Analysis of High-Speed Impact Behavior of Al 2024 Alloy Using Machine Learning Techniques

The large deformation of Al alloys subjected to high-speed impact may lead to catastrophic failure of aerospace structural components fabricated using these materials. The present work is focused to analyze dynamic behavior of Al2024 alloy subjected to very high impact velocities using FEA software (LS DYNA) along with machine learning (ML) techniques. The transient impact behavior of the alloy was estimated using modified Johnson–Cook visco-plastic model for a strain rate range of 100–3000/s. The residual velocities of Al 2024 subjected to high-speed impact were estimated through FEA, analytical routes, and with experimental studies. In the present work, five machine learning (ML) algorithms, such as K-Nearest Neighbor (KNN), Decision Trees (DT), Random Forests (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGB) algorithms are implemented to predict the crashworthiness of structures constituting Al2024 alloy. The procedures used for optimization of hyper parameters in each of the above ML models are discussed. The comparative analysis of ML models was made for its predictive accuracy in estimating the dynamic behavior of Al 2024 alloy based on their R2 scores, mean squared error, and mean absolute error. Extreme Gradient Boosting has performed better for the crashworthiness predictions providing least mean squared errors and higher R2 scores compared to other models.