#4257 closed defect (fixed)
Simulation with ida solver doesn't start in OMEdit
Reported by: | anonymous | Owned by: | Adeel Asghar |
---|---|---|---|
Priority: | high | Milestone: | 1.12.0 |
Component: | OMEdit | Version: | v1.11.0 |
Keywords: | Cc: |
Description
The simulation output is:
stdout | OMEditInfo | <p>~/OMEdit/Modelica.Electrical.Analog.Examples.CauerLowPassSC -port=34177 -logFormat=xmltcp -override=startTime=0,stopTime=60,stepSize=0.04,tolerance=1e-06,solver=ida,outputFormat=mat,variableFilter=.* -r=Modelica.Electrical.Analog.Examples.CauerLowPassSC_res.mat -jacobian=coloredNumerical -w -lv=LOG_STATS</p> assert | debug | <p>For the klu solver jacobian calculation method has to be kluSparse</p> stdout | OMEditInfo | <p>Simulation process finished successfully.</p>
but no results are appears.
Change History (5)
follow-up: 2 comment:1 by , 8 years ago
follow-up: 3 comment:2 by , 8 years ago
Replying to wbraun:
One problem is a wrong return value in solver step and is addressed in RP1459 for the runtime. The other issue is that for the ida solver we use a different default jacobian method than for dassl, and this should changed automatically.
@adeas31 Can we adjust it, so in case ida is selected the jacobian changes to kluSparse.
Instead of doing that I guess I should remove -jacobian
. Only use -jacobian
if its value is not coloredNumerical
(i.e., if user explicitly want to use other than the default).
comment:3 by , 8 years ago
Replying to adeas31:
Replying to wbraun:
One problem is a wrong return value in solver step and is addressed in RP1459 for the runtime. The other issue is that for the ida solver we use a different default jacobian method than for dassl, and this should changed automatically.
@adeas31 Can we adjust it, so in case ida is selected the jacobian changes to kluSparse.
Instead of doing that I guess I should remove
-jacobian
. Only use-jacobian
if its value is notcoloredNumerical
(i.e., if user explicitly want to use other than the default).
Yes, I agree we could just remove the jacobian option, and only use via simflags. Also the other DASSL related options are actually also available for IDA, so we should probably at least rename the options.
comment:5 by , 8 years ago
Milestone: | Future → 1.12.0 |
---|
One problem is a wrong return value in solver step and is addressed in RP1459 for the runtime. The other issue is that for the ida solver we use a different default jacobian method than for dassl, and this should changed automatically.
@adeas31 Can we adjust it, so in case ida is selected the jacobian changes to kluSparse.