AHMED BODY GRID DEPENDECY TEST

• Objective: The main aim of this challenge is to build a geometry model of Ahmed Body and perform a grid dependency test on three different setups, by varying the mesh appropriately and compare the results.
• Ahmed body and its importance

The Ahmed Body was first created by S.R. Ahmed in his research “Some Salient Features of the Time-Averaged Ground Vehicle Wake” in 1984. Since then, it has become a benchmark for aerodynamic simulation tools. The simple geometrical shape has a length of 1044 mm, height of 288 mm, and a width of 389 mm. It also has 50mm cylindrical legs attached to the bottom of the body and the rear surface has a slant that falls off at 20 degrees.

Fig.-Dimensional model of ahmed body Fig.- ANSYS SpaceClaim Model

The Ahmed body represents a simplified, ground vehicle geometry of a bluff body type. Its shape is simple enough to allow for accurate flow simulation but retains some important practical features relevant to automobile bodies. This model describes how to calculate the turbulent flow field around a simple car-like geometry using the Turbulent Flow, k-epsilon interface. The flow for this model is turbulent, which is based on the Reynolds number determined by the body length and inlet velocity. The simulation solves for the turbulent kinetic energy in addition to the velocity field. For this simulation, we need a larger mesh size than what is usually common to resolve the turbulent flow. More specifically, we use a finer mesh downstream of the model to capture the wake zone.

Ahmed body is placed in the following domain, measuring 8.04 x 1.34 x 1.39 meters, to compute the flow field. In order to compute results around the area of interest i.e. ahmed body, I created an enclosure, measuring 2.54 x 0.84 x 0.91 meters to refine the mesh locally and provide inflation layers around the ahmed body, in order get accurate results of simulation over the body.

Fig.-Ahmed body enclosed by two enclosures

Simulations for three different setups are performed in which the mesh is refined locally for each setup and I tried to model the final setup i.e. 3^rd case to have very fine meshing in order to get accurate results than the previous cases. I all the three cases, a Multizone meshing method is used for the outer encloser which creates Hexahedral type of mesh and it is known to be high quality of meshing type for a Box type model. Here, Steady Density-Based Solver has been used with Implicit Solution Method

Boundary Conditions:

Inlet Velocity = 50m/sec

Inlet Temperature = 300k

Outlet boundary condn. = No Slip

• CASE-1

In 1^st case, there is no change in the meshing properties and a regular meshing is done on the model.

Meshing Model:

In the above figure, we can see that the meshing is not done properly for the inside enclosure and the ahmed body walls and this is Coarse type of meshing with relevance = 0.

Meshing Properties:

Results:

Fig.- Residual plot Fig.- Velocity Profile at Cut Plane Z

Fig.- Vector Profile at Cut Plane Z Fig.- Pressure profile

• CASE-2

In 2nd case, Medium type of meshing with relevance = 50 is done for the whole model, which provides finer meshing of the body as compared to the previous case. Here, for the second bounding box and and car wall, a sizing mehtod is used in order to refined a mesh as compared to previous one and break the body into more number of elements because our area of interest is flow simulation over car wall and for that we need to have a close look for the area surrounding the ahmed body. By using sizing method, the second bounding box along with car wall is break into elements of size-0.07m and to have more finer meshing for the car wall, it is break into elements of size-0.03m.

Meshing Model:

Meshing Properties:

Fig.- Vector Profile at Cut Plane Z Fig.- Pressure Profile

• CASE-3

In order to get the most accurate results for the air flow simulation over the ahmed body, I used the Fine type meshing with relevance =100. In case 3, the 2nd bounding box and the walls of ahmed body are break into more finer elements with size of 0.05m and 0.01m. Our aim for this case is to have look into the area of interest and try to refined mesh as much as possible in order to get most accuarate results among all three cases. Here, one of the method of meshing i.e. Inflation is used at car walls with Max. thickness = 0.006m, No. of layers = 5 and Growth rate % = 1.2 in order to get results for the smooth flow of air over ahmed body.

Meshing Model:

Meshing Properties:

Results:

Fig.- Residual Plot Fig.-Velocity Profile at Cut Plane Z

Fig.- Vector Profile at Cut Plane Z Fig.- Pressure Profile

• Why the Pressure Goes Negative?

From the results we can see that the pressure goes negative for major part of the body and especially for the cylindrical four legs. There are many reasons for the pressure to be negative such as:

1. Fluid flows over a body at an inlet velocity, so in order to maintain flow, pressure decreases and this is a simple case of law of conservation of energy.
2. When a fluid flows over a curved surface of the body, flow seperation takes place which results in formation of bubbles and thus due to which the pressure values goes towards negative.
3. The pressure also turns to negative values in order to counter balance the effect of shear stress.
4. Pressure turns negative in a recirculation regions.

• CONCLUSION:

From the above results, we can cleary see the difference between results for all the three cases. Grid Dependency Test allows us to see that how the velocity profile turns from coarse to smoother, from case 1 to case 3. We can clearly see from the figures about the smoothning of the Vector-Velocity profile as we refined mesh locally from case 1 to case 3 and this smooth profile will help us to get accurate results of the simulation of air flow over ahmed body. One more thing we can notice from the figure-\'Velocity profile at cut palne Z\' is that, finer meshing can help us to see wake region more clearly. Thus, it\'s is concluded from above results is that, the more we refine the mesh i.e. the more is break down into number of elements in a proper way, the more accurate results we can get.