Permeability Estimates of Carbonate Reservoir by Using Flow Unit Theory and Bayes Discriminant Seismic Inversion Method

Abstract

Summary Permeability is a critical parameter for reservoir quality prediction, reservoir simulation and production prediction. Prediction of permeability in un-cored wells and no well area are important tasks in carbonate reservoir characterization. In general, permeability can be determined from three approaches: core analysis, well test analysis, and well log analysis. While both rock tests and well tests are ideal solutions to accurately obtain the formation permeability, they provide discrete data measurements for specific depths and they are often available only for a few wells in a field. They are also time consuming and expensive measurements. A more accurate permeability estimation technique based on flow units analysis and Bayes discriminant seismic inversion has been proposed in this study. Identification of flow units in cored wells is extended to un-cored wells by use of several well logs. The Bayes discriminant can be used to predict 3D permeability for no well area, which avoids the deterministic rock physics modelling, and the predicted results have a good match with the well logging data and flow units analysis results.