You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Users have reported that reading data from runs with lgr seems unintuitive. They were interested in finding the values from the main grid in the case with lgr included. Although the main grid is not used for simulation, averaged values are still stored for book keeping, and it is therefore possible to extract comparable values in both scenarios.
Approach:
# Simulation case without lgr
original_grid = ecl.EclGrid('ORIGINAL.DATA')
original_rest = ecl.EclFile('ORIGINAL.UNRST')
original_init = ecl.EclFile('ORIGINAL.INIT')
# Simulation case with lgr
lgr_grid = ecl.EclGrid('LGR.DATA')
lgr_rest = ecl.EclFile('LGR.UNRST')
lgr_init = ecl.EclFile('LGR.INIT')
tstep = 10
x = 1
y = 1
# Read saturation for both runs
original_soil = original_rest.iget_named_kw("SOIL", tstep)
lgr_soil = lgr_rest.iget_named_kw("SOIL", tstep)
assert (original_soil != lgr_soil)
lgr_soil = lgr_rest.iget_named_kw("SOIL", tstep * 2)
assert (original_soil == lgr_soil)
They found that length of lgr_soil is twice the length of original_soil (likely n-times the length, where n is number of lgr's + 1). What is a bit curious though is that the values where tstep is the same, didn't make much sense. They did not compare correctly with the values in the lgr grid. Still, reading from cells in the lgr grid, from what I can recollect, should be performed by fetching the lgr and then retrieving values from that. I have not looked any more into this.
Hopefully easy to reproduce.
The text was updated successfully, but these errors were encountered:
It's been a while since this issue was created, I must admit I don't recall everything here.
I thought this issue was reporting problems finding the values from the lgr grid itself. If one would want to find the values from the main grid one would call grid.iget_named_kw("KEYWORD", tstep * n+1), where n = number of lgr. Given the description in the issue it looks like the assertion on comparing main grid values are consistent. Does this provide you with expected output?
As long as the values are stored in the UNRST files from the simulator it is likely possible to fetch. Haven't looked into this though..
Users have reported that reading data from runs with lgr seems unintuitive. They were interested in finding the values from the main grid in the case with lgr included. Although the main grid is not used for simulation, averaged values are still stored for book keeping, and it is therefore possible to extract comparable values in both scenarios.
Approach:
They found that length of
lgr_soil
is twice the length oforiginal_soil
(likely n-times the length, where n is number of lgr's + 1). What is a bit curious though is that the values where tstep is the same, didn't make much sense. They did not compare correctly with the values in thelgr
grid. Still, reading from cells in thelgr
grid, from what I can recollect, should be performed by fetching thelgr
and then retrieving values from that. I have not looked any more into this.Hopefully easy to reproduce.
The text was updated successfully, but these errors were encountered: