Representing subsurface knowledge using RESQML
Conference Paper
Overview
Identity
Additional Document Info
View All
Overview
abstract
Copyright 2015, Society of Petroleum Engineers. Exploration and production workflows continue to evolve in complexity. We aspire to transfer both data and interpretations across a wide range of domains, often using a variety of software applications throughout the reservoir management life cycle. Keeping the model updated with new information while characterizing the range of uncertainty is a continual challenge. RESQML is the data exchange format used in the upstream oil and gas industry for transferring earth models between software applications in a vendor-neutral, open, and explicit format. RESQML VI. 1 focused strictly on data exchange; only individual elements (such as interpreted picks, surfaces and reservoir grid) could be exchanged. In RESQML V2.0 (published in September 2014), capabilities have been expanded to provide a mechanism for transferring the relationship information between RESQML data-objects and to define more data-objects, for example, wells and unstructured simulation grids. With these new relationship capabilities, it is now possible to exchange structural & stratigraphic and reservoir frameworks, with their associated properties. This paper describes how the RESQML format has evolved from a single exchange of independent geometry and property representations of horizons, faults, and corner-point structured grids, to a comprehensive consistent earth model exchange that can capture, represent and transfer many of the relationships and interpretations contained in a reservoir model. Examples are drawn from the use cases and unit tests that demonstrate the use of RESQML to support partial model transfers, to describe structural and stratigraphic frameworks used in larger scale models (such as velocity models), and associate features and interpretations of those frameworks with specific representations, especially reservoir grids. The paper shows how these enhancements can be used to update and execute sensitivity workflows on a reservoir model, while describing uncertainty in the data used to build the reservoir framework.
