Event handling for DAEs

[AnHeuermann]

How to handle events inside or between DAEs?

Possible discussion topics:

• What information are needed for the event iteration?
• Do different FMUs need different handling? Based on some capabilities of the DAE-FMU?
• How does the event search and iteration (in different versions) fit into the ME state machine?

[AnHeuermann]

An OpenModelica FMI-LS-DAE FMU want's to be handled by the importing tool something like the following diagram.
The important part for us, is that we (sometimes / usually) need the causalized ODE systems to do the correct thing for the event iteration. I think this was for synchronous events

Note, that in the current status of the DAE mode im-
plementation it is still mandatory to generate the causal-
ized code for proper handling of synchronous events and
discrete variables.

https://re.public.polimi.it/bitstream/11311/1065357/1/2017-BraunCasellaBachmann.pdf

The Excalidraw file:
FMI-LS-DAE event handling.zip

[elmir-nahodovic]

It is similar to how Dymola does it, except that also the state event localization is done in "ODE Mode" in our case. Of course, the ls-dae shouldn't mandate any specific way, you could even stay in "DAE mode" throughout the simulation.

I think the diagram is very nice, and it's important to have throughout these discussions.

One aspect I was thinking of is that the actual update of discrete states should be entirely left to the FMU with the call/iteration to updateDiscreteStates. Therefore the part "do event iteration in ODE" should just be the updateDiscreteStates loop.

I even think ODE vs DAE inside the updateDiscretStates might be left to the FMU, this is more about its internal implementation that perhaps shouldn't be handled by the importer, unless there is some good use-case for allowing the importer to control it