Zoneamento de poços através da inferência Fuzzy

Abstract

Well zoning may be understood as the geological characterization (location and facies description) of each layer crossed by the borehole trajectory. Well zoning is a common activity in conventional core analysis and important for petroleum geology, assisting the construction of stratigraphic column and also for petroleum engineering aiding in the development of the well exploitation plan. This work presents a method for well zoning wells of non cored boreholes, so that the information gained in these wells can contribute to improve the knowledge of sedimentology and oilfield engineering. The method showed here uses the core description for building the knowledge base of a fuzzy inference system, which operates with P parameter (a new combination of density log and sonic log), parameter M (M-N plot) and the natural gamma ray log and the deep resistivity log. Operation of this fuzzy inference system using log data from non cored borehole produces the well zoning of each non cored borehole. This method is presented with synthetic data satisfying the petrophysical model and the Archie Law, and real data of two cored boreholes from the Namorado oilfield, in the Campos Basin.