Shield construction surface deformation dynamic prediction method based on MIC-LSTM

The invention discloses a shield construction surface deformation dynamic prediction method based on MIC-LSTM, and the method comprises the following steps: 1, obtaining the surface deformation and surface deformation influence factors of a construction region where I shield segment rings are located in a shield construction process; 2, training an LSTM neural network model; and 3, according to the trained LSTM neural network model, predicting the surface deformation of each shield segment in the construction process of the construction area after the I shield segments. The invention is simplein step and reasonable in design, weight assignment of each influence factor of shield construction surface deformation is realized to distinguish the importance degree of each influence factor to the prediction model, and the LSTM neural network is adopted to perform modeling dynamic prediction, so that the prediction precision and efficiency of shield construction surface deformation are improved. 本发明公开了一种基于MIC-LSTM的盾构