create_rft_ertobs

This scripts creates .txt, .obs and a “welldatefile” to be used for GEN_DATA usage in assisted history match for matching of RFT pressure values in wells.

The script must be run from within RMS to be able to interpolate from MD to (x,y,z) (or the other way) along wellpaths.

Usage

Typical input is a CSV file:

Example CSV input
DATE	ERROR	MD	PRESSURE	WELL_NAME	ZONE
2018-03-01	3	1697.9	299.23	RFT_55_33-A-2	Valysar
2018-07-03	3	1673.4	298.52	RFT_55_33-A-3	Therys
2018-07-03	3	1690.7	280.43	RFT_55_33-A-3	Volon
2018-09-12	3	2578.7	288.60	RFT_55_33-A-4	Valysar
2018-09-12	3	3197.5	282.13	RFT_55_33-A-4	Volon
2018-04-28	3	1737.8	278.70	RFT_55_33-A-5	Valysar
2018-04-28	3	1769.1	286.55	RFT_55_33-A-5	Volon
2018-11-07	3	1759.9	280.85	RFT_55_33-A-6	Therys
2018-11-07	3	1777.4	286.41	RFT_55_33-A-6	Volon

For each row, corresponding to a specific RFT observation value, either (x,y,z) (in the columns EAST, NORTH and TVD), or the measured depth MD must be specified, (and the missing value(s) will be interpolated from the well trajectory).

The RKB of measured depth (MD) must be the same as for the well in the RMS project.

If ZONE is specified, the provided zone value is compared with the zone name valid for the relevant cell in the RMS model. If they do not match, a warning is emitted.

If a well has observations as multiple dates, the REPORT_STEP column in the welldatefile will be enumerated from 1 and upwards. The obs-files produced will always include the REPORT_STEP in the filename.

Observation files will always cover the measurement points for all dates for a well, but padded with -1 at dates (report steps) where there is no data.

Additional options are specified through a RMS dictionary, and create_rft_ertobs can be called from a RMS Python job with a script like

from fmu.tools.rms import create_rft_ertobs

# Paths are relative to rms/model
CONFIG = {
    "input_file": "../input/well_modelling/rft_observations.csv",
    "alias_file": "../../config/rms_eclipse_alias.csv",
    "exportdir": "../../ert/input/observations/rft",  # Must exist
    "project": project,  # The Python object representing your RMS project
    "gridname": "Simgrid",
    "zonename": "Zone",
    "verbose": True,
    "clipboard_folder": "RFT_ERT_observations", # Optional folder for storage in RMS
}

create_rft_ertobs.main(CONFIG)

The example provides a typical minimum. Available options to set are

input_file: CSV file with input data, one row for each RFT observations. The DATE column must be in ISO-8601 format (YYYY-MM-DD). Required columns are “DATE”, “MD”, “WELL_NAME” and “PRESSURE”.
alias_file: A CSV file with RMS well names in one column and Eclipse names in a different column. The column names for RMS and Eclipse has defaults, that can be changed through options. First line in this file is used as a header with column names.
rft_prefix: If specified, this is added as a prefix to all well names in the input file. If aliases are in use, the RMS names of wells in the alias file must include the prefix.
exportdir: A directory for where to dump the resulting txt, obs and well_date file. The directory must exist upfront.
welldatefile: A filename that will be written to (do not include the path, it is written to the directory specified in exportdir) that will contain data to be provided to GENDATA_RFT.
interpolation: Interpolation setting, choose between linear and cubic. Default is cubic. This affects how points along the wellpath are interpolated.
absolute_error: Floating point value for absolute error to use on those observations where “ERROR” is not specified in the CSV file.
relative_error: Relative error to be used for those observations where “ERROR” is not specified in the CSV file. The error is calculated as this relative error times the VALUE column in the CSV.
input_dframe: Alternative to using input_file, you may provide a Pandas dataframe with the same columns, directly without going via CSV on disk.
rms_name: The column name in the CSV alias_file with RMS names. Defaults to RMS_WELL_NAME.
ecl_name: The column name in the CSV alias_file with Eclipse names. Defaults to ECLIPSE_WELL_NAME.
zonename: Name of zone parameter in RMS grid (necessary for verifying if RFT observations are in the correct zone in the grid)
gridname: Name of the RMS gridmodel to be requested for mapping (x,y,z) to gridcells and into zone queries.
trajectory_name: The trajectory name for the wellpaths in the RMS project. Defaults to “Drilled trajectory”.
clipboard_folder: Optional name of clipboard folder for storing RFT points inside the RMS project. The folder will be created if not present.

create_rft_ertobs

Usage

See also