Search NASA⌕ Search

SEARCH · Search NASA

Results for “reservoir computing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 73 records · Page 4

Machine learning-assisted upscaling analysis of reservoir rock core properties based on micro-computed tomography imagery

Optimum solutions for geologic modeling and reservoir simulation in industries such as oil and gas recovery and carbon capture and storage require accurate characterization of reservoir properties, which are often heterogeneous. In this study, high-quality micro-computed tomography (CT) images (1.475-μm/pixel resolution) of a sandstone core acquired from the Bell Creek oil field, USA, were used to provide nondestructive analysis of pore- and core-scale heterogeneity across measurement scales of 94–566 μm. In addition to characterizing the as-received sample, the core sample was flooded with brine to evaluate the capacity of the core sample to receive injected fluids. The micro-CT images were systematically segmented into pore spaces and grains via machine learning (ML) steps including image preprocessing, label creation using a traditional ML method based on limited manual image annotation, and finally U-Net segmentation. The segmented image stacks were reconstructed into digital cubes of various scales of voxel lengths. The 3D porosity values were calculated for all the digital cubes, and the fractal dimensions of the cubes were estimated using a box-counting method. The results showed that smaller cubes had greater heterogeneity and that the porosity values could be accurately estimated by fractal dimension and voxel lengths using ML models. For the core sample with brine flooding, the ratio of pores filled by brine to the total pore space was related to the porosity and could also be accurately estimated by porosity, fractal dimension, and voxel lengths using ML models. In conclusion, the results of this study demonstrate that the concept of fractal dimension can be a useful vector to perform upscaling analysis of sandstone rock heterogeneity from the pore to core scale and that fractal dimensions can be used to estimate porosity values and pore space-filling capacity across those scales.

58 GEOSCIENCES↗

Generalized Multiscale Finite-Element Method (GMsFEM) for elastic wave propagation in heterogeneous, anisotropic media

It is important to develop fast yet accurate numerical methods for seismic wave propagation to characterize complex geological structures and oil and gas reservoirs. However, the computational cost of conventional numerical modeling methods, such as finite-difference method and finite-element method, becomes prohibitively expensive when applied to very large models. We propose a Generalized Multiscale Finite-Element Method (GMsFEM) for elastic wave propagation in heterogeneous, anisotropic media, where we construct basis functions from multiple local problems for both the boundaries and interior of a coarse node support or coarse element. The application of multiscale basis functions can capture the fine scale medium property variations, and allows us to greatly reduce the degrees of freedom that are required to implement the modeling compared with conventional finite-element method for wave equation, while restricting the error to low values. We formulate the continuous Galerkin and discontinuous Galerkin formulation of the multiscale method, both of which have pros and cons. Applications of the multiscale method to three heterogeneous models show that our multiscale method can effectively model the elastic wave propagation in anisotropic media with a significant reduction in the degrees of freedom in the modeling system.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Evaluation of formation water chemistry and scale prediction: Bakken Shale

Determination of in situ formation water chemistry is an essential component of reservoir management. This study details the use of thermodynamic computer models to calculate reservoir pH and restore produced water analyses for prediction of scale formation. Bakken produced water samples were restored to formation conditions and calculations of scale formation performed. In situ pH is controlled by feldspar-clay equilibria. Calcite scale is readily formed due to changes in pH during pressure drop from in situ to surface conditions. The formation of anhydrite and halite scale, which has been observed, was predicted only for the most saline samples. Finally, in addition, the formation of anhydrite and/or halite may be related to the localized conditions of increased salinity as water is partitioned into the gas phase during production.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

TRACING OUTFLOWS AND ACCRETION: A BIMODAL AZIMUTHAL DEPENDENCE OF Mg II ABSORPTION

We report a bimodality in the azimuthal angle distribution of gas around galaxies as traced by Mg II absorption: halo gas prefers to exist near the projected galaxy major and minor axes. The bimodality is demonstrated by computing the mean azimuthal angle probability distribution function using 88 spectroscopically confirmed Mg II-absorption-selected galaxies [W{sub r} (2796) {>=} 0.1 A] and 35 spectroscopically confirmed non-absorbing galaxies [W{sub r} (2796) < 0.1 A] imaged with Hubble Space Telescope and Sloan Digital Sky Survey. The azimuthal angle distribution for non-absorbers is flat, indicating no azimuthal preference for gas characterized by W{sub r} (2796) < 0.1 A. We find that blue star-forming galaxies clearly drive the bimodality while red passive galaxies may exhibit an excess along their major axis. These results are consistent with galaxy evolution scenarios where star-forming galaxies accrete new gas, forming new stars and producing winds, while red galaxies exist passively due to reduced gas reservoirs. We further compute an azimuthal angle dependent Mg II absorption covering fraction, which is enhanced by as much as 20%-30% along the major and minor axes. The W{sub r} (2796) distribution for gas along the major axis is likely skewed toward weaker Mg II absorption than for gas along the projected minor axis. These combined results are highly suggestive that the bimodality is driven by gas accreted along the galaxy major axis and outflowing along the galaxy minor axis. Adopting these assumptions, we find that the opening angle of outflows and inflows to be 100 Degree-Sign and 40 Degree-Sign , respectively. We find that the probability of detecting outflows is {approx}60%, implying that winds are more commonly observed.

79 ASTRONOMY AND ASTROPHYSICS↗

Simulation-trained machine learning models for Lorentz transmission electron microscopy

Understanding the collective behavior of complex spin textures, such as lattices of magnetic skyrmions, is of fundamental importance for exploring and controlling the emergent ordering of these spin textures and inducing phase transitions. It is also critical to understand the skyrmion–skyrmion interactions for applications such as magnetic skyrmion-enabled reservoir or neuromorphic computing. Magnetic skyrmion lattices can be studied using in situ Lorentz transmission electron microscopy (LTEM), but quantitative and statistically robust analysis of the skyrmion lattices from LTEM images can be difficult. In this work, we show that a convolutional neural network, trained on simulated data, can be applied to perform segmentation of spin textures and to extract quantitative data, such as spin texture size and location, from experimental LTEM images, which cannot be obtained manually. This includes quantitative information about skyrmion size, position, and shape, which can, in turn, be used to calculate skyrmion–skyrmion interactions and lattice ordering. We apply this approach to segmenting images of Néel skyrmion lattices so that we can accurately identify skyrmion size and deformation in both dense and sparse lattices. The model is trained using a large set of micromagnetic simulations as well as simulated LTEM images. This entirely open-source training pipeline can be applied to a wide variety of magnetic features and materials, enabling large-scale statistical studies of spin textures using LTEM.

McCray, Arthur R. C. (ORCID:0000000160774698)↗

GRAVITATIONAL WAVES FROM MASSIVE MAGNETARS FORMED IN BINARY NEUTRON STAR MERGERS

Binary neutron star (NS) mergers are among the most promising sources of gravitational waves (GWs), as well as candidate progenitors for short gamma-ray bursts (SGRBs). Depending on the total initial mass of the system and the NS equation of state (EOS), the post-merger phase can be characterized by a prompt collapse to a black hole or by the formation of a supramassive NS, or even a stable NS. In the latter cases of post-merger NS (PMNS) formation, magnetic field amplification during the merger will produce a magnetar and induce a mass quadrupole moment in the newly formed NS. If the timescale for orthogonalization of the magnetic symmetry axis with the spin axis is smaller than the spindown time, the NS will radiate its spin down energy primarily via GWs. Here we study this scenario for the various outcomes of NS formation: we generalize the set of equilibrium states for a twisted torus magnetic configuration to include solutions that, for the same external dipolar field, carry a larger magnetic energy reservoir; we hence compute the magnetic ellipticity for such configurations, and the corresponding strength of the expected GW signal as a function of the relative magnitude of the dipolar and toroidal field components. The relative number of GW detections from PMNSs and from binary NSs is a very strong function of the NS EOS, being higher (∼1%) for the stiffest EOSs and negligibly small for the softest ones. For intermediate-stiffness EOSs, such as the n = 4/7 polytrope recently used by Giacomazzo and Perna or the GM1 used by Lasky et al., the relative fraction is ∼0.3%; correspondingly, we estimate a GW detection rate from stable PMNSs of ∼0.1-1 yr{sup –1} with advanced detectors, and of ∼100-1000 yr{sup –1} with detectors of third generation such as the Einstein Telescope. Measurement of such GW signals would provide constraints on the NS EOS and, in connection with an SGRB, on the nature of the binary progenitors giving rise to these events.

79 ASTRONOMY AND ASTROPHYSICS↗

Supplementary Data for "Evaluation of the Economic Implications of Varied Pressure Drawdown Strategies Generated Using a Real-time, Rapid Predictive, Multi-fidelity Model for Unconventional Oil and Gas Wells" by Bello, K., Vikara, D., Sheriff, A., Viswanathan, H., Carr, T., Sweeney, M., O'Malley, D., Marquis, M., Vactor, R.T., and Cunha, L.

The Bello et al. study evaluates the impact of contrasting pressure drawdown on gas productivity and the resulting economics of a well in the Marcellus Shale of the Appalachian Basin. This research applies a techno-economic analysis approach to help identify potential ways pressure management strategies can be used to improve cumulative recovery of hydraulically fractured horizontal wells while maintaining project profitability. Gas production forecast outlook scenarios of the Marcellus Shale Energy and Environment Laboratory Laboratory's MIP-3H well were generated under varying pressure drawdowns using two approaches: 1) a novel physics-informed machine learning (PIML) workflow and 2) via traditional reservoir simulation in Computer Modeling Group’s (CMG) GEM Compositional & Unconventional Simulator. Cash flow and other economic metrics of interest were compiled on the production outlook using the U.S. Department of Energy's (DOE) National Energy Technology Laboratory (NETL) Unconventional Shale Well Economic Model (UShWEM).The sheets within this Microsoft ExcelTM workbook provide the economic metric outputs for the baseline condition and the one-at-a-time (OAT) sensitivity analysis of UShWEM's input parameters for each of the production scenarios evaluated.

Fracture Network Model↗

Alternative CAES Technology Using Depleted Unconventional Gas Wells and Subsurface Thermal Energy Storage (GeoCAES)

This project assessed the technical viability of a process called GeoCAES. The process stores electrical energy by injecting natural gas into shale gas formations using a compressor, storing it, and producing it through an expander to generate electricity. This data submission includes the models of temperature and pressure changes in the wellbore, surface plant equipment (compressor and expander), and the code used in CMG GEM reservoir modeling software to simulate injection and production. Note - the wellbore and surface plant equipment models use the REFPROP Excel Add-in from NIST (linked in submission) to calculate natural gas properties. Note - the reservoir model code requires a license for the Computer Modeling Group (CMG) GEM reservoir modeling software (linked in submission) to run it.

15 GEOTHERMAL ENERGY↗

Sedimentary Geothermal Feasibility Study: October 2016

The objective of this project is to analyze the feasibility of commercial geothermal projects using numerical reservoir simulation, considering a sedimentary reservoir with low permeability that requires productivity enhancement. A commercial thermal reservoir simulator (STARS, from Computer Modeling Group, CMG) is used in this work for numerical modeling. In the first stage of this project (FY14), a hypothetical numerical reservoir model was developed, and validated against an analytical solution. The following model parameters were considered to obtain an acceptable match between the numerical and analytical solutions: grid block size, time step and reservoir areal dimensions; the latter related to boundary effects on the numerical solution. Systematic model runs showed that insufficient grid sizing generates numerical dispersion that causes the numerical model to underestimate the thermal breakthrough time compared to the analytic model. As grid sizing is decreased, the model results converge on a solution. Likewise, insufficient reservoir model area introduces boundary effects in the numerical solution that cause the model results to differ from the analytical solution.

15 GEOTHERMAL ENERGY↗

Accelerating Multiphase Simulations With Denoising Diffusion Model Driven Initializations

This study introduces a hybrid fluid simulation approach that integrates generative diffusion models with physics‐based simulations, aiming at reducing the computational costs of flow simulations while still honoring all the physical properties of interest. Pore‐scale simulations enhance our understanding of applications such as assessing hydrogen and storage efficiency in underground reservoirs. Nevertheless, they are computationally expensive and the presence of non‐unique solutions can require multiple simulations within a single geometry. To overcome the computational cost hurdle, we propose a method that couples generative diffusion models and physics‐based simulations. While training the data‐driven model, we simultaneously generate initial conditions and perform physics‐based simulations using these. This integrated approach enables us to receive real‐time feedback on a single compute node equipped with both CPUs and GPUs. By efficiently managing these processes within a single compute node, we can continuously monitor performance and halt training once the model meets the specified criteria. To test our model, we generate realizations in a real Berea sandstone fracture which shows that our technique is up to 4.4 times faster than commonly used flow simulation initializations.

36 MATERIALS SCIENCE↗

Hydrodynamic and Structural Response Modeling of a Prototype Floating Membrane Reservoir System for Pumped Storage Hydropower

Hydrodynamic, computational fluid dynamics, and finite-element modeling were performed for a novel floating membrane reservoir system design for closed-loop pumped storage hydropower application. The conceptual design, which is now protected under an invention disclosure with a patent pending, offers a potential low-cost, low-impact solution to address the high costs, long investment return periods, and environmental disruptions encountered with traditional pumped storage development while offering modularity to enable replication at many locations. Prior to physical tests on a prototype, this paper documents the response analysis and modeling completed to simulate hydraulic and structural performance under different reservoir deployment and alignment arrangements, evaluate system stability, and refine the conceptual design. The results indicate that the excitation frequency from vortex shedding is at least an order of magnitude lower than the water sloshing frequencies in the reservoir, the structural natural frequency of the entire reservoir, and the vibrational frequencies of the side membranes, although excitation frequencies from other sources could cause mechanical resonance under certain conditions. To control destabilizing effects and prevent rocking motion and vibration, the authors conclude that the design could be improved by including a support structure around the floating reservoir, which could also provide floating-membrane containment, improve safety, and facilitate vertical reservoir movement. These conclusions, based on hydrodynamic and structural response modeling, help define specifications for upcoming full-scale prototype construction, deployment, and testing. This study demonstrates standard modeling technique application to an innovative design for which similar applications have not been previously evaluated. As a result, the refined design is capable of improving the scalability and feasibility of pumped storage hydropower in the United States and will be considered for commercialization following prototype testing.

13 HYDRO ENERGY↗

An extended trajectory-mechanics approach for calculating two-phase flow paths

A technique originating in quantum dynamics is used to derive a trajectory-based, semi-analytical solution for two-phase flow. The partial differential equation governing the evolution of the aqueous phase is equivalent to a family of ordinary differential equations defined along a path through the porous medium. The trajectories may be found by solving the differential equations directly or by post-processing the output of a numerical solution to the full set of governing equations. The trajectories, which differ from conventional streamlines, are found to bend downward in response to gravitational forces. The curvature is more pronounced as the dip of the porous layer containing the flow increases. Subtle changes in the relative permeability curve can lead to significant variations in the trajectories. The ordinary differential equation for the trajectory provides an expression for the travel time along the path. The expression produces a semi-analytical approximation to the model parameter sensitivities, the partial derivatives of the travel times with respect to changes in the permeability model. The semi-analytical trajectory-based sensitivities generally agree with those computed using a numerical reservoir simulator and a perturbation approach. The sensitivities are useful in tomographic imaging algorithms designed to estimate the spatial variation in permeability within a porous medium using multiphase observations.

58 GEOSCIENCES↗

Application of Machine Learning and Artificial Intelligence in Proxy Modeling for Fluid Flow in Porous Media

Reservoir simulation models are the major tools for studying fluid flow behavior in hydrocarbon reservoirs. These models are constructed based on geological models, which are developed by integrating data from geology, geophysics, and petro-physics. As the complexity of a reservoir simulation model increases, so does the computation time. Therefore, to perform any comprehensive study which involves thousands of simulation runs, a very long period of time is required. Several efforts have been made to develop proxy models that can be used as a substitute for complex reservoir simulation models. These proxy models aim at generating the outputs of the numerical fluid flow models in a very short period of time. This research is focused on developing a proxy fluid flow model using artificial intelligence and machine learning techniques. In this work, the proxy model is developed for a real case CO2 sequestration project in which the objective is to evaluate the dynamic reservoir parameters (pressure, saturation, and CO2 mole fraction) under various CO2 injection scenarios. The data-driven model that is developed is able to generate pressure, saturation, and CO2 mole fraction throughout the reservoir with significantly less computational effort and considerably shorter period of time compared to the numerical reservoir simulation model.

Amini, Shohreh↗

Modeling Nanoconfinement Effects Using Active Learning

Predicting the spatial configuration of gas in nanopores of is relevant in applications such as fluid flow forecasting and hydrocarbon reserves estimation. For example, shale reservoirs have suffered from computationally intractable multiscale problems, since fluid properties such as viscosity, density, and adsorption must be calculated by using expensive molecular dynamics (MD) simulations within each nanopore, whereas flow through these connected nanopores must be simulated at the micrometer scale. We utilize machine learning techniques to quickly and accurately model nanoscale confinement effects as an important step toward bridging the nano and micro scales. Our workflow is based on building and training physics-based deep-neural-networks models by learning from a database of MD calculations. The model accounts for the adsorption phenomenon by predicting the statistical distribution of gas inside nanopores. Because large databases of MD calculations are expensive to create, we investigate active learning (AL) as a data set construction strategy. In this workflow, new data are selected based on the model uncertainty via the query-by-committee approach. We show that our workflow obtains accurate models that generalize to real scanning electron microscopy geometries with 1/10th of the number of MD calculations required vs random data set generation. Our method enables the possibility of modeling nanoconfinement effects at the mesoscale, where complex connected sets of nanopores affect flow.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Simulating the pervasive fracture and fragmentation of materials and structures using randomly close-packed Voronoi tessellations.

Under extreme loading conditions most often the extent of material and structural fracture is pervasive in the sense that a multitude of cracks are nucleating, propagating in arbitrary directions, coalescing, and branching. Pervasive fracture is a highly nonlinear process involving complex material constitutive behavior, material softening, localization, surface generation, and ubiquitous contact. Two primary applications in which pervasive fracture is encountered are (1) weapons effects on structures and (2) geomechanics of highly jointed and faulted reservoirs. A pure Lagrangian computational method based on randomly close-packed Voronoi tessellations is proposed as a rational approach for simulating the pervasive fracture of materials and structures. Each Voronoi cell is formulated as a finite element using the reproducing kernel method. Fracture surfaces are allowed to nucleate only at the intercell faces. The randomly seeded Voronoi cells provide an unbiased network for representing cracks. In this initial study two approaches for allowing the new surfaces to initiate are studied: (1) dynamic mesh connectivity and the instantaneous insertion of a cohesive traction when localization is detected, and (2) a discontinuous Galerkin approach in which the interelement tractions are an integral part of the variational formulation, but only become active once localization is detected. Pervasive fracture problems are extremely sensitive to initial conditions and system parameters. Dynamic problems exhibit a form of transient chaos. The primary numerical challenge for this class of problems is the demonstration of model objectivity and, in particular, the identification and demonstration of a measure of convergence for engineering quantities of interest.

97 MATHEMATICS AND COMPUTING↗

Using Machine Learning to Predict Future Temperature Outputs in Geothermal Systems

Optimizing the power output, and economic value, of geothermal power plants over decades of operation is a major challenge in renewable energy. Optimizing the output requires the ability to predict the mass flow rates and the output temperatures of production wells based on the inputs of injection wells, as well as the time history of the system. Machine Learning (ML) that incorporates the known physics of geothermal systems is one possible solution to this challenge. In this work, we explore the ability of ML algorithms to predict future temperature outputs based on historical data. Considering the challenges with obtaining an empirical dataset from field data that is large enough to enable reliable ML, we propose an alternate approach: developing a high-fidelity reservoir model and using computational resources to build a dataset that enables ML. As a first step towards achieving this goal, we present preliminary results from applying ML to predict the temperature timeseries of simple modeled geothermal systems. We describe the application of relevant state-of-the-art ML approaches, such as the Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNN), to extract temporal structures in the model data. We assess the accuracy of the forecasts we obtain, compare the selected approaches, and share the lessons learned that would inform the process of training and utilizing ML algorithms for larger and more complex geothermal systems.

GEOTHERMAL ENERGY↗

iGeoT v1.0

iGeoT v1.0 automates into a stand-alone computer program the multicomponent chemical geothermometry code GeoT v2.1 and the numerical optimization engine of iTOUGH2. iGeoT allows for optimizations of GeoT runs using multiple water chemical analyses. The underlying geothermometry method is that previously developed by Reed and Spycher (1984) using the computed saturation indices of multiple minerals. GeoT automatically reconstitutes deep fluid compositions and estimates reservoir temperature from statistical evaluation of computed mineral saturation indices. The output include estimated temperatures following various statistical methods and their range of uncertainty. The computer program also outputs temperatures predicted using classical geothermometers. Input parameters include water composition and, optionally, gas composition, fraction of gas discharge, dilution/evaporation factor (in case of boiling or mixing with dilute waters), and end-member water composition (in case of mixing with other non-dilute waters). The dilution/evaporation factor, fraction of gas discharge, and concentration of various aqueous and gas species can be automatically estimated by numerical optimization. This program was designed for geothermal applications.

Spycher, Nicolas↗