A Conceptual Framework for Rock Data Integration in Reservoir Models Based on Ontologies

Abstract

In the domain of E&P petroleum chain, the authors have developed a study and proposed a conceptual framework for inserting direct rock data into reservoir models, through calibration of well logs. This type of data is often ignored or manual processed in actual petroleum reservoir modeling activities, due to its high cost of acquisition, or due to the high complexity for its modeling, interpretation and extrapolation. Direct rock data is the data that is acquired by direct observation of the rock, like descriptions of well cores, instead of indirect data, like seismic, well logs, etc. Directly observed data is important because it allows calibration of indirect interpretation methods, detection of errors in their evaluations and also to verify rock properties that is not possible when using indirect data. The authors claim that a well defined ontology can help in describing reservoir petrofacies, which are aggregates of rock data that can be detected through a special signature in the registers of indirectly collected rock data and, therefore, they can use to integrate rock data in reservoir models.