Ballistic impact: predicting penetration depth in UHPC targets with ML models

Abstract

This paper presents the application of Artificial Neural Network (ANN) model for predicting the penetration depth under projectile impact in Ultra High Performance Concrete (UHPC) targets containing steel fibers. Despite the availability of a large number of existing empirical models, the prediction of penetration depth remained inconclusive, partly owing to the phenomenon's complexity and partly due to the limitation of statistical regression. From the results of this study, it is evident that the ANN model is capable of predicting the penetration depth of UHPC more accurately than the other machine learning models (Linear Regression (LR), Decision Tree Regression (DTR), and Random Forest Regression (RFR)) and empirical formulae. The ANN model achieved a lower Root Mean Square Error (RMSE) of 11.68 compared to other machine learning models (RMSE – 16.66 to 19.74) and empirical equations (RMSE – 25.17 to 53.42), when applied to the test dataset. The velocity, impact energy, diameter of the projectile, and thickness of the UHPC targets are the most significant parameters (p-value <5%) for predicting the penetration depth using ANN and MLR models.