基于SSBA模型的深基坑支护结构变形预测研究  

Research on the deformation prediction of deep foundation pit support structures based on SSBA model

作　　者：高苏[1] 吴志强[1] 杨宏刚 邵海元 GAO Su;WU Zhiqiang;YANG Honggang;SHAO Haiyuan(School of Civil Engineering,Nantong Vocational University,Nantong 226007,Jiangsu,China;East China Architectural Design&Research Institute Co.,Ltd.,(Shanghai Underground Space and Engineering Design Research Institute),Shanghai 215000,China;Country Garden Real Estate Group Co.,Ltd.,Guangzhou 510623,Guangdong,China)

机构地区：[1]南通职业大学建筑工程学院,江苏南通226007 [2]华东建筑设计研究院有限公司(上海地下空间与程设计研究院),上海200001 [3]雅居乐地产置业有限公司,广东广州510623

出　　处：《水利水电技术(中英文)》2025年第3期212-221,共10页Water Resources and Hydropower Engineering

基　　金：国家自然科学基金青年基金项目(52108380);南通市科技计划项目(JC22022069);南通职业大学科研项目(23ZK05)。

摘　　要：【目的】支护结构的变形对深基坑的稳定性至关重要,为了更准确地预测深基坑开挖引起的支护结构变形,以变形数据的时空特征为切入点【方法】构建了融合斯皮尔曼(Spearman)相关系数、时空卷积神经网络(Spatio-temporal Convolutional Neural Networks)、双向长短期记忆神经网络(Bi-directional Long-Short Term Memory)和注意力(Attention)机制的新型变形预测模型(SSBA)。【结果】试验结果表明:通过参数相关性分析发现基坑变形与开挖深度、开挖阶段呈正相关,与支撑轴力呈负相关,且开挖深度对变形的影响程度最大;对监测数据进行分析发现,支护结构的变形小于规范值,表明内支撑系统有效地限制了墙体的侧向位移,支护结构设计合理。与4种基准模型相比,SSBA模型取得最小的均方根误差和平均绝对误差值,最大的确定系数值,表明该模型能更准确的预测支护结构变形;SSBA模型也能够较准确的预测不同测点的变形值,表明该模型据有较好的泛化能力和可靠性;选取苏州6号线某基坑的现场监测数据进行试验发现SSBA模型能更准确地预测地连墙变形,表明该模型具有较好的普适性。【结论】SSBA模型能较准确地预测基坑支护结构变形,为工程施工提供指导。[Objective]The deformation of support structures is crucial for the stability of deep excavations.To more accurately predict the deformation of support structures caused by deep excavation,[Methods]this paper proposes a novel deformation prediction model,SSBA,which integrates Spearman correlation coefficient,Spatio-temporal Convolutional Neural Networks(STCNN),Bi-directional Long-Short Term Memory,and Attention mechanism.[Results]Experimental result reveal that the deformation of the pit is positively correlated with excavation depth and phase,and negatively correlated with the support strut axial force.Excavation depth having the most significant impact on deformation.Analysis of the monitoring data revealed that the deformation of the support structure is less than the specified values,indicating that the internal support system effectively restricts the lateral displacement of the wall,and the support structure design is reasonable.Compared with four baseline models,the SSBA model achieved the smallest MAE and RMSE values and the largest R2 value,indicating that it can predict the support structure deformation more accurately.The SSBA model can also predict the deformation values at different measuring points accurately,demonstrating good generalization capability and reliability.Through experiments conducted using field monitoring data from a certain foundation pit on Suzhou Metro Line 6,it was found that the SSBA model can more accurately predict diaphragm wall deformations,indicating that the model has good generalizability.[Conclusion]SSBA model can predict the deformation of support structures more accurately and provide guidance for the engineering construction.

关 键 词：深基坑 地连墙变形 参数相关性 深度学习 注意力机制 

分 类 号：TU472[建筑科学—结构工程]