Study On Surrounding Land Subsidence Monitoring And Prediction Method Of Foundation Pit In Shanghai

With the urban construction becoming increasingly intensive and the depth of underground space becoming deeper,how to ensure the safety of surrounding environment and structure turn into key and difficult points in the process of foundation pit construction.Foundation pit engineering including soil retaining,support,precipitation,digging,monitoring and information construction,etc.should be regarded as a system to study.So it must to monitor,forecast and control the subsidence of surrounding structures and guarantee the safety of the foundation pit and surrounding environment in the excavation process.According to the features of the excavation and surrounding environment in Qiantan area,Shanghai,through the research of calculation theory,numerical simulation and prediction method;research on the predict surface subsidence of foundation pit excavation Combined with field monitoring data calculation and prediction model.in this paper.The neural network models are used to predict surface subsidence around foundation pit excavation,the main research content is as follows:1.The ground subsidence caused by the excavation of the foundation pit is inevitable;this paper expounds the theory and the inducement of the ground settlement around the foundation pit.The subsidence prediction method is divided into the theoretical calculation method based on soil mechanics and the prediction model based on measured data,compared with the theoretical formula,the applicability of the prediction method based on measured static and dynamic fit is better.The current situation and theories of static fitting curve method,grey model and neural network model are analyzed in detail in this paper.Monitoring programme and monitoring data of ground subsidence around the foundation pit are sorted according to the relevant data of foundation pit of in Qiantan area,Shanghai,the deformation law of representative monitoring points is analyzed,The prediction model is established by using hyperbolic fitting method,Hoshino method and grey system theory GM(1,1)model,the fitting prediction accuracy between three predicted and measured values is discussed.PSO-BP and GRNN are used to analyze and predict the subsidence of the surrounding ground caused by the next stage excavation based on time series method.PSO-BP model is used to study on prediction model based on subsidence factors and optimizing the structure of PSO-BP model.The prediction results show that the error between the predicted results and the monitoring data of the three neural network models is within a reasonable range and neural network model can well predict the surrounding subsidence of foundation pit,the prediction results can ensure that the surrounding structure of the foundation pit is in a safe state during the excavation of the foundation pit and the subsequent period of excavation.