Study On Prediction Model Of Ground Subsidence Caused By Excavation Of Deep Foundation Pit In Soft Soil Area

With the continuous progress of urban-rural integration in our country,the research on the development and utilization of underground space is even more important.At present,there are few studies on the deep foundation pit in Shaoxing area.Through the theoretical research on the deep foundation pit project in Shaoxing area,the development and utilization of urban resources in this area can be continuously promoted.There are many prediction methods for surrounding ground settlement caused by excavation of deep foundation pit and they can be divided into two types.The first category is the traditional theoretical method,which includes traditional algorithms and numerical algorithms.Due to many factors affecting the surrounding ground settlement,Class method is difficult to achieve fast and accurate prediction;the second type is based on measured data back-push settlement of the measured data inversion method,which in turn includes deterministic prediction method and uncertainty prediction method,such methods can be simple and fast Accurately predicting surface subsidence has strong applicability.In this paper,the second type of measured data deduction method,combined with the Shaoxing Guojin Shopping City foundation pit engineering site monitoring data,the excavation of deep foundation pit caused by the process of surface subsidence research.The main research results are as follows:(1)The analytic hierarchy process(AHP)is used to quantitatively analyze 12 risk factors that cause surface subsidence during excavation of deep foundation pit,and the most important five risk factors are summarized,which are geological conditions,distance from foundation pit,groundwater level,Surrounding additional loads and pit depth.(2)Summarize the main calculation methods of surface subsidence prediction around the foundation pit at present.Analyze the commonly used complex linear exponential function model(CLE model),double composite exponential function model(DME model)and normal composite exponential function model(CME model)Traditional mathematical prediction model for comparative analysis.The results show that the DME model is superior to the CLE model and the CME model and has higher prediction accuracy.(3)Based on the traditional model and theoretical analysis,a new high-order composite exponential function model(HCE model)is proposed.The parameters of the new model were determined and analyzed,and the accuracy of the new model was proved by combining the measured data of the project with the uncertainty prediction method of the measured data.(4)Based on the measured data of 30 deep foundation pits,a regression formula is proposed to estimate the maximum subsidence angle ? and the maximum settlement value ymax of the surface subsidence curve.Using these two new formulas,we can predict the value of the maximum settlement point xm and the maximum settlement value ymax of the ground surface more accurately,which can be better applied to verify the accuracy of the new model prediction.(5)According to the project of the foundation pit project of Shaoxing Guojin Shopping Town and the other two project examples,the accuracy of the new model is verified by the deterministic forecasting method of the measured data deduction method.The new model and the traditional Model for comparative analysis.The results show that the new model model(HCE model)is significantly better than the traditional model,with a higher prediction accuracy,the fitting effect is the best.