Incorporate ideas and requirements of 2018 proposal for DAE support in FMI

[chrbertsch]

see https://github.com/modelica/fmi-design/blob/master/WorkingGroups/02_DAE_PartialDerivatives/Meetings/2018-04-04_05-FMI-Design-Meeting-Dresden/DAE.pdf

[elmir-nahodovic]

One particular requirement is how to extend the ModelStructure for the residual equations and dependencies on the states and algebraic variables. Is a layered standard allowed to e.g. add another element to the ModelStructure - similar to ContinuousStateDerivative but for the residual equations?

[chrbertsch]

One particular requirement is how to extend the ModelStructure for the residual equations and dependencies on the states and algebraic variables. Is a layered standard allowed to e.g. add another element to the ModelStructure - similar to ContinuousStateDerivative but for the residual equations?

In a layered standard on top of FMI 3.0 this is not possible, as the FMUs has to be a valid FMI 3.0 FMU.
So one would either define new ignorable XML elements in a new minor 3.x version.
Or use one of the possibilities that the Layered Standard concept allows:

• define the new elements in standardized annotations
• define the new elements in the manifest XML file of the Layered Standard (see e.g. https://github.com/modelica/fmi-ls-xcp/blob/main/schema/fmi3LayeredStandardXcpManifest.xsd for an example from the FMI-LS-XCP.
• define an additional XML file that can be stored in the subfolder of the Layered Standard in the /extra directory.

For more information on the possibilities of extenson mechanisms of Layered standards see https://fmi-standard.org/docs/3.0.2/#VersioningLayered