Towards Multiscale Digital Rocks: Application of a Sub-Resolution Production Model to a multiscale Sandstone

Abstract

Many digital rock methodologies use a direct simulation approach, where only resolved pores are accounted for. This approach limits the types of rocks that can be analyzed, excluding some types of carbonates, unconventionals, and complex sandstones from the digital rock analysis. This is due to the challenge for single scale imaging to capture the full range of relevant pore sizes present in multiscale rocks. In this paper, a physical model is presented, within the context of an established direct simulation approach, to predict the production of hydrocarbons including the contribution of sub-resolution pores. The direct simulation component of the model employs a multiphase lattice Boltzmann method to simulate multiphase fluid flow displacement in resolved pores. In the production model, the amount of hydrocarbons present in the sub-resolution pores is identified and a physical description of the production behavior is provided. This allows a relative permeability curve to be predicted for rocks where mobile hydrocarbons are present in pores smaller than the image resolution. This simplified model for the oil movement in the unresolved pore space is based on a physical interpretation of different regions marked by simulation resolution limits in a USBM wettability test curve. The proposed methodology is applied to high-resolution microCT images of a sandstone that contains pores at multiple scales, some resolved and some not resolved. To allow for benchmarking, experimental routine and special core analysis data was also obtained. Good agreement to experimental results is observed, specifically in absolute and relative permeability. The presented multiscale model has the potential to extend the classes of reservoir rocks eligible for digital rock analysis and paves the way for further advancements in the modelling of multiscale rocks, particularly unconventionals and carbonates.