A prediction method for the crashworthiness of multi-cell tubes based on machine learning

Abstract

This study presents a machine learning framework to predict the crashworthiness of multi-cell tubes. Five distinct cross-sectional designs are selected, and various structural configurations are generated by sampling predefined parameters, which are only used for geometry generation, not for prediction. The training dataset is generated through finite element (FE) simulations. Using the autoencoder, structural features of the voxelized FE simulations are encoded into a one-dimensional latent space. When combined with thickness information, this latent representation provides a comprehensive description of the tube structure. Next, random forest, Support Vector Machines (SVM), and Multilayer Perceptron (MLP) are employed to develop prediction models, and the algorithm with the highest accuracy is selected. The final prediction errors for mean crushing force and peak force are 14.21% and 14.49%, respectively, demonstrating high accuracy. This prediction method reduces the cost associated with evaluating the axial crushing performance of multi-cell tubes and can accelerate the design process of multi-cell tubes.