PhD Student
Research Interests:
Meshfree Lagrangian Computational Fluid Dynamics: MPS and SPH
SPH (Smoothed Particle Hydrodynamics) and MPS (Moving Particle Semi-Implicit) are well-known CFD (computational fluid dynamics) methods that are used to solve fluid dynamics problems in mesh free Lagrangian framework. Both SPH and MPS are well suited for simulating large-deformation and fragmentation of free surface. They both use particle system for numerical simulation; however, there are some differences in numerical schemes and equations including pressure calculation and treatment of boundary conditions. In order to see more clearly the general behavior, and the strong and weak points of each method, a comparative study was done using the same test case, with the most similar setting for each method. An in-house program from Offshore System Simulation Laboratory, Texas A&M University, USA is used for the MPS simulations (Kim et al, 2014), while for SPH simulations, an open source SPH solver DualSPHysics program is used (Crespo et al., 2015). The underlying principles behind those strong and weak points, and possible ideas for improvements are also briefly discussed.
Comparison between in-house Moving Particle Simulation (MPS) model and an open source Weakly Compressible (Smoothed Particle Hydrodynamics (WC-SPH) model
Furthermore, to improve the pressure realization on the boundary particles and also to further improve the overall particle behavior, a simplified generalized wall B.C. (Adami et al., 2012) and collision model (Lee et al., 2011) was incorporated into the SPHysics program. The resulting scheme was then tested using various free-surface academic problem such as hydrostatic test, dam break test, sloshing test, and also wave making & dissipation test.
Boundary pressure improvement on SPHysics
Instantaneous free surface profile of the wave maker case
Absorbing boundary condition on a wave making simulation using WC-SPH
Journal Paper:
Proceeding Paper:
Bakti, F.P., Kim, K.S., Kim, M.H., Park, J.C. (2015), “Comparative Study on Particle Method for Numerical Simulation,” Int. Soc. of Offshore and Polar Eng. Proc. vol. 3, 424-431