Permeability Estimation in Carbonate Reservoirs Using Electrofacies in an Oil field in the Southwest of Iran

An electrofacies in defined by a similar set of log responses that characterize a specific bed and allow it to be distinguished from other beds. Electrofacies characterization is a simple and cost-effective approach to obtaining permeability estimates in heterogeneous carbonate reservoirs using commonly available well logs. Formation permeability is often measured directly from core samples in the laboratory or evaluated from the well test data. The first method is very expensive. Moreover, the well test data or core data are not available in every well in a field; however, the majority of wells are logged. We propose a two-step approach to permeability prediction from well logs that uses nonparametric regression in conjunction with multivariate statistical analysis. First, we classify the well-log data into electrofacies types. This classification does not require any artificial subdivision of the data population and it follows naturally based on the unique characteristics of well-log measurements reflecting minerals and lithofacies within the logged interval. A combination of principal components analysis (PCA), model-based cluster analysis (MCA), and discriminant analysis is used to characterize and identify electrofacies types. Second, we apply nonparametric regression techniques to predict permeability using well logs within each electrofacies. Three nonparametric approaches are examined, namely alternating conditional expectations (ACE), support vector machine (SVM), and artificial neural networks (ANN), and the relative advantages and disadvantages are explored. For permeability predictions, the ACE model appears to outperform the other nonparametric approaches. We applied the proposed technique to a highly heterogeneous carbonate reservoir in the southwest of Iran.