Using NMR Log to Estimate Permeability in One of the Carbonate Formations in South of Iran

Permeability is an elusive parameter in hydrocarbon reservoirs as it is very difficult, if not impossible, to determine precisely and directly from current subsurface logging technologies. In this research, an attempt is made to test some methods for estimating permeability as a function of depth from Nuclear Magnetic Resonance (NMR) logging in one of carbonate reservoirs in south of Iran. For accurate permeability estimation, an Artificial Neural Network (ANN) model with two different inputs is applied. In the first case, NMR porosity has been used as input data but in the second case there is no NMR data as input and core porosity has been used. Also three NMR models such as average-T2, free-fluid and Swanson model, have been used for permeability estimation. The results of all these methods are compared with the core permeability. The trends of permeabilities obtained by NMR models have good compatibility with core permeability, so they can be used for in-situ permeability estimation. The results of ANN model shows that using NMR porosity, beside traditional log data, as input for ANN leads to considerably increase in correlation coefficient relative using core porosity. So it can be used as a reliable method for permeability prediction.