Dynamic Response Prediction of Underwater Explosion Vessels

In order to ensure the safety of the underwater explosive vessel in service, it is necessary to analyse the dynamic response of the underwater explosive vessel. The dynamic response model based on Classification and Regression Tree is established to simulate the mapping relationship between load and container strain. A reliable and stable model is obtained by cross validation. At the same time, compared with Multi-Layer Perceptron model, CART model is faster and more accurate in training, and the prediction effect is obviously better than Multi-Layer Perceptron model, which further explains the validity of CART model and provides a better method for dynamic response prediction of container.