Using Digital Rock Modeling to Estimate Permeability and Capillary Pressure from NMR and Geochemical Logs

Abstract Permeability is one of the most important characteristics of hydrocarbon-bearing formations. Many approaches exist for estimating permeability. Formation permeability is often measured in the laboratory from core samples or evaluated from well test data. However, core analysis and well test data are expensive and available only from a limited number of wells in a field. On the other hand, logging data are available from almost all wells. Therefore, accurate prediction of the formation permeability using logging data becomes very attractive. This paper describes a unique integrated modeling approach to predict formation permeability. By combining digital rock physics (DRP) and downhole logging measurements, formation permeability and capillary pressure were predicted using computational analyses. In this approach, a representative geometrical rock model for the formation at each given depth of interest was numerically generated. These numerical rock models are constrained by the formation parameters derived from NMR and geochemical logging data, i.e., all input parameters for the geometrical rock model such as the rock density, porosity, grain-size distribution, and grain mineralogy, are directly or indirectly obtained from downhole logging measurements. Then, using the geometrical rock model as input, pore size information, capillary-pressure curve, and permeabilities were obtained by computational analyses, including fluid flow simulations based on the lattice-Boltzmann method (LBM). To demonstrate the method's feasibility and applicability, the proposed modelling approach was used on two sets of field data from a North Sea well and an Austin testing well. Permeability and capillary pressure were experimentally determined from laboratory measurements on cores and compared to the simulation results. The predicted permeability values from the proposed DRP approach were in good agreement with measured values from cores. The drainage capillary-pressure curves derived from rock models also matched well with laboratory measurements on core samples. The promising results generated from this study demonstrate the feasibility of a digital core analysis method that integrates a geometrical rock model with downhole logging measurements to accurately and reliably predict formation petrophysical properties in a cost-effective and timely manner.