Skip to content

digirock

digirock is a Python framework for modelling digital rock models. It uses four abstract building blocks Elements, Blenders, Transformers and Switches to build flexible and moduluar rock models, all on top of the Python scientific stack to make things easy (numpy, xarray and pandas).

  • The Element class is the fundamental building block, all other classes a based on it and it provides most of the functionality. Element classes are extended to have rock or fluid properties or indeed any other property you want to model.
  • The Blend class is used to combine any of the building block classes in different ways the blending method is implemented within a new class. Examples of a blend method include a Wood's Fluid or Voight-Reuss-Hill average for a mineral composite.
  • The Transform class is a pipeline that takes one of the building block classes and performs an operation on either the upgoing or downgoing properties of the rock. This is useful for transforming inputs or performing adjustments to lower level block outputs. For example, Nur's Critical porosity transform is a transformer.
  • The Switch class is used when you zones or regions in your data that use completely different models. The switch class for example can be used apply the correct fluid in different PVT Zones, or a different Rock model per facies zone.

digirock has primarily been implemented for clastic petro-elastic modelling but the framework is flexible enough for users to implement there own Elements and Models following the guidelines and examples from withing this documentation. If you write a new model, or implement an old one, please considering submitting back to the digirock project (contributing).

Quick Start

See the quick start example in the user guide.

Installation

Installing with pip

digirock is available via pip install.

Text Only
pip install digirock

Installing from source

Clone the repository

Text Only
git clone http://github.com/trhallam/digirock

and install using pip

Text Only
cd digirock
pip install .