Quantification of Multi-Phase Fluid Saturations in Complex Pore Geometries From Simulations of Nuclear Magnetic Resonance Measurements

摘要

We develop a numerical algorithm to simulate nuclearrnmagnetic resonance (NMR) measurements in the presence ofrnconstant magnetic field gradients. The algorithm is based onrnMonte Carlo conditional random walks in restricted andrnunrestricted space. Simulations can be performed of threedimensionalrn(3D) porous media that include both arbitraryrnbimodal pore distributions and multi-phase fluid saturations.rnThe ability to account for the presence of a constant externalrnmagnetic field gradient allows us to replicate actual wellrnlogging conditions that include the effect of CMPG pulsernsequences at a microscopic level. This is accomplished byrnsimulating pulse acquisition techniques that include multiplerninter-echo times (TE) similar to those currently used by thernwell-logging industry. Benchmark examples are presented tornvalidate the accuracy and internal consistency of our algorithmrnagainst previously published results for the case of a nullrnmagnetic field gradient. Validation examples are alsornpresented against actual NMR measurements performed onrncore samples of carbonate rock formations.rnInterpretation work is focused to the petrophysicalrnassessment of both partial oil/water saturations and porernstructures exhibiting hydraulic coupling. Simulation examplesrnare designed to quantify whether the inclusion of diffusionrnunder a magnetic field gradient can improve the interpretationrnof multi-phase fluid saturations when hydraulic coupling isrnsignificant. The simulation algorithm sheds light to new NMRrndata acquisition strategies that could be used to improve therndetection and quantification of (a) fluid types, (b) complexrnfluid saturations, and (c) complex pore geometries.