Updating MICP
Depends on https://github.com/OPM/opm-common/pull/4538
Please provide a more speaking title and a description if necessary. This will help with the manual.
Since this is quite technical and involves a lot of things that need to be documented, I am planning on doing myself a PR to the manual with all details after it is merged (as mentioned here).
Many thanks @totto82 for your comments, I think I have addressed them. Please let me know if these PR's needs more modifications.
Thanks for the update. I think what you did with the scaling factor is correct. Now, you only scale the equation for urea with a factor of 10. And not the variable so you don't need to scale back? Did you see any difference in non-linear behaviors (and results)? The only comment not addressed, I believe, is the first one, i.e., the alphabetic ordering of included files. After that, this can be merged IMO.
Thanks for the update. I think what you did with the scaling factor is correct. Now, you only scale the equation for urea with a factor of 10. And not the variable so you don't need to scale back? Did you see any difference in non-linear behaviors (and results)? The only comment not addressed, I believe, is the first one, i.e., the alphabetic ordering of included files. After that, this can be merged IMO.
I think the alphabetic order mentioned by @akva2 has been addressed (somehow it does not showed as Outdated in those comments). The variable does not need to be scaled back, since the scaling factor is applied to the flux, source, and storage terms. For the added test, without this weighting the Overall Linear Iterations is 128, while by weighting this it results in 125 (both of them 82 Newton Iterations). For this case the cell results look basically the same, but it is expected this differences are larger for tighter weights, which could be analogue to tightening the simulator cnv tolerances.
Thanks for the testing. From my point of view this is good to go. I will just rerun jenkins to make sure everything is fine.