Distance-based localiztion test script for testing full update workflow using ERT storage with ensemble experiments - #230
Closed
oddvarlia wants to merge 15 commits into
Closed
Distance-based localiztion test script for testing full update workflow using ERT storage with ensemble experiments#230oddvarlia wants to merge 15 commits into
oddvarlia wants to merge 15 commits into
Conversation
oddvarlia
marked this pull request as draft
August 15, 2025 12:22
dafeda
reviewed
Aug 19, 2025
| groups = list(param_config_all.keys()) | ||
| realizations = ensemble.get_realization_list_with_responses() | ||
| group_number = 0 | ||
| for group_name in groups: |
Collaborator
There was a problem hiding this comment.
You can use enumerate to avoid having to increment group_number manually.
for group_number, group_name in enumerate(groups)
dafeda
reviewed
Aug 19, 2025
| smoother = DistanceESMDA( | ||
| covariance=C_D, observations=observations, alpha=alpha, seed=seed | ||
| ) | ||
| smoother.prepare_assimilation(Y, truncation=0.99) |
Collaborator
There was a problem hiding this comment.
I think we can as part of the assimilate_batch function and only do the actual calculations if self.X3 and self.D` are not already set.
dafeda
reviewed
Aug 21, 2025
| # Turn on/off export of field parameters into 3D parameter files in ROFF format to be read by RMS. | ||
| WRITE_STATS_PARAMS = True | ||
|
|
||
| # Common definition of ERTBOX for all zones in Drogon |
Collaborator
There was a problem hiding this comment.
Does the grid file specified using the GRID keyword in the Drogon project contain all this information?
dafeda
reviewed
Aug 21, 2025
| OBS_FILE_PATH = ( | ||
| "/project/fmu/users/olia/drogon_20250623_11-50/resmod/ff/25.0.0/rms/model/" | ||
| ) | ||
| OBS_FILE = "summary_obs_with_localization_attributes.csv" |
Collaborator
There was a problem hiding this comment.
Can you push this file to the repo as well or is it perhaps too big?
dafeda
reviewed
Aug 21, 2025
| OBS_SUMMARY_WITH_LOCAL_ATTRIBUTES_FILE = OBS_FILE_PATH + OBS_FILE | ||
|
|
||
| # Hard-coded a classification of parameter groups into scalars, 2D or 3D field parameters | ||
| # NOTE: This should be available from ERT storage (ERT knows about the source the keyword FIELD, SURFACE GEN_KW) |
Collaborator
There was a problem hiding this comment.
Here's one way to get the parameter type:
with open_storage(storage_path, "r") as storage:
experiment = storage.get_experiment_by_name(experiment_name)
print(experiment.parameter_configuration["COND"].type)
# >> field
print(experiment.parameter_configuration["INIT_TEMP_SCALE"].type)
# >> gen_kw
dafeda
reviewed
Aug 21, 2025
| # (e.g for permeabilit or porosity) could be <zone_name>_<facies>_<petro_name> | ||
| # for petrophysical fields to be updated simultaneously with facies update from APS. | ||
| # For reservoir zones without facies modelling, the convention could be <zone_name>_<petro_name> | ||
| # NOTE: An additional parameter in FIELD keyword with zone name can solve the problem if naming convention |
Collaborator
There was a problem hiding this comment.
An issue has been created on this that needs more information:
oddvarlia
force-pushed
the
real_case_test
branch
from
September 4, 2025 08:56
00e0f0e to
974fcfb
Compare
Collaborator
|
I suggest we generate .csv files that are comma delimited and not delimited by space. This can avoid potential bugs if for example a zone name has a space in its name. |
oddvarlia
force-pushed
the
real_case_test
branch
3 times, most recently
from
September 7, 2025 21:05
a5846f6 to
0b1f9a9
Compare
oddvarlia
force-pushed
the
real_case_test
branch
3 times, most recently
from
September 23, 2025 10:59
7505b9c to
14a73ba
Compare
oddvarlia
force-pushed
the
real_case_test
branch
2 times, most recently
from
October 4, 2025 08:57
6992e30 to
c7e0d02
Compare
oddvarlia
force-pushed
the
real_case_test
branch
from
October 10, 2025 12:41
41ecce3 to
3dc8862
Compare
oddvarlia
force-pushed
the
real_case_test
branch
from
December 13, 2025 15:09
379e45b to
49af424
Compare
Collaborator
Author
|
The class DistanceESMDA is now merged into main. See merged PR 240. The test script with additional help scripts for RMS and yml files for drogon case is not merged. |
oddvarlia
force-pushed
the
real_case_test
branch
from
December 19, 2025 17:29
49af424 to
82a16a6
Compare
oddvarlia
force-pushed
the
real_case_test
branch
2 times, most recently
from
January 20, 2026 10:31
8baf006 to
780f49a
Compare
oddvarlia
force-pushed
the
real_case_test
branch
3 times, most recently
from
February 15, 2026 09:10
577d167 to
b5f3fae
Compare
oddvarlia
force-pushed
the
real_case_test
branch
from
February 27, 2026 09:49
b5f3fae to
ba89bfd
Compare
…ning on a case from ERT storage Added description of implementation steps Update create_obs_with_localisation_attributes update distance_based_on_real_case.py Working version with summary data Added seismic obs and RFT obs Attempt to save update to ERT storage Added tracer obs Added tracer obs file Added ESMDA update to scalars Modified the path to where on scratch to find the ensemble Ensure that updated ensemble in storage has iteration greater than 0 Added new scaling function Changed seismic_range Update scalar parameters Updated document lines Updated test_distance_based.ipynb to initialize ESMDA before each update Update code and use ertbox parameters from storage Update input to handle seismic and rft obs per zone Added adaptive localisation for scalara parameters Added functionality to simulate spatially correlated seismic observation errors Added option to specify D matrix as input to ESMDA and DistanceESMDA update distance_based_on_real_case.py to work with new version of storage Update experimental.py and test_experimental.py to be equal to main
oddvarlia
force-pushed
the
real_case_test
branch
from
March 6, 2026 17:52
ba89bfd to
ee10d05
Compare
Collaborator
|
Closing as this was not meant to be merged but was pushed for discussions. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The script is initially using Drogon case from this summer.
The distance-based localization depends on having position for observations (well prod observations and seismic points for 4D)
This is currently not available in ERT and storage directories created by ERT so the information must be taken from the FMU setup for seismic observations and the RMS forward model.
A script to extract position data from wells using RMS project is included
The plan is to get position data for the seismic observations as well by loading files with seismic data points including position from ERT project.
Main goal of this branch is to test and evaluate performance of distance-based localization before the method is implemented in ERT code.
This branch is built on the branch Distance-based localization per 15.08.2025