摘要
The development of a simplified 2-D numerical model was described forwind-driven circulation in reservoir u-sing standard k-ε turbulence model to specify eddy viscositydistribution. The governing equations are transformed and solved on variable vertical grids, whichallows refinement at the surface and bottom boundaries. The results of the model simulation for floware compared with analytical solutions for laminar and turbulent flows, experimental data in awind-flume and wind wave tank. The sensitivity analysis results show that use of large number ofdepth layers increases the accuracy for the bottom counter-current flow. Prediction of surface driftwas not very sensitive to surface grid refinement. The model was also applied to Baisha reservoirfor an assumed wind condition and showed to be able to simulate the general features of surfacedrift and return flow under variable flow depth. The model can serve as alternative means ofstudying wind-driven flow beside experiments. It also reduced the problem complexity associated with3-D circulation models while faithfully reproducing the drift and near bottom return currents.
The development of a simplified 2-D numerical model was described forwind-driven circulation in reservoir u-sing standard k-ε turbulence model to specify eddy viscositydistribution. The governing equations are transformed and solved on variable vertical grids, whichallows refinement at the surface and bottom boundaries. The results of the model simulation for floware compared with analytical solutions for laminar and turbulent flows, experimental data in awind-flume and wind wave tank. The sensitivity analysis results show that use of large number ofdepth layers increases the accuracy for the bottom counter-current flow. Prediction of surface driftwas not very sensitive to surface grid refinement. The model was also applied to Baisha reservoirfor an assumed wind condition and showed to be able to simulate the general features of surfacedrift and return flow under variable flow depth. The model can serve as alternative means ofstudying wind-driven flow beside experiments. It also reduced the problem complexity associated with3-D circulation models while faithfully reproducing the drift and near bottom return currents.
