Project 7 | Flow Separation
Type| Computational Fluid Dynamics (CFD)
Software | Ansys
Description
In this project, Ansys Fluent simulates airflow over different geometrical shapes. Ansys Post, a post-processing user interface, demonstrates that a change in geometric shape can drastically change the aerodynamic behavior.
Approach
For the geometrical shapes, a triangular wedge and an L-shaped wedge are created in Ansys SpaceClaim. SpaceClaim is used instead of Solidworks because of the simplicity of the shapes. Next, the mesh processing step employs a method known as “hard face meshing”, to create a uniformly divided number of cells on the solid surface. Using 2434 cells for the triangular shape and 1300 cells for the L-Shape, Ansys increases the precision of computed flow properties, and decreases uncertainty metrics. The Model for this project is based on the Viscous, Laminar Flow Model. And, for the solution method: a second order upwind for the Spatial Discretization is chosen.
Results
The following results are found. For the triangular shape, after 50 iterations, mass continuity converges at 4.03*10-3 kg/s in the x direction and 2.85*10-3 kg/s in the y direction. Airflow detaches at the corners of the triangle, and attempts to reattach far downstream where laminar flow is reassumed.
The following results are found for the L-shape. After 50 iterations, mass continuity converges at 9.38*10-3 kg/s in the x direction and 6.91*10-3 kg/s in the y direction. Airflow detaches towards the beginning of the shape, and great disturbance is caused at the onset of flow separation, generating turbulence. In both cases, the presence of a solid object caused the air to speed up, and this result can be seen in the Velocity Contour. Finally, in the region between the separating streamlines, pressure reaches an ultimate low, at a staggering -487 Pa for the triangular shape and -363 Pa for the L-Shape. Overall, changes include flow type, aerodynamic forces, and kinematic changes like velocity and acceleration.
References
Tamer A. AbdelMegid a, et al. “Revisiting the Lid-Driven Cavity Flow Problem: Review and New Steady State Benchmarking Results Using GPU Accelerated Code.” Alexandria Engineering Journal, 15 Oct. 2016, www.sciencedirect.com/science/article/pii/S1110016816302800.
Hu, Jianjun, et al. Impact of Time-Splitting Schemes on the Accuracy of FFD Simulations, www.colorado.edu/lab/sbs/sites/default/files/attached-files/c7_impact_of_time-splitting_schemes_on_the_accuracy_of_ffd_simulations.pdf. Accessed 10 Aug. 2023.
Kim, J., and P. Moin. “Application of a Fractional-Step Method to Incompressible NavierStokes Equation.” NASA Technical Memorandum85898, Mar. 1984.
