Functionality and configuration
Data which sim2sumo supports
sim2sumo can extract and upload the following data to Sumo. For more information on these datatypes, see the res2df docs.
Default data types
Note
These data are uploaded by default, with no additional configuration required:
summary
rft
satfunc (relperm curves)
gruptree (production network)
wellcompletiondata
Warning
The following datatypes need additional configuration:
grid (INIT “initial” and restart “dynamic” 3D grid properties)
pvt (PVT tables)
trans (transmissibilities)
vfp (lift curves; one table for all lift curves)
To upload other data to Sumo, see the SUMO_UPLOAD forward model.
Configuration of sim2sumo
sim2sumo can be configured in your FMU model’s global_variables.yml file. This file is generated using fmu-config. If you are not familiar with updating global variables, see the fmu-config docs.
The simplest way to configure sim2sumo is to add a sim2sumo key to the global_variables_master.yml file.
The SIM2SUMO forward model looks for sim2sumo config in ../../fmuconfig/output/global_variables.yml by default. If your config file isn’t located here, you need to tell SIM2SUMO where to find it:
FORWARD_MODEL SIM2SUMO(<S2S_CONF_PATH>=path/to/config/file.yml)
Data types
To specify which data to upload, provide the datatypes as a list
sim2sumo:
datatypes:
- summary
- rft
- gruptree
- ...
To include all datatypes use a list with a single item “all”. This will also include the grid datatype:
sim2sumo:
datatypes:
- all
3D grid properties
To upload both “static” (INIT) and dynamic “restart” (UNRST) 3D grid properties, additional configuration is required.
Examples can be found here.
Paths to files
By default, SIM2SUMO looks for all DATA files in the runpath <user>/<case>/realization-*/<ensemble>.
It is possible to configure where results will be extracted from in two ways:
Specify file paths, or file stubs (without an extension):
sim2sumo:
datafile:
- eclipse/model/DROGON
- eclipse/model/DROGON-0.DATA
Specify folder paths:
sim2sumo:
datafile:
- eclipse/model/
Running several DATA files in one runpath, or running a prediction case from a restart?
Warning
Data consumers expect to see one of each datatype for each realisation, e.g.
one summary file for each realisation. If you run several simulations in the
same runpath (<user>/<case>/realization-*/<ensemble>) or if you symlink
to a restart file when running a prediction ensemble, you should specify the
path of the datafiles to upload. If you don’t do this, you may find your
results are not displayed or are displayed incorrectly in applications
reading data from Sumo.
Example: a prediction run is found in
realization-*/pred/eclipse/model and a symlinked restart file is located
in realization-*/pred/restart/model. The symlinked restart file can be
omitted by specifying just the folder path of the prediction run:
sim2sumo:
datafile:
- eclipse/model/