(1. China Railway Bridge and Tunnel Technologies Co., Ltd, Nanjing, Jiangsu 210061, China; 2. State Key Laboratory of Geomechanics and Geotechnical Engineering Safety, Institute of Rock and Soil Mechanics, Chinese Academy of Sciences, Wuhan, Hubei 430071, China; 3. University of Chinese Academy of Sciences, Beijing 100049, China)
Abstract：Abnormal subsidence of open caisson foundations may induce risks such as delayed sinking and sudden sinking during construction, and thus real-time and accurate prediction of subsidence is essential to ensure construction safety. An intelligent subsidence prediction model is proposed for open caisson sinking based on multi-source data fusion and the Extra Trees Regressor (ETR) algorithm. By integrating multi-source monitoring data, including soil pressure and structural stress, the model extracts nonlinear relationships between data features and subsidence to achieve dynamic prediction. Based on the construction of the super-large open caisson foundation for the north anchorage of the Zhangjingao Yangtze River Bridge’s North Channel Bridge, two single-source prediction models (based on soil pressure on the cutting edges or structural stress) and two multi-source fusion models (feature fusion and decision fusion). The results demonstrate that the multi-source decision fusion model achieves optimal performance, with a root mean square error of 0.003 85 and a coefficient of determination of 0.88. The model requires an average computation time of 1.98 s per prediction cycle, and the average relative error of cumulative subsidence prediction is only 0.19%, which meets both accuracy and timeliness requirements for engineering applications. Sensitivity analysis reveals that the model exhibits high stability and robustness, as its performance remains insensitive to variations in the number of base decision trees. Furthermore, a graded early-warning mechanism is proposed, integrating prediction results with engineering thresholds to establish a closed-loop “monitoring-warning-feedback” management system, providing reliable technical support for risk control during open caisson construction.
徐 安1，魏祥平1，蒋 凡1，严和仲1，董学超2，3*，郭明伟2，3. 基于多源数据融合的超大锚碇沉井基础施工下沉智能预测[J]. 岩石力学与工程学报, 2026, 45(S1): 411-422.
XU An1, WEI Xiangping1, JIANG Fan1, YAN Hezhong1, DONG Xuechao2, 3*, GUO Mingwei2, 3. Intelligent subsidence prediction of super large anchor caisson foundations during construction based on multi-source data fusion. , 2026, 45(S1): 411-422.
[1] 穆保岗，朱建民，龚维明. 大型沉井设计、施工及监测[M]. 北京：中国建筑工业出版社，2015：1–63.(MU Baogang，ZHU Jianmin，GONG Weiming. Design，construction and monitoring of large open caissons[M]. Beijing：China Architecture and Building Press，2015：1–63.(in Chinese))
[2] 郭明伟，董学超，沈孔健，等. 超大沉井基础取土下沉端阻力变化规律研究[J]. 岩石力学与工程学报，2021，40(增1)：2 976–2 985.(GUO Mingwei，DONG Xuechao，SHEN Kongjian，et al. Study on the variation of the bottom resistance during sinking stage of super large caisson foundation[J]. Chinese Journal of Rock Mechanics and Engineering，2021，40(Supp.1)：2 976–2 985.(in Chinese))
[3] 董学超，郭明伟，王水林，等. 基于集成学习的超大型沉井基础下沉倾斜程度预测[J]. 岩石力学与工程学报，2023，42(增1)：3 812–3 822.(DONG Xuechao，GUO Mingwei，WANG Shuilin，et al. Inclination prediction of a super-sized open caisson foundation during sinking process based on ensemble learning[J]. Chinese Journal of Rock Mechanics and Engineering，2023，42(Supp.1)：3 812–3 822.(in Chinese))
[4] 董学超，郭明伟，王水林. 基于LightGBM的超大沉井下沉状态预测及传感器优化布置[J]. 岩土力学，2023，44(6)：1 789–1 799. (DONG Xuechao，GUO Mingwei，WANG Shuilin. Sinking state prediction and optimal sensor placement of super large open caissons based on LightGBM[J]. Rock and Soil Mechanics，2023，44(6)：1 789–1 799.(in Chinese))
[5] 陈培帅. 深厚淤泥层大型陆上沉井施工控制技术研究[博士学位论文][D]. 西安：长安大学，2021.(CHEN Peishuai. Research on key technology of large land caisson in deep silt overburden[Ph. D. Thesis][D]. Xi?an：Chang?an University，2021.(in Chinese))
[6] 周金枝，何 奇. 鹦鹉洲长江大桥北锚碇沉井下沉稳定性分析[J]. 科学技术与工程，2017，17(28)：131–136.(ZHOU Jinzhi，HE Qi. Stability analysis of sinking for north anchorage caisson of Yingwuzhou Yangtze River Bridge[J]. Science Technology and Engineering，2017，17(28)：131–136.(in Chinese))
[7] 杨 智. 超大型沉井基础下沉阻力分布特征及下沉稳定性研究[硕士学位论文][D]. 淮南：安徽理工大学，2024.(YANG Zhi. Research on the distribution characteristics of sinking resistance and sinking stability of super large caisson foundations[M. S. Thesis][D]. Huainan：Anhui University of Science and Technology，2024.(in Chinese))
[8] 蒋 凡，刘 华，岳 青，等. 超大沉井基础施工控制智能感知系统设计研究[J]. 地下空间与工程学报，2022，18(增1)：233–242.(JIANG Fan，LIU Hua，YUE Qing，et al. Design and research of intelligent sensing system for construction control of super-large caisson foundation[J]. Chinese Journal of Underground Space and Engineering，2022，18(Supp.1)：233–242.(in Chinese))
[9] 张计炜. 深厚软土地区超大型沉井下沉行为及稳定性分析研究[硕士学位论文][D]. 杭州：浙江大学，2020.(ZHANG Jiwei. The research on subsidence behavior and stability analysis of ultra large open caisson in deep soft soil area[M. S. Thesis][D]. Hangzhou：Zhejiang University，2020.(in Chinese))
[10] 谭 斌. 沉井施工突沉及偏斜的研究及处理[J]. 施工技术，2009，38(增2)：446–447.(TAN Bin. Research and treatment of sudden settlement and oblique in caisson construction[J]. Construction Technology，2009，38(Supp.2)：446–447.(in Chinese))
[11] 陈保国，贺洁星，骆瑞萍，等. 沉井下沉过程的运动学特性与下沉控制研究[J]. 岩土力学，2022，43(增2)：425–430.(CHEN Baoguo，HE Jiexing，LUO Ruiping，et al. Kinematic characteristics and sinking control of open caisson during sinking process[J]. Rock and Soil Mechanics，2022，43(Supp.2)：425–430.(in Chinese))
[12] 冯兆祥，王 建. 泰州长江公路大桥南锚基础沉降计算研究[J]. 桥梁建设，2010，(1)：49–51.(FENG Zhaoxiang，WANG Jian. Calculation and research of settlement of south anchorage foundation of Taizhou Changjiang River Highway Bridge[J]. Bridge Construction，2010，(1)：49–51.(in Chinese))
[13] 潘 济，于宏伟. 温州瓯江北口大桥中塔沉井取土下沉施工技术[J]. 工程技术研究，2021，6(5)：78–80.(PAN Ji，YU Hongwei. Construction technology for the sinking via soil excavation of the middle tower caisson of Oujiang River North Estuary Bridge in Wenzhou[J]. Engineering and Technological Research，2021，6(5)：78–80.(in Chinese))
[14] 李 鹏. 沪通长江大桥主塔沉井基础施工技术[J]. 工程技术研究，2020，5(20)：76–77.(LI Peng. Construction technology of the open caisson foundation for the main tower of the Hutong Yangtze River Bridge[J]. Engineering and Technological Research，2020，5(20)：76–77.(in Chinese))
[15] 李思阳. 大型沉井基础下沉全过程受力特性及突沉机制研究[硕士学位论文][D]. 成都：西南交通大学，2020.(LI Siyang. Study on mechanical characteristics and sudden sinking mechanism of large caisson foundation during the whole sinking process[M. S. Thesis][D]. Chengdu：Southwest Jiaotong University，2020.(in Chinese))
[16] 孙彦晓. 大型沉井下沉对土体变形影响的研究[硕士学位论文][D]. 南京：东南大学，2019.(SUN Yanxiao. Research of soil behavior with the sinking of large caisson[M. S. Thesis][D]. Nanjing：Southeast University，2019.(in Chinese))
[17] 潘亚洲，王 琛，梁发云. 大型沉井下沉阻力分布特征及相关工程问题[J]. 结构工程师，2020，36(6)：134–143.(PAN Yazhou，WANG Chen，LIANG Fayun. Sinking resistance distribution characteristics of large open caissons and related engineering problems[J]. Structural Engineers，2020，36(6)：134–143.(in Chinese))
[18] 蒋炳楠. 沪通大桥超深大沉井下沉阻力及突沉现场监测研究[硕士学位论文][D]. 成都：西南交通大学，2016.(JIANG Bingnan. Resistance and suddenly sinking monitor research of deep and large open caisson of Hutong Bridge[M. S. Thesis][D]. Chengdu：Southwest Jiaotong University，2016.(in Chinese))
[19] JIANG B N，YU T，PENG W，et al. Analysis of mechanism of sand gushing and sudden sinking of ultra-large deep-water open caisson in sand via field and model tests[J]. Applied Ocean Research，2023，130：103436.
[20] 施 洲，李思阳，杨仕力，等. 超大型沉井基础下沉中后期摩阻力特性及突沉机制研究[J]. 岩石力学与工程学报，2019，38(增2)：3 894–3 904.(SHI Zhou，LI Siyang，YANG Shili，et al. Study on the characteristics of friction resistance and the mechanism of sudden sinking in the middle and late sinking stages of super large caisson foundation[J]. Chinese Journal of Rock Mechanics and Engineering，2019，38(Supp.2)：3 894–3 904.(in Chinese))
[21] 张治成，邓燕羚，郑锋利，等. 深厚软土地区大型沉井突沉行为分析[J]. 地下空间与工程学报，2020，16(3)：933–943.(ZHANG Zhicheng，DENG Yanling，ZHENG Fengli，et al. Analysis on sudden sinking behaviors of massive open caisson in deep-thick soft clay area[J]. Chinese Journal of Underground Space and Engineering，2020，16(3)：933–943.(in Chinese))
[22] 邓燕铃. 深厚淤泥软土地基超大型沉井基础的稳定性及承载力研究[硕士学位论文][D]. 杭州：浙江大学，2019.(DENG Yanling. The research on the stability and bearing capacity of ultra large open caisson foundation in deep silt soft soil[M. S. Thesis][D]. Hangzhou：Zhejiang University，2019.(in Chinese))
[23] 马远刚，刘彦峰，黄 锐. 深厚淤泥层大型沉井突沉机制及预警指标研究[J]. 桥梁建设，2019，49(增1)：33–38.(MA Yuangang，LIU Yanfeng，HUANG Rui. Study of mechanism and early warning indexes of abrupt sinking of large open caisson in deep soft clay layers[J]. Bridge Construction，2019，49(Supp.1)：33–38.(in Chinese))
[24] 施 洲，纪 锋，余万庆，等. 基于集对分析理论的大型沉井基础施工动态风险评估[J]. 东南大学学报：自然科学版，2021，51(3)：419–425.(SHI Zhou，JI Feng，YU Wanqing，et al. Dynamic risk assessment of huge caisson foundation construction based on set pair analysis theory[J]. Journal of Southeast University：Natural Science Edition，2021，51(3)：419–425.(in Chinese))
[25] 董学超. 基于数据驱动的大型沉井基础下沉智能监控研究[博士学位论文][D]. 北京：中国科学院大学，2023.(DONG Xuechao. Investigation on intelligent monitoring of sinking of large open caisson foundations based on data-driven strategies[Ph. D. Thesis][D]. Beijing：University of Chinese Academy of Sciences，2023.(in Chinese))
[26] 蒋炳楠. 超大型深水沉井下沉受力特性及翻砂突沉机制研究[博士学位论文][D]. 成都：西南交通大学，2020.(JIANG Bingnan. Research on characteristics of sinking and mechanism of sand gushing and sudden sinking of a large，deep-water，open caisson[Ph. D. Thesis][D]. Chengdu：Southwest Jiaotong University，2020.(in Chinese))
[27] 周和祥. 深大沉井下沉阻力及其分布特性研究[博士学位论文][D]. 成都：西南交通大学，2019.(JIANG Bingnan. Research on sinking resistance andits distribution characteristics ofdeep and large open caisson[Ph. D. Thesis][D]. Chengdu：Southwest Jiaotong University，2019.(in Chinese))
[28] 朱建民，龚维明，穆保岗，等. 超大型沉井首次接高受力及变形规律初探[J]. 岩土力学，2012，33(7)：2 055–2 060.(ZHU Jianmin，GONG Weiming，MU Baogang，et al. Research on stress and settlement of super-large open caisson during first lifts[J]. Rock and Soil Mechanics，2012，33(7)：2 055–2 060.(in Chinese))
[29] 王正振，戴康乐，苏天涛，等. 大型沉井下沉过程拉应力影响因素分析[J]. 长安大学学报：自然科学版，2024，44(1)：47–57.(WANG Zhengzhen，DAI Kangle，SU Tiantao，et al. Analysis of factors affecting tensile stress of large caisson during sinking[J]. Journal of Chang?an University：Natural Science Edition，2024，44(1)：47–57.(in Chinese))
[30] 董晓朋. 超深大沉井施工期间结构内力试验研究[硕士学位论文][D]. 成都：西南交通大学，2018.(DONG Xiaopeng. Experimental research on structural internal force of the ultra-deep and large caisson during construction[M. S. Thesis][D]. Chengdu：Southwest Jiaotong University，2018.(in Chinese))
[31] SUDHAMATHI T，PERUMAL K. Ensemble regression based Extra Tree Regressor for hybrid crop yield prediction system[J]. Measurement：Sensors，2024，35：101277.
[32] DEVI M S，MATHEW R M，SUGUNA R. Regressor fitting of feature importance for customer segment prediction with ensembling schemes using machine learning[J]. International Journal of Engineering and Advanced Technology，2019，8(6)：952–956.
[33] 中煤长江基础建设有限公司. 张靖皋长江大桥ZJG–A5标段补充勘察岩土工程勘察报告[R]. 南京：中煤长江基础建设有限公司，2022.(China Coal Changjiang Foundation Construction Corporation. Supplementary geotechnical investigation report for ZJG-A5 Section of Zhangjinggao Yangtze River Bridge[R]. Nanjing：China Coal Changjiang Foundation Construction Corporation，2022.(in Chinese))
[34] 中铁桥隧技术有限公司. 张靖皋长江大桥北航道桥北锚碇沉井基础施工监控方案[R]. 南京：中铁桥隧技术有限公司，2022.(China Railway Bridge & Tunnel Technologies Co.，Ltd. Monitor scheme for the construction of north anchorage open caisson foundation of Zhangjinggao Yangtze River Bridge[R]. Nanjing：China Railway Bridge & Tunnel Technologies Co.，Ltd，2022.(in Chinese))
[35] 中交一公局集团有限公司. 张靖皋长江大桥A5标北锚碇沉井下沉专项施工方案[R]. 北京：中交一公局集团有限公司，2022.(China First Highway Engineering Co.，Ltd. Special construction scheme for the sinking of north anchorage open caisson of A5 Section of Zhangjinggao Yangtze River Bridge. Beijing：China First Highway Engineering Co.，Ltd，2022.(in Chinese))
[36] 周小雄，龚秋明，殷丽君，等. 基于BLSTM-AM模型的TBM稳定段掘进参数预测[J]. 岩石力学与工程学报，2020，39(增2)：3 505–3 515.(ZHOU Xiaoxiong，GONG Qiuming，YIN Lijun，et al. Predicting boring parameters of TBM stable stage based on BLSTM networks combined with attention mechanism[J]. Chinese Journal of Rock Mechanics and Engineering，2020，39(Supp.2)：3 505–3 515.(in Chinese))
[37] 宋 莉. 基于最大熵理论的逆强化学习算法研究[博士学位论文][D]. 北京：北京化工大学，2023.(SONG Li. Research on inverse reinforcement learning based on maximum entropy theory[Ph. D. Thesis][D]. Beijing：Beijing University of Chemical Technology，2023.(in Chinese))
[38] 周志华. 机器学习[M]. 北京：清华大学出版社，2016：171–190.(ZHOU Zhihua. Machine learning[M]. Beijing：Tsinghua University Press，2016：171–190.(in Chinese))
[39] 张世殊，王 鹏，张 开，等. 基于全过程信息的多源数据融合隧道超前地质预报方法[J]. 岩石力学与工程学报，2025，44(4)：838–849.(ZHANG Shishu，WANG Peng，ZHANG Kai，et al. An advanced geological forecast method of tunnels based on multi-source data fusion of full-process information[J]. Chinese Journal of Rock Mechanics and Engineering，2025，44(4)：838–849.(in Chinese))
[40] 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.(ZHANG Ping，REN Song，WU Fei，et al. Prediction method of rock burst in deep buried tunnel based on multi-source data fusion[J]. Journal of Southeast University：Natural Science Edition，2024，54(3)：707–716.(in Chinese))
[41] 汪明元，陈松庭，王耿鑫，等. 多源数据融合的土性参数预测方法[J]. 浙江工业大学学报，2024，52(4)：430–436.(WANG Mingyuan，CHEN Songting，WANG Gengxin，et al. Multi-source data fusion method for prediction of soil parameters[J]. Journal of Zhejiang University of Technology，2024，52(4)：430–436.(in Chinese))
2026, Vol. 45 
