Full text loading...
-
Comparing Permeability Estimates of a Brazilian Carbonate Reservoir Using Zoning, Multiple Regression and Alternating Conditional Expectations
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 81st EAGE Conference and Exhibition 2019, Jun 2019, Volume 2019, p.1 - 5
Abstract
Carbonate reservoirs have a wide variation of permeability which valuation is made from well logs, laboratory measurements or formation tests. The permeability estimates were developed in this study from logs of a carbonate reservoir in Southeast Brazil. The following sequence was employed to perform this in the reference well: main components, cluster and discriminant function analysis; and, construction of multiple linear regression and alternating conditional expectations models with and without electrofaceis zoning. Next, the modified Lorenz plot and the flow zone indicators were utilized to determine the hydraulic flow units. All this methodology was then used in a blind test. Models without electrofacies functioned better in accord with the error metrics, especially in the case of the alternating conditional expectations. Hydraulic flow units were the most promising technique in the reference well but did not work fine in the blind test. It is reasonable to say that the three approaches are useful in predicting permeability, but, when they are applied together with electrofacies, this needs care because the poor classification of the reservoirs can lead to erroneous estimates.