Fast permeability estimation using NMR well logging data log-normal decomposition

Abstract

Nuclear magnetic resonance (NMR) well logging and NMR core analysis provide a non-invasive approach to a fast reservoir rock pore structures evaluation. The NMR data is presented as a time distribution called transverse relaxation time spectrum (T2 spectrum) and, with proper adjustment, the area under the T2 distribution curve is directly related to the porosity of the reservoir rock. Based on the assumption that immovable and free fluid are present in pores of different sizes, NMR well logging and core analysis use these differences to characterize the fluids found inside the rock pores. In this way, a fixed T2 value can be selected as a cutoff. The amplitudes of the T2 spectrum that have their T2 coordinates below this fixed value are related to the fluids that are in the small pores (immovable fluids), and those that have their T2 coordinates above this value are related to the large pores (free or movable fluids). This T2 value is called T2cutoff and can be estimated in a laboratory with NMR data acquisition on water-saturated core samples. This is done by comparing the NMR acquisition data from a plug fully saturated with water with the NMR acquisition data taken from the same partially saturated plug. In this work, a new method to estimate the pore size distribution, a variable T2cutoff curve, and the Timur–Coates permeability curve of the formation along the reservoir rock is presented. The method is based on the decomposition in log-normal components of the T2 spectra of NMR well logging and on the permeability and porosity of plugs, measured in a laboratory.