Quantifying Discretization Effects on Brain Trauma Simulations
Abstract
Numerical models of the brain are becoming an important tool for developing protective headgear for the Soldier. One nonphysical parameter that is often taken for granted in these models is the material representation by the mesh; arbitrarily formed meshes can propagate error when resolving interactions among the skull, cerebrospinal fluid, and brain. We compared Lagrangian, pure Eulerian, and embedded Lagrangian/Eulerian schemes under both 2-dimensional cyclic rotation and planar blast scenarios. The 3 discretization techniques used differ not only in their finite element implementations but also necessarily in their representations of cerebrospinal fluid. The quantities analyzed in each case were the variations in stress magnitude, spatial distribution, and wave patterns that arise inside the brain. The effects of various time scales, degrees of rotation, and mesh resolutions were compared across methodologies to quantify the solutions and efficiencies of each technique and to understand how their formulations contribute to results. Preliminary results show that the wave patterns and spatial distributions for the rotational simulations are significantly different among the methodologies and that there is much more agreement among the blast loading simulations. Further investigation into specific aspects of several of the models is recommended.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2016
- Accession Number
- AD1001413
Entities
People
- Sophia M. Haire
Organizations
- American Society for Engineering Education