Análise multivariada de dados de poços petrolíferos e modelagem geoestatística do campo curimã, bacia do Ceará

• Ano de defesa: 2022
• Autor(a) principal: Normando, Márcio Nunes
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Caracterização de reservatórios; Modelagem geológica; Análise de incertezas e sensibilidade
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/66710

The geological modeling is one of the most important phases in oil reservoir characterization, mainly regarding workload, uncertainties and impact on final results. It consists of the construction of structural, stratigraphic, and petrophysical models; the calculation of the volume of oil in place; and the analysis of uncertainties. In the other hand, defining a reservoir itself, whether conventional or not, lies on the analyses of a large amount of data, thus requiring robust techniques to deal with. The main goal of the present Thesis is to characterize the geometry and petrophysical properties of the Curimã Field, the main producer in the Ceará Basin (Brazil), having a reservoir interpreted as deltaic in the Paracuru Formation. This research integrated well log and seismic data obtained from a variety of disciplines (geophysics, sedimentology, petrophysics, geology, and engineering), aiming to define and build a reliable geological reservoir model. To better define the reservoir, the most complete log section possible (in terms of available petrophysical data) of a unique discoverer well was classified in electrofacies by applying multivariate statistics. In turn, to represent reservoir heterogeneities and integrate geological concepts and observations, geostatistical techniques were applied to a specific geological modeling of the Paracuru Formation reservoir, by using several well log and seismic data. The results for the electrofacies classification allow subdivide the Paracuru Formation in three intervals, evidencing that its topmost interval holds the better properties for an oil reservoir due the arenaceous composition and low shaliness. In turn, the integration of general data using geostatistical techniques to simulate reservoir properties provided a reliable geological model that matches, in terms of original volume in place, with that published by the Brazilian National Agency of Oil, Natural Gas and Biofuels (ANP). In addition, the sensitivity and uncertainty analysis of the variables used to build the geological model allowed to rank them according to their impact on the volume calculation - thus, from greatest to the lowest impact: i) the porosity cutoff for the NTG calculation; ii) the mean porosity value for sandstones; iii) the depth of water-oil contact. The final results were based on the analysis of 500 simulation runs and generated optimistic, most probable and pessimistic scenarios (P10, P50, and P90) of the volume in place. These results will aid a detailed dynamic assessment of the exploitation potential of the Curimã Field.