Ahmed Body

Ahmed body and its importances

Reducing the global consumption of vehicle fuel is a crucial and urgent necessity if we are to respond swiftly and appropriately to climate change. Research on drag reduction which enhances fuel efficiency, thus represents a very significant environmental concern. To reduce aerodynamic drag effectively, we surely need a comprehensive understanding of the flow structure around the vehicle. The air drag surrounding an automobile is mainly consituted of pressure drag and friction drag.

CFD simulations are an efficient tool in automotive industry used extensively for design. This allows the designers to obtain a quick and proper analysis before go ahead to the final step. Ahmed car body is the standard model that can be used as validating case in industry and simulation. Ahmed Body is a generic car type bluff body with a slang back has a different angles from 0 - 40 degrees. It is used as a benchmark test case for simulating the external aerodynamic and flow characteristics over a car model. CFD analysis in a car industry is used in the determination of the applied forces and that of the vehicles wake duing moving, and often used to analyse the effect of the wake on the vehicle\'s efficiency and compatibility in comparision with other vehicle\'s.

AIM & OBJECTIVES:

Aim of this project is to refine the mesh around the area of interest i.e. locally. Perform the grid dependency test on this setup by varying the mesh appropriately ( 3 set-ups) and try to bring the final setup as close to the academic limit as possible.

Geometry

Taking air as a working fluid, we have

• Density = 1.225 kg/m^3
• Cp = 1006.43 J/kg-K
• Inlet Velocity = 50m/s
• Thermal conductivity = 0.0242 W/m-K
• Dynamic Viscosity = 1.7894e-05 kg/m-s

Assumptions:

• There is No heat transfer between the flow and the geometry.
• Air Velocity at the inlet section is constant value duringthe solution.
• All the object boundaries will be dealt as walls with noSlip shear conditon.
• Because of Mach Number is less than 0.3, the flow is being considered as incompressible in nature.

1. Baseline setup

Here, as per the lecture the enclosure around the Ahmed body has been created and its mesh diagram has been shown below

The cell count of the above mesh is 84443

The element size = 413.6 mm

Setup and results:

The Energy equation has been turned on because we are using the k-epsilon model for solving its continuity and momentum equations.

residual plot:

Velocity plots and Velocity vectors:

It is observed that, the wake region is formed near the exit of the Ahmed Body, it is due to the computational meshes are very coarse hence adaptive meshing technique should be considered. So, lets add local regions of refinement across the Ahmed body and region of wake as well.

1. Refined Mesh I

I have created the second enclosure of the following size

The Following refinement has been done on the Mesh properties:

The cell count of the above mesh is 200476

The element size = 100 mm

Setup and results:

It is observed that, the wake region is formed near the exit of the Ahmed Body, it is due to the computational meshes are still very coarse hence more meshing refinement technique should be considered. So, lets add local regions of refinement across the Ahmed body and region of wake as well.

Refined 2:

The cell count of the above mesh is 334443

The element size = 75 mm

Setup and results:

Post Processing:

It is observed that, the wake region is formed near the exit of the Ahmed Body, it is due to the computational meshes are still very coarse hence more meshing refinement technique should be considered. So, lets add local regions of refinement across the Ahmed body and region of wake as well. Due to the restriction of licencing I have made the mesh count close to 0.5 million only.

Refined 3:

The cell count of the above mesh is 530034

The element size = 50 mm

Post Processing:

• There are some notable changes in above cases.
• The no of meshes not only increased from case 1 to case 3 but also it has been refiened in the targeted areas such as The local enclosures, boundary layers, Inflation layers at the support of Ahmed body etc from the baseline setup.
• The reason for doing it because the finer mesh provides a smoother and accurate values.
• In the finer setups, it is easier to visualization the contours and behaviours of variables.
• The velocity contours indicates that the key to drag reduction is to weaken (if not eliminate) the longitudinal vortex created at the side edges of the rear slant.
• But of course it will increases the computational times as well.

Grid Independent Test:

A CFD solution can neverbe trusted unless we check whether the results depends on the grids or not, or else we have any experimental data. The output results of a coarser mesh and finer mesh can be neither be same. So upto what extent, we need to vary the mesh to get the expected level of tolerance can be found out from Grid Independent test.

This is done by varying the mesh size from coarse to fine and checking the output result for each mesh. When varying the mesh does not affect the result much, then we can stop and select that minimum mesh size for our final solution updated. For our case of above scenario, the residuals have been constant from 798 iteration. In most of the industiral and scientific uses we can check it by PARAMETRIC ANALYSIS as well, which can be done by creating separate parameter sets for meshes and desired results/outputs and exporting the calculation into csv files, plotting graphs and analysis, which is little tedious process, but nevertheless which would shows similar results as for this setups.

Q. Variation of Pressure, why it goes negative?

Pressure variation of refined 2 setup

Pressure variation of refined 3 setup:

As already I have mention in the Ahmed Body importances in the first section, an Ahmed Body is a bluff model vehicle type in which upto 85% of total drag is pressure drag and remainder is friction drag and it is also true that the rear end contibutes as much as 91% to total pressure drag and is especially predominant at high speeds (i.e. at high Re), as confirmmed by Hucho and Svoran (reference).

• Its front is designed to be blunt so that the flow cannot separate and aerodynamic forces depend instead largly on the flow structure created on the rear.
• For our level of understanding required for this challenge, we should know from calculations and observations the drag generated by flow separation is small so the total drag is decreased. (of course, details of explanations can be found on many publications on internet)
• The absolute pressure value decreased on the slant surfaces, because the longitudinal vortices are the region of low pressure which contibute to overall pressure drag.

P.S. A better understanding and results can be get from various level of computations with different slant angles, incoming velocities, road with different frictions etc.