Research On Data Processing And Prediction Model Of Foundation Pit Deformation Monitoring

In the construction of foundation pit,the deformation monitoring and prediction of foundation pit structure is of great significance,and has become one of the hot topics in the field of Surveying and Mapping Engineering in China.Combined with the actual deformation of a foundation pit project in Urumqi,according to the actual deformation monitoring data of the foundation pit design project,the combined model method is used to accurately predict the overall deformation trend of the foundation pit,so as to provide technical support and data support for strengthening the safety construction management of the foundation pit,which has important guiding significance for strengthening the construction design and safety construction management of the foundation pit,And it can effectively ensure the safety of foundation pit deformation and normal construction use in the whole process of foundation pit construction.The main contents of this paper are as follows(1)The basic monitoring content of the foundation pit slope deformation is studied.This paper introduces the summary of foundation pit excavation engineering,geological conditions,the selection of monitoring instruments and monitoring measurement methods,the layout of monitoring points and datum points,and the compliance with the requirements of relevant technical standards and specifications in the measurement,etc,The experimental results show that the overall monitoring value of each foundation pit monitoring project is within the control range,and the construction of foundation pit slope supporting structure is in a relatively stable state in the whole construction stage of foundation pit excavation,which ensures the safe excavation and smooth construction of foundation pit.(2)Taking jcd8,the monitoring point with the largest settlement in the monitoring area,as an example,SVM regression model and ARMA model are used to predict and analyze the data of the monitoring point.The first mock exam shows that the prediction results are unstable.Therefore,there is a certain risk when using the prediction results of single model data processing to make decisions.In this paper,the first mock exam method is proposed.Combined with the advantages of the above two single models,a svm-time series combination model is established.Comparing the first mock exam model with the model,the model can overcome the shortcomings of single model prediction,and improve the accuracy of the model prediction,and make the prediction model more stable and reliable.Finally,with the help of Numerical analysis software platform,the monitoring data are preprocessed and the prediction model is established for programming.(3)In order to verify the feasibility of this conclusion,another jcd7 monitoring point with the second largest settlement is taken as an example.SVM model,time series model and their combined models are used to predict and analyze the monitoring point data respectively.The experimental results further verify that the combined model has higher prediction accuracy and smaller error.The research conclusion has a certain reference value for foundation pit engineering under similar geological conditions.