The module implements two types of functionality for initializating simulation setup: from an input deck or from an output deck (restart file). The latter is not fully complete.
Input deck: Releases prior to 2012a featured the 'readGRDECL' function for inputting a selection of keywords that mostly described grids and cell-based properties. The new input support, consists of two main functions, 'readEclipseDeck' and 'convertDeckUnits'. The former does the actual data reading while the latter converts various units of measurements (e.g., permeability values) into MRST's strict SI-only unit conventions. Along with the input reading, we have also added several functions to construct fundamental MRST objects (e.g., grids, fluids, rock-properties or wells) from the descriptions of a particular simulation deck.
Restart files: Recently, we have also included functionality for reading and processing ECLIPSE unformatted output files. These can then be used to set up a new simulation or continue from an existing one.
In the first part of this tutorial, you can see how functionality from the deckformat module is used to initialize the well-known SPE1 benchmark case. That is, we read the input deck, extract grid, rock properties, construct a fluid object, and set up a simulation schedule.
This example reads ECLIPSE unformatted output files from a simulation based on the SPE-9 benchmark. The MRST grid structure (G) is generated using output from INIT and EGRID files, whereas the solution structure is generated using output from an UNRST file.
This example continues the reading of unformatted output and shows how to only extract a few of the simulation steps, converting the input so that it can be used for visualization, etc
Here, the input from a previous SPE9 simulation is used as input to the interactive flow-diagnostics viewer from the diagnostics module.
Published October 1, 2012