Week 4: Project - Steady state simulation of flow over a throttle body

AIM :- Steady state simulation of flow over a throttle body.

THEORY AND INTRODUCTION :-

In automotive application, modelling of the various singularities in pipe systems still remain challenging. The aim of this study is to develop a new throttle valve model which can be included in a one dimensional engine simulation code. First experimental study was conducted to evaluate an existing literature model and results show draback in this model. Then after a new model was develop based on the experimental study and CFD analysis, with the throttle valve model as a function of flow charecterastics and geometrical parameters. After including a new model in the engine simulation code, the engine performance at partial load was determined and results show good agreement with the experimental study.

Most of the spark ignition internal combustion engines operate with a stoichiometric air gasoline ratio since it enable the catelyst to control emissions level. To control engine power, a throttle valve is used. The objective is to control air and consequently gasoline mass flow. The internal combustion engine design is often obtain by simulating engine behaviour at full load operation. However, this method has two drawbacks, the first is that full load rarely used in real vehicle applications, and the second is that inlet and exhaust system are better designed if the mass of the in-cylinder air-gasoline mixture is correctly evaluated. This last point depends upon the pressure waves in the inlet and exhaust manifold, with pressure waves being simulated using one-dimensional gas dynamic codes (Benson, 1982).

SOFTWARE USED FOR THIS PROJECT :-

1. CFD software, Converge studio and Converge science v3.0
2. Cygwin or basic command prompt
3. Paraview to show the animation

IMPORTANT STEPS TO INVOLVED WHILE SOLVING THIS CHALLENGE :-

1. We first download required geometry from the website of skill-lync (which is solid work model in STL format).
2. Set the case setup according to the given problem in the challenge. As this steady state challenge we have set the case as steady state challenge.
3. Now we will check the given case for any problem and the export it filesfor running it into cygwin.
4. Cygwin will create the output files of the case by solving them.
5. Now we will open paraview software and open the case there. We will use glymph and streamline option to visualise and post process the results more clearly.
6. Now by using converge software line plotting option, we will plot the graphs for velocity, pressure, mass flow rate and total cell.

CASE SETUP :-

Case setup for the given challenge are as follows-

1. Materials :-

Here I consider air as a default predefined mixture by which we can test the different gasses with air.

2. Global transport proparties :-

This is the default predefined global transport parameters. Several two and three dimensional flow examples are presented below to access the performance of the purpose variable turbulent Schmidit and prandtl number models. All computations were performed using the commercial code CFD++ which is a TVD-based second order upwind finite volume unstructured flow solver. Two anisotropic turbulence closers were employed. A cubic k-epsilon model and a second moment closure. All computations were performed on fine enough grids to produce numerically reliable and well converged predictions.

3. Reaction mechanism :-

Here is the simple reaction i.e. only nitrogen and oxygen where oxygen is 23% and nitrogen is 77% which is normally. Two reaction means wo species.

4. Run parameters for solver :-

In this challenge I used steady state solver because throttle valve inside the pipe and inside the pipe have constant flow. We dont need to test case so that we have to use Full hydrodynamic simulation mode.

5. Run parameters foe Misc :-

Here using the case solve momentum, solve species and steady state monitor and uncheck the case of use shared memory so that to create the inputs.in file.

6. Simulation time parameter :-

Here defined some specific conditions-

Start time = 0 cycle

End time = 15000 cycles

Initial time step = 1e-9 second

Minimum time step = 1e-9 second

Maximum time step = 1 second

Maximum convection CFL limit = 1.

7. Navier stroke solver in solver parameters :-

Here I use PISO method. PISO involves one predictor step and two corrector steps and is designed to satisfy mass conservation using predictor-corrector steps.

Advantage and disadvantage of PISO methods :-

1. Generally gives more stable results and takes less CPU time but not suitable for all processes.
2. Suitable numerical schemes for solving the pressure-velocity linked equation.
3. For laminar backward facing step PISO is faster than SIMPLE but it is slower conserning flow through heated fin.
4. If momentum and scalar equation have weak or no coupling then PISO is better than SIMPLEC.

8. Misc in solver parameters :-

Here I use 0-No strict conserve.

9. Momentum equation in solver parameters :-

In momentum equation SOR solver type is used with No preconditionar.

10. Pressure equation in solver parameters :-

In pressure equation Converge BICGSTAB solver type is used with No preconditioner.

11. Defined boundary :-

Here I defined all boundary conditions with specified region names.

12. Elbow wall boundary :-

For steady state wall motion is stationary type and boundary is wall type. Law of wall UDF is used with zero normal gradient.

13. Outflow boundary :-

This is the outlet boundary with outflow type where pressre is 1 bar and velocity boundary type is NE. Air as a species also added in it.

14. Inflow boundary :-

This is the inlet boundary with 1.5 bar pressure is usedand UDF boundary is NE, by adding air as a species.

15. Throttle valve boundary :-

Throttle valve is also wall type boundary with temperature boundary conditions is NE.

16. Region and initialization :-

Here, I define the region for the boundary with atmospheric pressure and room temperature.

17. Turbulence modelling :-

Here, I used by default RNG k-epsilon turbulence model and it's specifications.

18. Base grid :-

Here I used 1mm as a base grid in X, Y and Z directions. By reffering this grid I use Fixed embeding as follows.

19. Fixed embeding :-

Here, I used Permanent mode embeding on throttle valve with scale factor as 3 and embed layers are 2.

20. Output files :-

This is the output files as-

Time inerval for writting 3D output data files =1000

Time interval for writting restarting output = 1000.

21. Grid control :-

For grid control Fixed embeding are used for refining the mesh quality of elbow and throttle valve.

22. Steady state monitor :-

This is the steady state monitor, by which monitor variable is AVG_velocity, targeted on output boundary type . with

Monitor start delay = 200 cycles and

sample size = 100 cycles

Minimum sample size = 0.0001 seconds.

23. Post convert files :-

Here I post converted the files for Paraview window and it require 19 seconds.

RESULTING PLOTS :-

Base grid size = 1e-3m along X, Y and Z directions.,

1. Mass flow rate for inlet :-

The mass flow rate plot for inlet is somewhat straight in line over all cycles, It means no more changes occurs while mass flow at inner side.

2. Mass flow rate for outet :-

Here have the more comparable difference for mass flow rate, from inlet to outlet mass flow rate increases slowly but at outlet it increases tremendously because there have a more compressible mixtures at outlet. Mass flow rate increases it means the rate of change of velocity increases. The formula for mass flow rate is-

m = density*velocity*Area = constsnt, in Kg/s.

This is also called as continuity equation.

3. Pressure contour plot :-

This is the pressure contour plot which increases linearly. The formula for pressure is-

Pressure = Force/Area, in N/m^2

Here if any twoquantity known then we can calculate third quantity easily. The shape of plot is like staircase.

4. Pressure contour for surface with edge :-

This is the pressure contour for surface with edge, Here we can see that the quality of mesh which is fully refined.

NOTE :- In laptop tochpad there have no option for panning so that we can not zoom specific section and watch carefully that's why I can not showing the mesh near the throttle valve by zooming. At inlet pressure is lower but at outlet it is higher.

5. Pressure plot in converge :-

This is the pressure plot in converge studio v3.0. It increases at start but later on it decreases slowly by increasing number of cycles.

6. Pressure contour for surface :-

This is the pressure contour for surface which is same as that of pressure contour at surface with edge, the only difference is to meshing. Here we can show that only surface and there showing only edges.

7. Total cell count plot :-

This is straightline over the all cycles. It increases as quality of refined mesh increases. The number of cycles increases as mass flow rate and other factors increases.

8. Velocity contour plot :-

This is the velocity contour plot, which decreases due to pressure increases. The pressure decreases gradually because of higher rate of velocity increases.

9. Velocity contour for surface with edge :-

This is the velocity contour for surface with edge, here velocity decreases from inlet to outlet because pressure rate increases.

10. Velocity contour for surface :-

This is the velocity contour for surface same as that of surface with edge. At throttle velocity is minimum and at inlet it is maximum.

11. Pressure animation :-

12. Velocity animation :-

CONCLUSIONS :-

From above detail results we can conclude that-

1. Pressure is increases from inlet to outlet,

2. no more change in mass flow rate at inlet but at outlet it increases gradually, also

3. the pressure does decreases at every cycles because pressure and velocity are inversaly proportional to each others.

4. Total cell count increases as the quality of refine mesh increases.

5. Embeding quantity is no more consider than grid size, so that there have no create more difference.

6. Creating animation files for pressure and velocity to showing clear results.

In this way, I set up the results for steady state simulation for flow over throttle body, Post processing the results and showing the pressure and velocity contours, Showing the refined mesh quality, Showing the plots for pressure, velocity, mass flow rate and total cell count, Also creating animation file for pressure and velocity contours.

In this way, I successfully completed this project by explaining each plot and terms used in setup and showing the clear results with the help of animation files. Setting up the case by considering the steady state simulations and showing that what is the effect does on throttle valve.