Warning about differentiation in dymola 2020

[Mathadon]

In dymola 2020 I sometimes get:

Differentiating (if noEvent(abs(dp) > dp_turbulent) then noEvent((if dp > 0 then 1 else (if dp < 0 then -1 else 0)))*k*sqrt(abs(dp)) else (1.40625+(0.15625*dpNormSq-0.5625) *dpNormSq)*m_flow_turbulent*dpNorm)
under the assumption that it is continuous at switching.

which resides in IBPSA.Fluid.BaseClasses.FlowModels.basicFlowFunction_dp. Adding smooth(2, in the function body solves the problem, however we already have smooth=2 in the annotation so we should not get this error. Something to report to the developers?

[mwetter]

Yes, this sounds like a bug (or missing feature) of the Modelica tool.

[Mathadon]

reported

[Mathadon]

Feedback from Claytex: they consider the behavior to be ok and consider it a safety feature to avoid people not getting a warning. This does not seem to be in line with the Modelica specification:

smoothOrder defines the number of differentiations of the function, in order that all of the differentiated outputs are continuous provided all input arguments and their derivatives up to order smoothOrder are continuous

which implies that smoothOrder=2 results in continuous derivatives so I don't think there should be a need for this double-check.

Do we add the smooth(2,.) to our functions or do we want to argue further about this? Perhaps directly with Dassault.

[mwetter]

@Mathadon : It would be best if you can directly contact Dassault. I see no need to double-declare the property with smoothOrder=2 and smooth(2, ...). This appears to be an improvement that should be added to Dymola.

[Mathadon]

Forwarded to Dassault by Claytex. I propose to continue with the hotfix (smooth(2,) in the CheckValve pull request for now.

[mwetter]

OK, I don't see any harm in using smooth(2, ...) until it is fixed.

[Mathadon]

then this is closed by #1224

[mwetter]

@Mathadon : The same smooth(2, ...) annotation should be added to IBPSA.Fluid.BaseClasses.FlowModels.basicFlowFunction_m_flow which we require in some of our new models that we work on. I will make the change.