The petroleum industry requires new technologies to improve the economics of exploration and production. Digital rock physics is a methodology that seeks to revolutionize reservoir characterization, an essential step in reservoir assessment, using computational methods. A combination of X-ray computed microtomography, digital pore network modeling, and 3D printing technology represents a novel workflow for transferring digital rock models into tangible samples that can be manufactured in a variety of materials and tested with standard laboratory equipment. Accurate replication of pore networks depends on the resolution of tomographic images, rock sample size, statistical algorithms for digital modeling, and the resolution of 3D printing. We performed this integrated approach on a sample of Idaho Gray Sandstone with an estimated porosity of 29% and permeability of 2200 mD. Tomographic images were collected at resolutions of 30 and 7 μm per voxel. This allowed the creation of digital porosity models segmented into grains and pores. Surfaces separating pores from grains were extracted from the digital rock volume and 3D printed in plastic as upscaled tangible models. Two model types, normal (with pores as voids) and inverse (with pores as solid), allowed visualization of the geometry of the grain matrix and topology of pores, while allowing characterization of pore connectivity. The current resolution of commodity 3D printers with a plastic filament (30 μm for pore space and 16 μm for grain matrix) is too low to precisely reproduce the Idaho Gray Sandstone at its original scale. However, the workflow described here also applies to advanced high-resolution 3D printers, which have been becoming more affordable with time. In summary, with its scale flexibility and fast manufacturing time, 3D printing has the potential to become a powerful tool for reservoir characterization.
IntroductionThe petroleum industry has always been faced with a problem of correlation across multiple scales of investigation, for example, among seismic, well log, and core data. Although seismic profiles and wireline logs capture field-scale features, and while petrography and computed microtomography (CT) provide insight into pore-to bed-scale features of reservoir rocks, uncertainty in petrophysical properties due to differences in scale still persists. Moreover, calculations of petrophysical properties from microscopy images do not always match experimental data from cuttings and core plugs due to deficiencies in computational algorithms used for pore network modeling and fluid transport simulations. The physical pore network is an essential element of petroleum reservoir that is defined by the sizes, orientations, and connectivity of pores in a rock. Thus, accurate detection of pore space in reservoir rocks is crucial for a proper assessment of porosity-permeability relationships that ultimately affect prediction of hydrocarbon flow and ultimate recovery.