Application Of Kalman Filter-BP Neural Network Combination Model In Metro Deformation Monitoring

The main content of engineering deformation monitoring is to use measuring instruments to measure the main body of the building in the field,and further use scientific methods to analyze and process the collected deformation monitoring data,and finally make accurate prediction of the deformation trend of the deformation body.The main purpose is to ensure the safety of Engineering construction.When predicting the predicted value of subway deformation monitoring with a traditional single model,the prediction result often deviates greatly from the expected value.This thesis mainly compares the model of deformation monitoring data prediction of the main construction project of Hugang East Road Station of Metro Line 1 in Hohhot City.Firstly,the modeling and data processing flow of Kalman filter model,BP neural network model and BP neural network model based on Kalman filter are expounded.Secondly,a simple statistical analysis of the collected deformation monitoring values is performed.Finally,the three models are used to analyze and predict the collected data,and the prediction effect is compared and the effect of the combined model is verified.The research work of the thesis mainly includes the following aspects:1.This paper introduces the significance of engineering deformation monitoring work in engineering construction,and elaborates the Kalman filter model and BP neural network model in the traditional deformation monitoring data analysis and prediction model.Aiming at the inaccuracy of single model prediction,which leads to the large deviation between the predicted value and the actual measured value,Kalman filter-BP neural network combined model is proposed.2.Introduce the deformation monitoring project of Hugang East Road of Hohhot Metro Line 1 in this case study,explain the data collection work of the relevant national norms,technical routes and deformation monitoring points for data analysis,the reliability of the data source is ensured.3.The application of Kalman filter model and BP neural network model in metro engineering deformation prediction are studied.Based on the two,BP neural network model based on Kalman filter is adopted.This model can combine the advantages of the two models.In this paper,the combined model is applied to the settlement prediction of the subway project,and the prediction result is better than the single model,which provides a new idea for the settlement prediction of the subway construction project.