Closed ghanique-altair closed 1 month ago
Thanks for the contribution Please give me a few day to think about this, maybe there is a better way of doing this I coupled with steady state solvers before, then it was working
Any news ?
AdvanceInTime
already get the current global time, did you try using this?
For sure I tried, but in the case, the time sent to the steady solver is one step late.
But AdvanceInTime is called before your proposed SetTime method
AdvanceInTime get the currentTime from the transient solver, so you have to wait for the next call of AdvanceInTime to be able to sent that time value to a steady solver, so it is one step late.
Ok now I understand
Given that the global time is already in the processInfo of the coupled solver, I am leaning towards passing this one to the solver-wrappers. Checking it now ...
can you check if it works with my commit?
Now you can access the global time from the solver wrapper
Ok, thanks to your modification I can access to the global time of the cosimulation in my Flux solver_wrapper by using 'solver'._parent_process_info[KM.TIME]. This way will be deliver in the Flux solver_wrapper (delivered in Flux setup). Can you commit your modification in the main branch of Kratos ?
Can you commit your modification in the main branch of Kratos ?
yes, will happen automatically as soon as the pipelines passed
📝 Description This development enable to send the global time of a cosimulation to a steady solver through the solver_wrapper. This is only usefull for steady solvers which need to update parameters depending on the global time and then reproduce a so-called transient behavior.
🆕 Changelog