Achievement

Chemical Engineering trainee Nathan Lane reports the development of a tetrahedral meshing scheme which utilizes topologically accurate network models to place nodes in a computational domain a priori. The intelligent placement of nodes gives the ability to coarsen tetrahedral meshes and still preserve connectivity and porosity of the void space. A recent set of Finite Element Method (FEM) simulations at low speeds through samples of sandstone suggest that coarsening tetrahedral meshes based on network models can reduce the degrees of freedom of the system by almost 2 orders with little response to the calculated value of permeability. Therefore, image based FEM model for porous media simulations can yield results equivalent to Lattice Boltzmann models (LBM) often used for pore scale simulation at a fraction of the computational cost with LBM. Nathan is being coadvised by Professors Thompson (Chemical Engineering), White (Petroleum Engineering) and Wilson (Civil Engineering).