PetroGrapher:: managing petrographic data and knowledge using an intelligent database application

This paper describes the PetroGrapher system, an intelligent database application to support petrographic analysis, interpretation of oil reservoir rocks, and management of relevant data using resources from both knowledge-based system technology and database technology. In this project, the visual tacit knowledge applied in petrographic analysis was rendered explicit through the collection of cases (rock descriptions), which were then used in the development of a domain ontology organized in a partonomy. Expert-level basic features, which we call 'visual chunks', were identified. The cases were further compared against the ontology to elucidate the relations between features in descriptions of rocks, visual chunks and expert interpretations. The domain knowledge was represented through a set of frames and knowledge graphs. The knowledge graphs are applied to recognize the visual chunks in the user data and retrieved the related interpretation. The system was developed as a structure tightly coupled with a relational database system, which acts as a repository for the knowledge base and the user data, and an object-oriented component, which preserves the semantics of data and develops inferences. The system was validated by three groups of users with different levels of expertise. (C) 2003 Elsevier Ltd. All rights reserved.