Study and analysis of Ahmed body

Fuel efficient cars are in very high demand nowadays, a major factor determining fuel efficiency is aerodynamic drag produced by the vehicle body. The higher the aerodynamic drag the lesser is the fuel efficiency. This increases the demand for the study of vehicle aerodynamics which when done experimentally racks up a huge cost and time. So, in order to study vehicle aerodynamics researchers and engineers turn to computational fluid dynamics(CFD) and this is where the Ahmed body steps in. Ahmed body is a simplified generic vehicle body used by researchers and engineers alike to study vehicular aerodynamics. Since Ahmed body is a popular and the most conventional shape out there, there are vast amounts of experimental data readily available, which is quite useful for CFD engineers to verify results obtained from CFD simulation ensuring that their CFD setup is accurate. That is why Ahmed body is quite important and always in high demand for studies.

In this project, flow over Ahmed body is studied and grid dependency test is performed. The flow is simulated for steady state using a density based solver as the velocity considered here for inlet is 50m/s which is nearly 1/6 th speed of sound. The viscous model considered for the simulation is k-epsilon model with energy eqautions considered for the simulation.To perform grid dependency test velocity is observed at a line made 2m from origin. The solution is set for convergence when the error is reduced to 1e-3.

Ahmed body geometry:

An enclosure is created using enclosure option having a length 2m from the front of the Ahmed body, 5m from the back of Ahmed body and 0.5m from the two sides of the Ahmed body for the unrefined mesh.

Unrefined mesh:

The unrefined mesh has 16563 nodes and 83882 elements.

Mesh metrics for the unrefined mesh:

Convergence plot:

The plot did not converge because of low mesh quality so the solution is stopped and the above plot is taken when the error fluctuation becomes constant

Velocity profile taken from mid cut plane:

Pressure profile taken from mid cut plane:

For the upcoming meshes a second enclosure is created to locally refine the meshes so that definitive results can be obtained from the Ahemed body, the second enclosure has a length of 0.5m from the front of the Ahmed body, 1m from the back of the Ahmed body and 0.26m from the two sides of Ahmed body.

Refined mesh 1:

Multizone method of meshing is adopted here and the local enclosure is further refined in order to capture data more accurately in the region near Ahmed body and inflation layers are added. The mesh thus created has 62532 nodes and 201799 elements.

Mesh metrics for refined 1:

Convergence plot:

Since the mesh is refined now, the solution converged at 640 iterations.

Velocity profile taken from the mid cut plane:

Pressure profile taken from the mid cut plane:

Refined mesh 2:

The mesh is again refined here to push the quality of the results obtained to as close as possible in ANSYS Academic liscence. Here the mesh contains 134297 nodes and 466247 elements close to the 0.5 million elements limit.

Mesh metrics for refined mesh 2:

Convergence plot:

Further refining the mesh, the solution takes more iteration to converge at 1040 iterations.

Velocity profile taken from the mid cut plane:

Pressure profile taken from the mid cut plane:

Velocity at 2m from origin to study grid independence:

Velocity is taken at a line created 2m from origin, one can see here that as the mesh is getting more and more refined the velocity lines are changing, this is because the results are getting affected as the number of elements in the mesh are increased. The process of mesh refinement should be kept on repeating until the velocity lines no longer change so that the results beecome grid independent. But it is not possible to do this in academic liscence as the limit is 0.5 million elements. So results obtained upto mesh refinement possible is attached below:

The pressure goes negative in the back of the Ahmed body and in the top of the front portion of the Ahmed body. The pressure goes negative in the back of the Ahmed body because of the wake region formation which is indicative of the drag generated on the body. The larger the wake region the larger the drag experienced by the vehicle. Therefore researchers try to modify the shape of the body so that less drag is experienced by the vehicle this increases fuel efficiency. Whereas the negative pressure generated in the top of the front portion of the body is due to the flow seperation from the Ahmed body. Engineers try to optimize the flow seperation in this region to improve the downforce for vehicles which increases traction and improves torque.