Methods And Applications Of Digital Rock Physics For Complex Reservoir Studies

Digital rock physics(DRP)is a newly developed technology stemmed from the evolution of statistical physics and computer sciences.DRP obtains 3D digital images of real core samples through high-resolution X-ray CT(or other techniques)and then turns the 3D information into digitized core structure body by image processing methods.The physical procedures including fluid flow,electrical particle flow and elastic wave propagation,etc.,are carried out by applying lattice Boltzmann method(LBM)or other numerical methods through virtual reality procedure.In this way,the reservoir rock parameters such as pore-throat size distribution,porosity,permeability,formation resistivity factor,elastic properties,and nuclear magnetic resonance(NMR)properties can be ?measured? or calculated.DRP can have a number of advantages over traditional lab measurements: Firstly,it can achieve pore size scale observations and measurements hence substantially improves the resolution of these studies and enable to reveal the physical mechanisms of micro-and nano-scale;secondly,it may significantly reduce the time for measuring some critical parameters(e.g.relative permeability)and hence to reduce the measurement cost;last but not the least,DRP can provide more parameters for reservoir studies and development more efficiently.In this study,firstly,the methods of digital core reconstruction are analyzed and reviewed,the basic principle of X-ray CT scanner for obtaining digital images is elaborated,and the basic procedure of DRP image processing is summarized.After analyzing and improving the numerical simulation methods for calculation of rock properties,the parallel lattice Boltzmann method(PLBM)is developed and implemented on a high-performance-computing-cluster(HPCC).This algorithm has now been installed to our in-house software ?SAKHER2.0? as a module,which can be used to do DRP image processing and numerical simulations.By comparing the simulated results obtained by ?SAKHER2.0? with lab measurements of real rock samples,the PLBM and the software have been validated for its accuracy and efficiency.This improved PLBM is applied to calculate the porosity and directional permeability for three parts(upper,middle,and lower)of each for five carbonate rock samples.By comparing the simulations with physical lab measurements,the heterogeneity of these samples and the permeability anisotropy of local sampling formation are studied.The results show that the carbonate rocks samples have strong heterogeneity and the horizontal permeability of local sampling formation is higher than its vertical permeability.On this basis,a new DRP workflow is proposed to evaluate the heterogeneity of carbonate rock samples.Some parameters of the improved PLBM are optimized,which is then suitable for simulating micro flow of shale gas.After the validation of the algorithm by comparing it with the analytical solution of a tube flow model,the modified PLBM is applied to study some factors(e.g.rarefaction effect,velocity slip,and roughness effect)that have an impact on micro flow of shale gas in micro/nano pores.Based on these studies,the expression of Knudsen number,the accommodation coefficient of velocity slip,and the correction parameter of apparent permeability are modified and then the gas flow in real pores is simulated with Nano-CT images of two organic-rich mudrock samples.In addition,multi-scale image analysis technics of DRP are studied by depicting the processing of a rock sample that has digital images with three different scales and resolutions.A method of multi-scale image analysis is proposed and well used for upscaling analysis and calculation,which is that some parameters(e.g.,porosity,image resolution,sample size,etc.)are considered and applied to do statistical analysis using the computed rock properties of multi-scale images.