No abstract

Fine-scale models that represent first-principles physics are challenging to represent at larger scales of interest in many application areas. In nanoporous media such as tight-shale formations, where the typical pore size is less than 50 nm, confinement effects play a significant role in how fluids behave. At these scales, fluids are under confinement, affecting key properties such as density, viscosity, adsorption, etc. Pore-scale Lattice Boltzmann Methods (LBM) can simulate flow in complex pore structures relevant to predicting hydrocarbon production, but must be corrected to account for confinement effects. Molecular dynamics (MD) can model confinement effects but is computationally expensive in comparison. The hurdle to bridging MD with LBM is the computational expense of MD simulations needed to perform this correction. Here, we build a Machine Learning (ML) surrogate model that captures adsorption effects across a wide range of parameter space and bridges the MD and LBM scales using a relatively small number of MD calculations. The model computes upscaled adsorption parameters across varying density, temperature, and pore width. The ML model is 7 orders of magnitude faster than brute force MD. This workflow is agnostic to the physical system and could be generalized to further scale-bridging applications. Multi-scale physics problems are found in all scientific disciplines. Prominent examples can be found in material science 1-3 , biology 4 , chemistry 5-9 , and geosciences 10-12. Typically, information from computationally intensive fine-scale models have to be translated or upscaled into faster coarse-scale models to solve the problem at the scale of interest. A problem of great scientific and economic interest is the flow of hydrocarbon in nanoporous shale. Traditional porous media approaches such as the LBM allow for complex pore geometries but need to be provided with effective properties that account for nanoconfinement effects in order to accurately simulate mass transport at the continuum scale 13. Atomistic simulations such as Molecular Dynamics (MD) capture nanoconfinement effects accurately, but are limited to a few pores as they are computationally intractable to simulate for mesoscopic pore geometries. There is a need for approaches that efficiently bridge these two scales without compromising accuracy. Recently, Machine Learning (ML) has shown great promise in accelerating physics-based models that makes it feasible to build a scale-bridging framework 14-16. The applications include fracture propagation in brittle materials 17 , computational fluid dynamics 18 and molecular dynamics 19. On another dimension, Machine Learning

The dynamics of asphaltene molecules is highly impacted by both the nature of the solvent and the physical conditions of the system. We performed molecular simulation to investigate the dynamic behavior of asphaltene during gas flooding. We also consulted the experimental observations for validation purposes when available. Two structures representing the archipelago and continental types are used, whose aggregation and interactions are studied in methane (C1), propane (C3), carbon dioxide (CO2), heptane (C7), and toluene as pure solvents, binary mixtures of toluene and either C1, C3, or CO2, and a representative oil composition. The continental structure is used afterward to evaluate the impact of temperature, pressure, and resin content on the aggregation dynamics in CO2 mixtures. Interestingly, the solvating power of CO2 is dependent on the asphaltene structure where inhibitor-like behavior is observed for the continental structure and precipitator-like behavior is observed for the archipelago structure. The solvent quality is highly correlated with the solvent s ability to replace the interactions among asphaltene molecules with interactions between asphaltene and solvent. The aggregate size is reduced by temperature and enhanced by pressure in CO2. However, limited effect is reported for resins on asphaltene dynamics in CO2. The aggregation of asphaltene is impacted by the physical state of CO2 as its solvating power to asphaltene is significantly enhanced in its supercritical state. Nonetheless, this impact is limited when CO2 is introduced to a representative oil mixture.

No abstract

Predicting the spatial configuration of gas molecules in nanopores of shale formations is crucial for fluid flow forecasting and hydrocarbon reserves estimation. The key challenge in these tight formations is that the majority of the pore sizes are less than 50 nm. At this scale, the fluid properties are affected by nanoconfinement effects due to the increased fluid-solid interactions. For instance, gas adsorption to the pore walls could account for up to 85% of the total hydrocarbon volume in a tight reservoir. Although there are analytical solutions that describe this phenomenon for simple geometries, they are not suitable for describing realistic pores, where surface roughness and geometric anisotropy play important roles. To describe these, molecular dynamics (MD) simulations are used since they consider fluid-solid and fluid-fluid interactions at the molecular level. However, MD simulations are computationally expensive, and are not able to simulate scales larger than a few connected nanopores. Alternatively, mesoscale simulation methods that handle larger domains with complex pore geometries (i.e. the lattice-Boltzmann method) cannot directly account for nanoconfinement effects, resulting in grossly inaccurate predictions. To incorporate these effects into larger domains with complex geometries, it is necessary to accelerate the computation. We present a method for building and training physics-based deep learning surrogate models to carry out fast and accurate predictions of molecular configurations of gas inside nanopores. Since training deep learning models requires extensive databases that are computationally expensive to create, we employ active learning (AL). AL reduces the overhead of creating comprehensive sets of high-fidelity data by determining where the model uncertainty is greatest, and running simulations on the fly to minimize it. The proposed workflow enables nanoconfinement effects to be rigorously considered at the mesoscale where complex connected sets of nanopores control key applications such as hydrocarbon recovery and CO2 sequestration.