Prediction of Foam Rheology Models Parameters Utilizing Machine Learning Tools.

Rheological models are usually used to predict foamed fluid viscosity; however, obtaining the model constants under various conditions is challenging. Hence, this paper investigated the effect of different variables on foam rheology, such as shear rate, temperature, pressure, surfactant types, gas phase, and salinity, using a high-pressure high-temperature foam rheometer. Power-law, Bingham plastic, and Casson fluid models fit the experimental data well. Therefore, the data were fed to different machine learning techniques to evaluate the rheological model constants with different features. In this study, seven different machine learning techniques have been applied to predict the rheological models' constants, including decision tree, random forest, XGBoost (XGB), adaptive gradient boosting, gradient boosting, support vector regression, and voting regression. We evaluated the performance of our machine learning models using the coefficient of determination (R2), cross-plots, root-mean-square error, and average absolute percentage error. Based on the prediction outcomes, the XGB model outperformed the other ML models. The XGB model exhibited remarkably low error rates, achieving a prediction accuracy of 95% under ideal conditions. Furthermore, our prediction results demonstrated that the Casson model accurately captured the rheological behavior of the foam. Additionally, we used Pearson's correlation coefficients to assess the significance of various properties in relation to the constants within the rheological models. It is evident that the XGB model makes predictions with nearly all features contributing significantly, while other machine learning techniques rely more heavily on specific features over others. The proposed methodology can minimize the experimental cost of measuring rheological parameters and serves as a quick assessment tool.

Implementation of contact line motion based on the phase-field lattice Boltzmann method.

This paper proposes a strategy to implement the free-energy-based wetting boundary condition within the phase-field lattice Boltzmann method. The greatest advantage of the proposed method is that the implementation of contact line motion can be significantly simplified while still maintaining good accuracy. For this purpose, the liquid-solid free energy is treated as a part of the chemical potential instead of the boundary condition, thus avoiding complicated interpolations with irregular geometries. Several numerical testing cases, including droplet spreading processes on the idea flat, inclined, and curved boundaries, are conducted, and the results demonstrate that the proposed method has good ability and satisfactory accuracy to simulate contact line motions.

Fractured Geothermal Reservoir Using CO2 as Geofluid: Numerical Analysis and Machine Learning Modeling.

The effect of natural fractures, their orientation, and their interaction with hydraulic fractures on the extraction of heat and the extension of injection fluid are fully examined. A fully coupled and dynamic thermo-hydro-mechanical (THM) model is utilized to examine the behavior of a fractured geothermal reservoir with supercritical CO2 as a geofluid. The interaction between natural fracture and hydraulic fracture, as well as the type and location of geofluids, influences the production temperature, thermal strain, mechanical strains, and effective stress in rock/fractures in the reservoir. A mathematical model is developed by using the fully connected neural network (FCN) model to establish a mathematical relationship between the reservoir parameters and the temperature. The response surface methodology is applied for qualitative numerical experimentation. It is found that the developed FCN model can be utilized to forecast the temporal variation of temperature in the production well to a desired level using FCN. Therefore, the numerical simulations developed with the FCN method can be useful tools to investigate the temperature evolution with higher accuracy.

Predicting wettability of mineral/CO2/brine systems via data-driven machine learning modeling: Implications for carbon geo-sequestration.

Effectively storing carbon dioxide (CO2) in geological formations synergizes with algal-based removal technology, enhancing carbon capture efficiency, leveraging biological processes for sustainable, long-term sequestration while aiding ecosystem restoration. On the other hand, geological carbon storage effectiveness depends on the interactions and wettability of rock, CO2, and brine. Rock wettability during storage determines the CO2/brine distribution, maximum storage capacity, and trapping potential. Due to the high CO2 reactivity and damage risk, an experimental assessment of the CO2 wettability on storage/caprocks is challenging. Data-driven machine learning (ML) models provide an efficient and less strenuous alternative, enabling research at geological storage conditions that are impossible or hazardous to achieve in the laboratory. This study used robust ML models, including fully connected feedforward neural networks (FCFNNs), extreme gradient boosting, k-nearest neighbors, decision trees, adaptive boosting, and random forest, to model the wettability of the CO2/brine and rock minerals (quartz and mica) in a ternary system under varying conditions. Exploratory data analysis methods were used to examine the experimental data. The GridSearchCV and Kfold cross-validation approaches were implemented to augment the performance abilities of the ML models. In addition, sensitivity plots were generated to study the influence of individual parameters on the model performance. The results indicated that the applied ML models accurately predicted the wettability behavior of the mineral/CO2/brine system under various operating conditions, where FCFNN performed better than other ML techniques with an R2 above 0.98 and an error of less than 3%.

Machine Learning-Based Accelerated Approaches to Infer Breakdown Pressure of Several Unconventional Rock Types.

Unconventional oil and gas reservoirs are usually classified by extremely low porosity and permeability values. The most economical way to produce hydrocarbons from such reservoirs is by creating artificially induced channels. To effectively design hydraulic fracturing jobs, accurate values of rock breakdown pressure are needed. Conducting hydraulic fracturing experiments in the laboratory is a very expensive and time-consuming process. Therefore, in this study, different machine learning (ML) models were efficiently utilized to predict the breakdown pressure of tight rocks. In the first part of the study, to measure the breakdown pressures, a comprehensive hydraulic fracturing experimental study was conducted on various rock specimens. A total of 130 experiments were conducted on different rock types such as shales, sandstone, tight carbonates, and synthetic samples. Rock mechanical properties such as Young's modulus (E), Poisson's ratio (ν), unconfined compressive strength, and indirect tensile strength (σt) were measured before conducting hydraulic fracturing tests. ML models were used to correlate the breakdown pressure of the rock as a function of fracturing experimental conditions and rock properties. In the ML model, we considered experimental conditions, including the injection rate, overburden pressures, and fracturing fluid viscosity, and rock properties including Young's modulus (E), Poisson's ratio (ν), UCS, and indirect tensile strength (σt), porosity, permeability, and bulk density. ML models include artificial neural networks (ANNs), random forests, decision trees, and the K-nearest neighbor. During training of ML models, the model hyperparameters were optimized by the grid-search optimization approach. With the optimal setting of the ML models, the breakdown pressure of the unconventional formation was predicted with an accuracy of 95%. The accuracy of all ML techniques was quite similar; however, an explicit empirical correlation from the ANN technique is proposed. The empirical correlation is the function of all input features and can be used as a standalone package in any software. The proposed methodology to predict the breakdown pressure of unconventional rocks can minimize the laboratory experimental cost of measuring fracture parameters and can be used as a quick assessment tool to evaluate the development prospect of unconventional tight rocks.

Improving deep learning performance for predicting large-scale geological [Formula: see text] sequestration modeling through feature coarsening.

Physics-based reservoir simulation for fluid flow in porous media is a numerical simulation method to predict the temporal-spatial patterns of state variables (e.g. pressure p) in porous media, and usually requires prohibitively high computational expense due to its non-linearity and the large number of degrees of freedom (DoF). This work describes a deep learning (DL) workflow to predict the pressure evolution as fluid flows in large-scale 3-dimensional(3D) heterogeneous porous media. In particular, we develop an efficient feature coarsening technique to extract the most representative information and perform the training and prediction of DL at the coarse scale, and further recover the resolution at the fine scale by spatial interpolation. We validate the DL approach to predict pressure field against physics-based simulation data for a field-scale 3D geologic [Formula: see text] sequestration reservoir model. We evaluate the impact of feature coarsening on DL performance, and observe that the feature coarsening not only decreases the training time by [Formula: see text] and reduces the memory consumption by [Formula: see text], but also maintains temporal error [Formula: see text] on average. Besides, the DL workflow provides predictive efficiency with 1406 times speedup compared to physics-based numerical simulation. The key findings from this research significantly improve the training and prediction efficiency of deep learning model to deal with large-scale heterogeneous reservoir models, and thus it can also be further applied to accelerate workflows of history matching and reservoir optimization for close-loop reservoir management.