Executive Summary : | The research focuses on the drying of capillary porous media using Pore Network Modeling and Multiphase Lattice Boltzmann Method (LBM). The project aims to develop a parallelized LBM for the application of drying porous media, which requires parallel computing devices like HPCs and GPUs. The methodology is divided into seven parts:
1. Development of the Double Distribution Function (DDF) modeling for multiphase LBM. The multiphase model is built on an object-oriented platform for user-friendly utilization and incorporates various collision models, stencils, and forcing schemes.
2. Scaling up computation features of the LBM tool. The tool is efficiently parallelized using MPI programming with non-blocking domain decomposition method, allowing researchers to choose the appropriate parallel computing hardware based on their assessment.
3. Implementation of Solute transport (DDF) feature in the LBM tool. This feature plays a major role in reactive flows and natural phenomena like salt precipitation and dissolution.
4. Implementation of Heat transfer (DDF) feature in the LBM tool. This feature allows for simulation of any coupled phenomenon in fluid flow, with non-isothermal drying of porous media as a principal candidate for validation.
5. Development of Hybrid LBM Models. The LB-FD model incorporates the advection-diffusion equation with finite difference method, while the LB-PN model is a switching model between the PNM and LBM for faster multiphase simulations.
6. Implementation of particle dynamics in the model. The fully developed model is versatile for various applications of multiphase flows in porous media, including CO2 sequestration in porous rocks. |