A computational analysis for quantitative evaluation of petrol-physical properties of rock fluids based on Bloch NMR diffusion model for porous media

Abstract

Estimating permeability from grain-size distributions or from well logs is attractive but very difficult. The difficulties are inherent in many petroleum-bearing reservoirs and complex mineralogy earth formations where existing nuclear magnetic resonance (NMR) models require modifications to work effectively. In this paper, we present a generally applicable and simple approach which may yield useful information from NMR signals of different petro-physical properties and porosity. This approach is a model of the Bloch NMR diffusion equation for complex pore geometries in which the transverse magnetization is obtained as function of reservoir chemical (relaxation) and physical properties. The NMR signal is also shown to be dependent on the tortuosity and relaxation rate of rocks fluid so that reservoirs comprised of mixed lithology and mineralogy can be easily evaluated. The computational tools obtained in this study are useful for repetitive data processing which is otherwise difficult due to hardware limitations and logistic issues. • We model NMR diffusion equation for quantitative analyses of complex pore geometry. • We obtained expression for M y as function of T 1 , T 2 and petrol-physical properties. • The colour maps show the model is able to differentiate components of fluid mixture. • The model interpolates (colour maps) for points of interest not on the data table. • The model could not show contrast between brine and oil via diffusion process.