Application of Grey Theory in Prediction of Ground Settlement in Foundation Pit

Abstract

The grey system is a mathematical method to solve the incomplete information system; this model is often combined with BP neural network in foundation pit engineering to produce a new model. Among them, GM (1, 1) model is widely used in foundation pit engineering. In the process of subway construction, the foundation pit is often the most important one of them, therefore, this paper takes a subway station as an example for surface subsidence analysis, and selects GM (1, 1) model and kalman filter as the basic prediction model, programs with MATLAB, and predicts the results respectively, compares the results with the actual values to analyse the good results of the prediction. Through comparative analysis, it indicates that the prediction results of GM (1, 1) in surface settlement prediction are ideal and more suitable for practical engineering.