Data-Driven Method for Predicting Soil Pressure of Foot Blades within a Large Underwater Caisson

The soil pressure on the bottom surface of foot blades is an important monitoring point during sinking process large underwater caissons. Complex soil-structure interactions occur process, making it difficult to accurately predict blades. Accurate construction processes often rely data from in field. In this study, a data-driven approach used establish relationship between amount caisson and Furthermore, by improving splitting method original Classification Regression Tree (CART) algorithm, single model’s numerical prediction 80-foot pressures realized. improved CART model, multilayer perceptron (MLP), long short-term memory (LSTM), linear regression model are compared through comprehensive multiparameter evaluation method. Finally, article discusses deployment scheme comparing analyzing time period 10 : 00 July 29, 2020, 23 August 7, 2020. experimental results can satisfy engineering demands provide basis for further intelligent control sinking.