Machine learning accelerated search for the impact limit of the graphene/aluminum alloy whipple structure

Abstract

Abstract This paper proposes a Whipple structure to enhance the impact resistance of graphene/aluminum alloy composites by varying the interlayer spacing between graphene and aluminum alloy. The increased interlayer spacing provides more deformation space for the graphene to absorb more deformation energy, and enables the formation of a debris cloud from the bullet fragments and graphene fragments, significantly reducing the impact energy per unit area of the next material. The impact limit serves as a critical metric for assessing the impact resistance of the Whipple structure. Based on molecular dynamics simulations, we developed a machine learning model to predict the protection of aluminum alloy, and quickly determined the impact limits of velocity, bullet radius, and interlayer spacing by using the machine learning model. An empirical equation for the impact limit of interlayer spacing was established. The results showed that non-zero interlayer spacing can significantly improve the impact resistance of the hybrid structure; to fully exploit the superior impact resistance of this Whipple structure, the number of graphene layers should be at least 3. Furthermore, at high impact velocities and large bullet radii, the impact limit of the interlayer spacing exhibits a substantial correlation with the number of graphene layers. These results provide valuable information for the design of the impact resistance of the graphene/aluminum alloy composites.