An improved framework for UHMWPE-based soft body armor design

Abstract

Design of soft armor used typically by law enforcement officers has conflicting requirements of cost, thickness, and weight. While a trial-and-error approach is likely to yield a design that meets certification requirements, it is unlikely to lead to an optimal solution that balances the three conflicting requirements. In an earlier work, thin shell finite elements were used in conjunction with machine learning. While the results were encouraging, the finite element (FE) models did not capture all the impact behavioral metrics needed to understand and improve the design of the shoot pack. To overcome the deficiencies, two major changes have been made. First, capabilities of a newly developed constitutive material model for visco-elastic, visco-plastic orthotropic materials have been augmented for use in modeling the soft armor shoot pack. The required input is tailored for materials such as the shoot pack, thereby greatly simplifying the input and speeding up the finite element calculations. Second, solid finite elements are used instead of thin shell elements. This change helps capture the behavior that is observed in laboratory ballistics tests more accurately, such as the punch shear phenomenon. Regression analysis is used to assist in tuning the material parameters that cannot be found via laboratory experiments. The final validation of the developed framework is carried out by comparing FE predictions with laboratory ballistic test data. The predicted number of penetrated layers and the back-face signature from the developed framework are close to the experimental values, and qualitatively the damaged shoot pack resembles the tested article.