Português English

Dissertação de Fernando Cicconeto

Detalhes do Evento

Aluno: Fernando Cicconeto
Orientadora: Profa. Dra. Mara Abel
Título: GeoReservoir: An Ontology for Deep-marine Depositional System Description

Linha de Pesquisa: Aprendizado de Máquina, Representação de Conhecimento e Raciocínio

Data:  25/02/2021
Horário: 10h

Esta banca ocorrerá excepcionalmente de forma totalmente remota. Interessados em assistir a defesa poderão acessar a sala virtual através do link: https://mconf.ufrgs.br/webconf/00006784

Banca Examinadora:
– Prof. Dr. Michel Perrin – ESNMPFrança
– Prof. Dr. Sandro Rama Fiorini – IBM Research Center Brasil
– Profa. Dra. Renata de Matos Galante – UFRGS

Presidente da Banca: Profa. Dra. Mara Abel

Abstract: Ontology is the logical theory that accounts for a domain’s vocabulary intended meaning, which allows us to develop computational artifacts that explicitly and formally define the conceptualization associated with a particular set of vocabulary inside a domain. This master thesis presents the result of the ontological analysis of the terminology adopted in sedimentological studies and proposes a domain ontology to support the description of deep-marine depositional system geological occurrences. The domain in focus is of special interest primarily because of its economic importance: deep-marine deposits constitute the main type of petroleum reservoir explored nowadays in the world. Further on, the task in focus demands new approaches in conceptual modeling methodologies because the available terminology describes an interpretation of the studied occurrences instead of a rational description of the visually recognized objects. The ontological substrate helps in choose and define the precise vocabulary requested for the descriptive capturing of data. Therefore, the project aimed to collect and disambiguate the Geology terminology to describe reservoir occurrences in deep-marine depositional systems and build a domain ontology for reservoir description: the GeoReservoir. A team of professional reservoir geologists supported the knowledge acquisition process. We developed the ontology in iterative steps from an initial prototype to a complete taxonomy of entities, relations, and qualities, increasingly refined by the experts. The process implied the specialization of the concepts of two previously existent upper-level ontologies to build GeoReservoir: the Basic Formal Ontology (BFO) and the GeoCore ontology. This framework provided us the necessary structure and organization to focus our analysis in the domain, reducing the development effort. The resulting ontology includes a taxonomy of sedimentary units, lithologies, geological structures, and contact types. This ontology is complemented by a spatial relations ontology that formalizes the mereotopological relations of the described geological objects. The geological occurrences of Karoo turbidites in South Africa offer the data of a case study for testing and initial validation of the proposed ontology. The ontology is under continuous expansion through the inclusion of terminology for other depositional environments. A public accessible repository stores an implemented version of the model and the application case study.

Keywords: Ontology, Petroleum Geology, Reservoir geometry