,19(增2):852-858.
[14] 隆能增,任松,吴斐,等.酸性干湿循环下泥质砂岩劣化及声发射演化特征识别研究[J].岩土力学,2024,45(9):2653-2668.
[15] 陈一帆,吴倩,蒋凌,等.基于声信号识别的焊后残余应力处理质量检测方法[J].工程设计学报,2022,29(3):272-278.
[16] 魏博文,柳波,徐富刚,等.融合PSO-SVM的混凝土拱坝多测点变形监控混合模型[J].武汉大学学报(信息科学版),2023,48(3):396-407.
[17] 袁青,陈培帅,钟涵,等. 基于优化FAHP-TOPSIS法的高压富水花岗岩断层涌水预测[J].隧道建设(中英文),2019,39(5):766-774.
[18] 吴治生,张杰. 岩溶隧道涌水影响因素、预测方法及危害分析[J].铁道工程学报,2012,29(11):58-61, 86.
[19] 《中国公路学报》编辑部. 中国隧道工程学术研究综述·2015[J]. 中国公路学报, 2015, 28(5):1-65.
[20] 高洁,郑东,卢腾泽.高水压条件下隧道水压力智能控制技术研究[J].施工技术,2017,46(20):74-77.