Hi,
Comments inline ...
On 2015-04-13 11:06, Per Oberg wrote:
I have now put the library in it's current version on Github, se [1]
below. The latest version includes the fixes i mentioned in my last post.
Very good! I wanted to ask if that's possible (to put it in github)
in my last email but forgot to do so.
The examples that are working in openModelica 1.9.2 Beta 1 are:
* energyPreservationTest.mo
* throttleSanityCheck.mo
The example that is not working is
* SIEngineTest
*Code to reproduce:*
cd("C:/Temp/Modelica-MVEM/MVEMLib")
clear()
loadModel(Modelica);
loadFile("package.mo")
instantiateModel(MVEMLib.Examples.SIEngineTest)
simulate(MVEMLib.Examples.SIEngineTest, stopTime=15, method="dassl",
tolerance=1e-005)
Unfortunately I don't have a track-account yet for my new address. Maybe
I should get one.
You have an old one. I'll change the email address
in it and let you know the details via email.
Cheers,
Adrian Pop/
[1] https://github.com/droberg/Modelica-MVEM.git
Best regards
Per Öberg
------------------------------------------------------------------------
Wolfram MathCore AB
Teknikringen 1F
SE-583 30 Linköping
Sweden
Phone: +46 13 32 85 00 <tel:%2B46%2013%2032%2085%2000>
Fax: +46 13 21 27 01 <tel:%2B46%2013%2021%2027%2001>
Please visit us at: www.wolframmathcore.com
<http://www.wolframmathcore.com/> or www.wolfram.com
<http://www.wolfram.com/>
------------------------------------------------------------------------
*Från: *"Adrian Pop" <adrian....@liu.se>
*Till: *openmodelicainterest@ida.liu.se
*Skickat: *fredag, 10 apr 2015 19:54:23
*Ämne: *Re: Comments and questions about Media support
Hi,
Comments inline ...
On 2015-04-10 15:45, Per Oberg wrote:
> Hello
>
> Some years ago I developed a library for mean value engine modeling in
> Modelica (as part of the openProd project), see [1] below. I recently
> went back and checked if there had been any development on the parts
> that failed in openModelica last time, and to my surprise it now looks
> very good and some of the models works quite well.
Good to know.
>
> So, first of all: Good job to you all!
> (When I tried it last time it didn't work at all due to all kinds of
> reasons)
>
> To get it to work reasonable well I had to change some questionable code
> that probably wasn't right in the first place. This made the smaller
> examples work and their output is the same as when running with dymola.
> For the larger example it still fails however. It flattens but I cant
> get it to simulate.
>
> The reported error is as follows:
>
> record SimulationResult
>
> resultFile = "",
>
> messages = "Failed to build model: MVEMLib.Examples.SIEngineTest"
>
> end SimulationResult;
>
> OMC-ERROR:
>
> "Warning: The initial conditions are not fully specified. Use
> +d=initialization for more information.
>
> Warning: There are iteration variables with default zero start
> attribute. Use +d=initialization for more information.
>
>
[c:/dev/OpenModelica/Compiler/SimCode/SimCodeUtil.mo:3474:7-3474:46:writable]*Error:
> Internt fel function createNonlinearResidualEquationsComplex* failed
> for: ({-sIEngine.exhaustTemperatureDrop.InPut.dmXi[1],
> -sIEngine.exhaustTemperatureDrop.InPut.dmXi[2],
> -sIEngine.exhaustTemperatureDrop.InPut.dmXi[3],
> -sIEngine.exhaustTemperatureDrop.InPut.dmXi[4],
> -sIEngine.exhaustTemperatureDrop.InPut.dmXi[5],
> -sIEngine.exhaustTemperatureDrop.InPut.dmXi[6]},
> sIEngine.adiabaticBurner.dummyOutput) =
>
MVEMLib.Basic.Media.AdiabaticBurner.fuelAirMedium.calcBurnedFractions({sIEngine.adiabaticBurner.UnburnedInPut.dmXi[1],
> sIEngine.adiabaticBurner.UnburnedInPut.dmXi[2],
> sIEngine.forcedMassFlow.InPut.dmXi[1]}, sIEngine.exhaustManifold.gas.T)
>
>
>
> Is there anything I can do about this, like specifying initial values
> better or is it a bug?
As far as I can see right now it does look like a bug.
>
>
> [1] http://www.fs.isy.liu.se/Software/Modelica-MVEMLib/
Per, is this the latest version or you have some newer code?
You mentioned above some changes to the code, are they included?
I ask because we need to debug this with the code your used.
By the way what model did you ran from the library to get
to this errors?
I suggest you add a Trac ticket with more details and we'll
look into it.
Cheers,
Adrian Pop/
>
>
> Per Öberg
>
> ------------------------------------------------------------------------
> Wolfram MathCore AB
> Teknikringen 1F
> SE-583 30 Linköping
> Sweden
> Phone: +46 13 32 85 00 <tel:%2B46%2013%2032%2085%2000>
> Fax: +46 13 21 27 01 <tel:%2B46%2013%2021%2027%2001>
>
> Please visit us at: www.wolframmathcore.com
> <http://www.wolframmathcore.com/> or www.wolfram.com
> <http://www.wolfram.com/>
>
