Deformation Analysis Of Foundation Pit Excavation Based On Grey Markov Model Prediction

Accurate prediction of deep foundation pit deformation in underground engineering is of great significance to the safety prevention and treatment of underground engineering construction.At present,there are many prediction methods for deep foundation pit deformation.Based on the foundation pit monitoring project of Zhou Pu Station of Shanghai Rail Transit Line 18,this paper proposes a metabolic grey prediction model based on particle swarm optimization and Markov optimization.The prediction model is tested by residual error,relative error,average relative error and posterior error,and the accuracy of the model is evaluated accurately.The main work of this paper includes the following points:(1)The grey system theory and Markov chain are systematically introduced,and the correlation of each index in deep foundation pit monitoring data is analyzed.The correlation between subsequence and parent sequence is reflected by point correlation degree.(2)The GM(1,1)model is established to predict the monitoring points,which has good adaptability to the exponential data series,but poor prediction accuracy for the data with large shocks.(3)The grey GM(1,1)prediction model is optimized by Markov chain,so that the optimized model can reflect the vibration characteristics of data,and the accuracy of the model is tested by residual error,posterior error and small error probability.(4)On the basis of grey Markov prediction model,combined with MATLAB calculation software programming,the relative error sequence parameters are whitened by particle swarm optimization,and the grey Markov prediction model based on particle swarm optimization is established.On this basis,the metabolism grey Markov prediction model is proposed.The model is applied to the foundation pit deformation of Zhou Pu Station of Shanghai Rail Transit Line 18,and the analysis results show that the accuracy of the model is higher than that of grey Markov prediction model.Comparing the predicted values of the three models with the measured values,the results show that the grey Markov forecasting model optimized by particle swarm optimization is better than the single GM(1,1)forecasting model,while the metabolic grey forecasting model is better than the GM(1,1)forecasting model optimized by particle swarm optimization,which shows that the metabolic grey Markov forecasting model proposed in this paper has certain effectiveness in predicting the deformation of deep foundation pit of underground engineering.