Study And Application Of Combined Forecasting Model In Deformation Monitoring Of Foundation Pit

In the recent years,with the continuous progress of the society and the development of engineering construction,a variety of pit engineering growth rapidly.As a complex and variable field in underground engineering construction,the foundation pit engineering is the key point of construction safety,and the safety monitoring and deformation prediction are the important research topics in engineering field.In the construction of foundation pit,the construction of the foundation pit and the surrounding buildings will inevitably change.It is of great significance to study how to improve the accuracy of foundation pit deformation prediction reasonably and effectively.This paper first introduces the manifestation of the deformation of the foundation pit,and then discusses the purpose and requirements of the foundation pit monitoring,the design basis and principle of the monitoring plan,the items and methods of monitoring.Then we introduce the time series model,the gray system model and the BP neural network model for the prediction of foundation pit deformation,and elaborate the construction methods of these models in detail.Aiming at the shortcomings of single model prediction,this paper introduces the combination forecasting model to study its significance and its combination method.In this paper,the time series and BP neural network optimal variable weight combination model are established by parallel method and implemented in the form of algorithm.After analyzing the advantages and disadvantages of the gray system and the neural network,it is found that the two models have a complementary relationship in the case of a small amount of data.The gray model and BP neural network model are established by means of data optimization,and the algorithm flow is described.Finally,the paper introduces the engineering examples,and analyzes the monitoring data with Eviews and the commercial mathematical software MATLAB in the background of the underground road engineering of Zhengdong New District integrated transportation hub.Firstly,ARIMA model,BP neural network model,ARIMA and BP neural network optimal variable weight combination model are used to predict the deformation of foundation pit.Through comparison,the ARIMA and BP neural network optimal variable weight combination model are compared with single ARIMA model and the conclusion of BP neural network model with high.In this paper,the GM(1,1)model is used to analyze the situation,and then the GM(1,1)and BP neural network model are used to predict the data.The combined model is combined with a single GM(1,1)model.The results show that the combined model of GM(1,1)and BP neural network is more accurate.It can be inferred that the trend of deformation can be judged by the combination forecasting model,which plays an important role in ensuring the safety of the project.