Permeability predictions in carbonate reservoirs using optimal non-parametric transformations: an application at the salt creek field unit, kent county, TX

摘要

In this paper, we have utilized a non-parametric transformation and regression technique called ACE (altermating conditional expectation) to estimate permeability from well logs at the Salt Creek Field Unit (SCFU), Texas, a heterogeneous reef carbonate reservoir. Previous attempts to derive permeability correlations at the SCFU have been less than satisfactory, leading to an over-dependence on porosity derived resevoir descriptions to predict fluid flow. Using non-parametric regression, we have now established a relationship between permeability and several common well logs that are available field-wide. These include density porosity, neutron porosity, shallow resistivity, deep resistivity and gamma ray logs. The approach adopted here also allowed us to integrate our geologic understanding of teh reservoir into the non-parametric regression, further optimizing the final correlation. We have successfully predicted permeability in a majority of the uncored wells with acceptable accuracy at SCFU. These results have led to an enhanced reservoir characterization based on flow (permeability) rather than storage (porosity). This benefits both daily operations and reservloir simulation efforts. This first, full-field application of ACE in a carbonate reservoir has demonstrated the strength and potential wide-scale use of non-parametric methods to predict permeability in heterogeneous reservoirs.