DescriptionThis research project will focus on using high performance computing for accelerating the numerical methods for modeling the flows of viscoelastic liquids. The computational schemes developed for studying viscoelastic flows are based on an adaptive finite-volume discretization of the Navier-Stokes equations combined with various constitutive laws for viscoelastic liquids. The numerical method is based on a volume of fluid algorithm for tracking the interface, in case of the presence of a second phase. The numerical framework has parallel support using the MPI library, dynamic load-balancing, and parallel offline visualization. The parallel performance of the code will be tested and improved upon. Adaptivity combined with parallelization are essential components of the numerical framework, since transient computations of viscoelastic flows are often very intensive (due to small timestep and high mesh resolution required to resolve high stress regions) and therefore require a large computational resources.
OrganizationNew Jersey Institute of Technology
DepartmentMathematical Science
Sponsor Campus GridOSG-XSEDE
Principal Investigator
Shahriar Afkhami
Field Of ScienceMathematical Sciences