Challenge

Assignment 7

Explore tutorial number 9 and write a detailed report 
FRM corresponds to fast running models. This done by decreasing the time steps the GT Suite uses to solve individual components. In order, to know which components are the ones that consumes maximum amount of time we basically start with the high fidelility model, run it and then go to RLT Contour Map ->
Pipes/FlowSplits -> Flow Control -> "Minimum Courant Time Step (Explicit Circuits Only) in GT POST.

 The following graph would appear . The components with the lowest number would be the most time conuming ones...
So in order to start which components we could edit in order to decrease the run time of the simulation and bring the time closer to the real run time.

To Start, go to the tools in the GT ISE and the FRM Tags to is the components that need to be more time efficient.
Afterwards , got o the FRM converter to convert the high Fidelility Model to FRM.

Now the strategy to convert the model is simple the FRM converter show you the FRM Tags parts that could be made more efficient.
We would be choosing these segments one by one and the commanding the FRM converter that do we want to Simplify for Accuracy or Speed. Generally, a segment is first caliberated for ccuracy and afterwards if the timespet is furthe wanted to be shorted then further it is caliberated for the speed.

this would lead to form parameters that would be caliberated for usually one case and then the parameter value that is optimized is used for other cases ...

After completing one segment for example "EXHAUST MANIFOLD" we move to the next segment "EXHAUST PIPES" and repeat the above list process.

This process can be repeated for other segments as shown in the tutorial.  
After hitting the FRM Conveter you would be asked to input results of the parameters and their allowable tolerances that if crossed showed be highlighted in red color when running the model after caliberation.
As above listed strategy for FRM Converter , Step1 is about exhaust manifold , so the procedure would be  as follows:

1- Simplify for Accuracy

2- Parameter would be generated i.e. discretization length" and "heat transfer multiplier", Give an initial guess to the this paramteer in the case setup.

3- Caliberate for on case. 

4- Extend this parameter values for all cases by running the all cases after caliberation process.
  
As above listed strategy for FRM Converter , Step2 is about exhaust pipes , so the procedure would be  as follows:

1- Simplify for Accuracy

2- Parameter would be generated i.e. Diameter of Exhaust Pipe", Give an initial guess to the this parameter in the case setup.

3- Caliberate for on case. 

4- Extend this parameter values for all cases by running the all cases after caliberation process.
  
As above listed strategy for FRM Converter , Step3 is about intake manifold , so the procedure would be  as follows:

1- Simplify for Accuracy

2- Parameter would be generated i.e. Heat Transfer Model", Give an initial guess to the this parameter in the case setup.

3- Caliberate for on case. 

4- Extend this parameter values for all cases by running the all cases after caliberation process.

5- Check to see if any reults are outside the accuracy tolerances
  
As above listed strategy for FRM Converter , Step4 is about boost pipes , so the procedure would be  as follows:

1- Simplify for Accuracy

2- Parameter would be generated i.e. boost pipes diameter", Give an initial guess to the this parameter in the case setup.

3- Caliberate for on case. 

4- Extend this parameter values for all cases by running the all cases after caliberation process.

5- Check to see if any reults are outside the accuracy tolerances
  
As above listed strategy for FRM Converter , Step3 is about intake pipes, so the procedure would be  as follows:

1- Simplify for Accuracy

2- Parameter would be generated i.e. diameter of intake pipes", Give an initial guess to the this parameter in the case setup.

3- Caliberate for on case. 

4- Extend this parameter values for all cases by running the all cases after caliberation process.

5- Check to see if any reults are outside the accuracy tolerances
  
As above listed strategy for FRM Converter , Step6 is about additional changes , so the procedure would be  as follows:

1- Simplify for Speed now instead of Simplify for Accuracy

2- Parameter would be generated i.e. Heat Transfer Model", Give an initial guess to the this parameter in the case setup.

3- Caliberate for on case. 

4- Extend this parameter values for all cases by running the all cases after caliberation process.

5- Check to see if any reults are outside the accuracy tolerances
  
If additional decrease in time is required then further cylinder slaving could be done
THE FINAL FRM MODEL
Notice the snapshot below. It was the original SI_4cyl_GDI_Turbo_WOT . the image below this image is the FRM converted model's simulation time snapshot.

Notice how much time its consuming with respect to different cases (RPM) compared to the image beneath this image which is the FRM model.
Build FRM Model for following configuration using FRM builder approach
Bore 102 mm stroke 115 mm CR 17
No of cylinder 6
CI engine
Twin Scroll Turbine
GT Controller
Run all cases

Step 1:

Go to Tools and then on FRM Builder.

A window whose snapshot is provided below would prompt.
Step 2:

Choose between SI or CI Engine
Step 3:

Enter the Geometry of the Engine.
Step 4:

Choose between Twin Scroll Turbo and Variable Geometry Turbine Turbo.
Step 5:

Select either 6 cylinder or 4 cylinder Engine
Step 6:

Choose a control logic type
FINAL FRM MODEL
Model Simulation Time per Case 
The reults are obtained quite fast in this approach compred to the one with high fidelity
Your portfolio, next

Ship work like this
in your own course.

Talk to a career counsellor

Free 20-minute call. We'll map the right course to where you want to work.

By submitting you agree to our Terms and Privacy Policy.

Talk to a career counsellor

Free 20-minute call. We'll map the right course to where you want to work.

By submitting you agree to our Terms and Privacy Policy.