nieweber
commented
1 year ago Thank you for your contribution to the MSL-TFS interface. I had a first look on your implementation today. In your TFS->MSL interface you are directly coupling the pressure and mass-flow of both approaches. This is something that should be treated with care to avoid the spread of large non-linear equation systems across components and hence are taking the advantages of our modeling approach. It would probably make sense to have a quick call for better understanding of your general idea and your test cases and how to proceed. Internally we are working on an alternative solution and conducted similar tests.
Also I saw that in your "target test case" you are plugging in a component from the MSL into a closed cycle. It would be preferable to model the whole subsystem in the MSL and plug it to the TFS to keep the benefits of the TFS modeling approach.
hubertblervaque
You will find in Interfaces (and also in Undirected.Interfaces) a new package with my current works about the development of plugs between MSL and TFS fluid connectors as mentioned in Issue #125. For directed flow, I initially try to model these plugs with existing components (subpackage "ByAssembly") then only by a set of equations (subpackage "ByCode"). Basic unit tests are running for both. It must be working for all kind of medium for "ByAssembly" plugs but I have not yet code the case of mixture medium for "ByCode" plugs. I structure the interfaces of Undirected plug and the unit test I would like to get but no model are still tested. I guess this Undirected flow could be contructed from the two "Directed" plugs. Feel free to contact me if needed. Could you please inform when you have a vision of the integration of this task in your planning.