We use the material point methods to overcome many difficulties related to numerical simulation of objects with complex geometries subjected to extreme deformations and stresses. These issues include complex geometries of 3-D printed materials, mesh or material point generation from CT- scanned images, and distortion of elements in typical Lagrangian methods. Different from many other mesh-based methods, in MPM we can use positions of material points to accurately represent the location and shape of an object. The material points can be generated from stereolithography (STL) files commonly used in advanced manufacturing and computer aided design applications. The background mesh in material point method calculations is not required to be body conforming. The efficiency, accuracy, convergence, and convenience of the method are illustrated using examples.
For the pre- and post- processing of the numerical results, we benchmark the significant efficiency improvements of a new material point generation method and then discuss the presentation of numerical results using the STL file format on material surfaces. Traditionally, the material points are generated based on the computational meshes. With the new efficient material point generation technique, the material points can now be generated independent of the mesh. This new capability causes a new issue related to different numbers of initial material points in adjacent cells. An efficient numerical algorithm has been developed to address the issue. With the algorithm, we will also show the freedom gain of using material point methods.
This work is sponsored by ASC, C2, and LDRD programs of Los Alamos National Laboratory under auspices of U.S. Department of Energy.