Geothermal model breaking at certain treshholds

[Mai MRST]

Dear MRST Team,

I have been running a series of geothermal simulations and have observed that the models fail under certain conditions. Specifically, increasing the injection temperature appears to lower the threshold injection rate at which the simulations become unstable and break.

Could you provide insight into what might be causing this behaviour?

Attached is a plot showing the pressure and temperature evolution at the well for the timesteps before model failure.

Best wishes,

Mai

[Olav Møyner]

Hi!

What do you mean when you say that the models are breaking? The model stops converging at some point during the timesteps? Or do you get another error message?

The geothermal module uses an equation of state internally that has a range of validity. If you go far beyond the limits of the model, it is likely that you get unphysical values which will lead to issues. I do not remember off-hand what the ranges are, but it could be that you reach the boundaries of the model at 550 K. Do you get any warnings or error messages in the simulation output?

Best regards,

Olav

--
You received this message because you are subscribed to the Google Groups "MRST-users: The Matlab Reservoir Simulation Toolbox User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sintef-mrst...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sintef-mrst/cc2a95b7-a9bf-4392-8ba3-603e0505d211n%40googlegroups.com.

[Mai MRST]

Hi Olav, thank you for your reply. 

Indeed, I receive a warning that is similar to following: Solver did not converge in 25 iterations for timestep of length 20 Days. Cutting timestep.

Considering that the project I am working on injects high-temperature fluid, but this approach causes the model to fail, are there alternative methods to achieve high-temperature injection, perhaps by employing a different model?

And out of curiosity, is there currently or are you planning on creating an EOS/model that can handle supercritical fluid?

Best wishes,

Mai