srebughini / asali Goto Github PK

This issue is related to the elapsed time comparison of the API version of ASALI.
The comparison shows that the C version is too slow compared to the Fortran and C++ version, therefore we need some help to improve it.

We are looking for contributors that can help us writing a more efficient C code for ASALI.

Feature request

ASALI should be improved with modern programming languages. I suggest to start working with:

• Carbon
• Rust
• Anything in your minds...

Solution that we would like to have

This improvement should be added to the api branch.
Since new contributors are always welcome, here a short TODO list to make the implementation of new programming languages easier:

• Develop the files parser (usually called database-generator): ASALI properties estimation method is based on 4 different files, containing fitting parameters for the formula that should be implemented. NASA standardized the file structure long time ago, so please dont change it!!! To make the use of ASALI-API more user-friendly, we decided to hard-code this parameters. For this reason, the file parser should read those files and convert them into hard-coded vector/matrix. An example can be found here.
• Develop the set/get function: ASALI requires to set some properties (e.g. Temperature, Pressure, Composition) and get other properties (e.g Viscosity, Density). Here, you can find a list of the get and set function required, with also the type of variable required.
• Develop the core formulas: ASALI is based on mathematical formulas to estimate gas mixture properties. Once you need them, feel free to comment this issue.
• Develop the example: To improve the user experience of ASALI-API, we add examples to show how to use it.

Comments

@andreee94: you might be interested in this

Dear Stefano,

I have installed Cantera via Conda following the instructions on https://cantera.org/install/conda-install.html

Now, how do I install ASALI? I can't find the program folder for Cantera to replace in the .compile.sh file.

Thanks for the help.

Describe the bug
I ran a python script with Cantera to calculate mixture averaged diffusion coefficients with a certain T, P and composition (mass fraction). I tried to calculate the same coefficients with ASALI. If I choose mass fractions for the composition in the python script, I only get the same answer if I select mole fraction in the ASALI GUI. vice versa.

To Reproduce
Steps to reproduce the behavior:

1. run the attached script, which is setup for mass fraction
2. run ASALI, use the default parameters and got to transport properties. Enter the same mass fraction composition, T and P as the Py Script
3. The diffusion coefficients in ASALI will be different
4. Hit back on the GUI
5. Change mass to mole fraction in the ASALI GUI. Now the mixture diffusion coefficients will match between ASALI and the Python script

Expected behavior
I expected that the mixture averaged diffusion coefficients would be the same in both the Python script and in ASALI when both the Python script and ASALI accept the composition in mass fraction

Desktop (please complete the following information):

• OS: Ubuntu
• Version 22.04

Additional context
diffusionCoeff.py.txt