Non-convergence when considering permeability heterogeneity
Hello again, the non-convergence problem is still unresolved when considering permeability heterogeneity in my TOUGH2/ECO2N CO2 storage model. In this model I used the default/common parameter values given in the TOUGH2 related articles.
The model was working correctly when considering homogeneous permeability. But once the permeability heterogeneity (by adding a domain named 'SEED' in block ROCKS and assigning each cell a PM value) was added, the model failed to converge.
According to the output file, I found the pressure in all cells (except for the boundary cells that were fixed) increased by serveral times even if most of them are far from the CO2 plume. The time step reduced from days to seconds before the simulation halted.
Any help would be appreciated.
The input and output files and a figure illustrating the conceputal model are attached.
Regards,
Jim
2 replies
-
Jim,
Two things:
(1) It seems you tried to run this file with ECO2M instead of ECO2N (but that's not the issue).
(2) If you have a dummy domain SEED in your input file with DROK=POR=0.0, permeability modifiers are read from block ELEME, Columns 41-50. All your boundary elements (grouped at the end of the ELEME block) are blank in the slot reserved for PMX, i.e., the PMX is zero, which sets the permeabilities to zero and thus closes the system (which explains the pressure increase).
Solution: Set PMX=1.0 for the last few elements with material BOUND and it will run.
Best,
Stefan