Study On Deformation Monitoring And Prediction Of Deep Foundation Pit

The safety and order of deformation monitoring of deep foundation pit is the precondition of deformation analysis and prediction,which is important in the process of information construction.Deformation prediction of deep foundation pit is the ultimate purpose of deformation monitoring,and the data obtained from prediction is an important reference for safety construction decision.Therefor e,it is an important method to establish appropriate deformation monitoring model that ensures the safety of foundation pit construction.Based on the analysis and prediction of the monitoring data,the results obtained from the reference are valuable to the construction decision.Against the background of Changsha Metro Line four Yingwanzhen station engineering,with the combination of theoretical research and field test research,the grey prediction model and BP neural network model are used to predict the future development of deformation,and the error sources of the two models are analyzed.Then the particle swarm optimization algorithm is used to optimize the two models,and the improved model is used to predict the monitoring data.Finally,the predicted results are compared and analyzed.The main contents of this paper are as follows:(1)The basic principle and modeling method of grey system theory are introduced in detail,and the modeling process of GM(1,1)model is deduced in detail.The error sources of GM(1,1)model are analyzed systematically.Finally,the parameters of the GM(1,1)model are optimized by using the global optimization of particle swarm optimization,and the optimized PSO-GM(1,1)model is proposed.(2)In this paper,the artificial neural network and BP neural network are introduced in detail,and the error back propagation algorithm is theoretically deduced.The shortcomings of BP neural network are pointed out.Finally,the weights and thresholds of BP neural network are modified by the particle swarm optimization algorithm,which is a global search algorithm,and the PSO-BP neural network model is proposed.Combined with the advantages of grey prediction and BP network,a PSO-GM-BP combination model is proposed and used to predict.(3)By comparing the results of several models,we can draw the conclusion that GM(1,1)model and BP neural network model can predict accurate results.The prediction of the two models based on particle swarm optimi zation algorithm has higher accuracy and stronger applicability,the prediction effect of PSO-GM-BP model is the best,which can effectively guide the construction of foundation pit engineering.