This paper investigates the sensitivity of width proximity and mesh grid size to the wake characteristics of Momentum Reversal Lift (MRL) turbine using a new computational fluid dynamics (CFD) based Immersed Body Forc...This paper investigates the sensitivity of width proximity and mesh grid size to the wake characteristics of Momentum Reversal Lift (MRL) turbine using a new computational fluid dynamics (CFD) based Immersed Body Force (IBF) model. This model has been added as a source term into the large eddy simulation (LES), which is developed for solving two phase fluids. The open source CFD code OpenFOAM was used for the simulations. The simulation results showed that the grid size and width proximity have had massive impact on the flow characteristics and the computational cost of the tidal turbine. A fine grid size and large width inflicted longer computational time. In contrast, a coarse grid size and small width reduced the computational time but showed poor description of the flow features. In addition, a close proximity of the domain’s wall boundary to the turbine affected the free surface, the air body, and the flow characteristics at the interface between the two phases. These results showed that careful investigation of a suitable grid size and spacing between the wall boundary and the turbine is important to minimise the effect of these parameters on the simulation results.展开更多
文摘This paper investigates the sensitivity of width proximity and mesh grid size to the wake characteristics of Momentum Reversal Lift (MRL) turbine using a new computational fluid dynamics (CFD) based Immersed Body Force (IBF) model. This model has been added as a source term into the large eddy simulation (LES), which is developed for solving two phase fluids. The open source CFD code OpenFOAM was used for the simulations. The simulation results showed that the grid size and width proximity have had massive impact on the flow characteristics and the computational cost of the tidal turbine. A fine grid size and large width inflicted longer computational time. In contrast, a coarse grid size and small width reduced the computational time but showed poor description of the flow features. In addition, a close proximity of the domain’s wall boundary to the turbine affected the free surface, the air body, and the flow characteristics at the interface between the two phases. These results showed that careful investigation of a suitable grid size and spacing between the wall boundary and the turbine is important to minimise the effect of these parameters on the simulation results.
