Separator level and volume calculations

[Alyasiri37]

Hello,

I have made the following modifications to the Separator class:

1- Included calculation for separator level from liquid volume(based on separator type) instead of the old percentage
2- Cross-section calculation for separator phases
3- Liquid volume calculation from actual liquid level
4- Getter and setter for liquid level updated to convert from percentage to original units.(both accept percentage)
5- Added new variables for radius, separator volume, and cross section
6- Updated class functions to implement those changes

In short, the separator class should now handle actual level instead of volumetric ratios, based on separator orientation. However, one of the tests have failed (ProcessSystemRunTransientTest.java) due to the assert statement on line 230. I have not modified the test and would like to know more about this value reference. Is it based on simulation from another tool? how to check calculation accuracy?

Fork branch is on
https://github.com/Alyasiri37/neqsim/tree/separator-level

And it is already rebased with the main project.

[EvenSol]

Hi.
This seems like great modifications. I will review and give my feedback.
Even

[EvenSol]

Good update. I made a pull request here:
#1561

[Alyasiri37]

Glad to hear that professor!
Thanks.

[EvenSol]

Had to move it here:
#1562

[EvenSol]

@Alyasiri37 I was trying to update the Colab example according to your updates. It would be nice if you can have a look at the example and suggest updates:
https://colab.research.google.com/drive/1WkRtkrZskI0lNaVBpRZYuie8MB_jsx7b?usp=sharing

[Alyasiri37]

@EvenSol Sure. Will have a look.

[Alyasiri37]

@EvenSol
Overall, calculations seem to be fine. To keep dynamics realistic, I suggest keeping the flow rate around 150 kg/hr bearing in mind the current density, separator volume and simulation duration. This makes level fluctuations easier to observe. For level controllers, this gain set worked for me (26, 400, 0). Also PR EOS would be more suitable for the given example.

I will compare dynamic separator control against DWSIM once I have time.