initialization in Dymola influenced by annotation flag evaluate

Question

As already mentioned in the title, initialization of my (multibody) model is influenced by annotation(evaluate).

With new parameter par implemented in the model, initialization always fails if I use :

parameter par annotation(evaluate=false)

but if I use

evaluate=true,

it works.

The first option is unfortunately not a good solution for me, as I want to be able to set the parameter value after compilation time.

What exactly happens to the model structure when one of the parameters is set to evaluate=false? What should I check besides initial conditions?

Maybe part of your question is answered here: http://book.xogeny.com/behavior/equations/annotations/#evaluate — matth, Oct 28 '15 at 09:57

score 1 · Accepted Answer · answered Nov 27 '15 at 12:56

What is value of the parameter?

There are a number of different scenarios where this can occur:

Boolean/Integer values that determine the size of arrays etc. If they have Evaluate=false you will have problem even translating your model.
Zero values for e.g. mass that influence the index-reduction; in that case you should log the index-reduction in addition to initial conditions.
Anything else that influence the structure of the systems of equations for the initialization; you can look at that log - and also add debug-messages for non-linear system of equations (e.g. to detect if they are singular). The axis of revolute joints are often in the last category.

1 Answers1