Search NASA⌕ Search

SEARCH · Search NASA

Results for “Large-scale inverse problems”

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 55 records · Page 3

Multigrid Optimization for Large-Scale Ptychographic Phase Retrieval

Ptychography is a popular imaging technique that combines diffractive imaging with scanning microscopy. The technique consists of a coherent beam that is scanned across an object in a series of overlapping positions, leading to reliable and improved reconstructions. Ptychographic microscopes allow for large fields to be imaged at high resolution at the cost of additional computational expense. Herein, we propose a multigrid-based optimization framework to reduce the computational burdens of large-scale ptychographic phase retrieval. Our proposed method exploits the inherent hierarchical structures in ptychography through tailored restriction and prolongation operators for the object and data domains. Our numerical results show that our proposed scheme accelerates the convergence of its underlying solver and outperforms the ptychographical iterative engine, a workhorse in the optics community.

97 MATHEMATICS AND COMPUTING↗

DART-PFLOTRAN: An ensemble-based data assimilation system for estimating subsurface flow and transport model parameters

Ensemble-based Data Assimilation (EDA), based on the Monte Carlo approach, has been effectively applied to estimate model parameters through inverse modeling in subsurface flow and transport problems. However, implementation of EDA approach involves a complicated workflow that include setting up and executing ensemble forward model simulations, processing observations and model simulation results for parameter updates, and repeat for sequential or iterative EDA. To facilitate the management of such workflow and lower the barriers for adopting EDA-based parameter estimation in subsurface science, we develop a generic software frame-work linking the Data Assimilation Research Testbed (DART) with a massively parallel subsurface FLOw and TRANsport code PFLOTRAN. The new DART-PFLOTRAN leverages both the core data assimilation engines in DART and the computational power afforded by PFLOTRAN. In addition to the standard smoother and filtering options, DART-PFLOTRAN enables an iterative EDA workflow based on the Ensemble Smoother for Multiple Data Assimilation method (ES-MDA) to improve estimation accuracy for nonlinear forward problems. Here, we verify the implementation of ES-MDA in DART-PFLOTRAN using two synthetic cases designed to estimate static permeability and dynamic exchange fluxes across the riverbed, respectively, from continuous temperature measurements made across a depth profile. One-dimensional hydro-thermal simulations are performed in both cases to relate temperature responses with the parameters of interest. In the case of estimating dynamic parameters, we demonstrate the flexibility of DART-PFLOTRAN in automating sequential ES-MDA workflow, which will significantly reduce the time researchers spend on managing complex workflows in similar applications. Both studies yield accurate estimations of the parameters compared to their synthetic truth, while ES-MDA leads to more accurate estimation when a high level of nonlinearity exist between observed responses and unknown parameters. With a code base in Python and Fortran, DART-PFLOTRAN paves the way for applications in large-scale subsurface inverse modeling by automating the complex workflow of sequential ES-MDA that can be executed on various computing platforms.

97 MATHEMATICS AND COMPUTING↗

ML-based Dimension Reduction Strategies

Deep learning (DL)--based surrogate models have achieved success in various applications in carbon capture and storage (CCS). However, the model training on high-dimensional spaces is computationally expensive and impractical for large-scale and complex geological models, because the models usually contain hundreds of thousands to millions of grid cells, each with a set of parameters. Furthermore, the high cost of generating training data with sufficient variation is another limitation of model training on high-dimensional spaces, which may result in overfitting and reduce the model efficiency and prediction performance. We proposed the workflow incorporating dimension reduction methods and deep learning models, which aim to extract the latent variables of input parameters and output state variables, and then build the mapping function at the latent spaces. The proposed workflow can significantly reduce the computational complexity in solving both forward and inverse problems compared to models trained on high-dimensional spaces. Dimensionality reduction models showed great potential in workflows for fast reservoir simulation, history matching, prior model generation, visualization, and more, ultimately enhancing DL model performance in related SMART Work Packages.

Hosseini, Seyyed↗

Towards real-time monitoring: data assimilated time-lapse full waveform inversion for seismic velocity and uncertainty estimation

SUMMARY Rapid development of time-lapse seismic monitoring instrumentations has made it possible to collect dense time-lapse data for tomographically retrieving time-lapse (even continuous) images of subsurface changes. While traditional time-lapse full waveform inversion (TLFWI) algorithms are designed for sparse time-lapse surveys, they lack of effective temporal constraint on time-lapse data, and, more importantly, lack of the uncertainty estimation of the TLFWI results that is critical for further interpretation. Here, we propose a new data assimilation TLFWI method, using hierarchical matrix powered extended Kalman filter (HiEKF) to quantify the image uncertainty. Compared to existing Kalman filter algorithms, HiEKF allows to store and update a data-sparse representation of the cross-covariance matrices and propagate model errors without expensive operations involving covariance matrices. Hence, HiEKF is computationally efficient and applicable to 3-D TLFWI problems. Then, we reformulate TLFWI in the framework of HiEKF (termed hereafter as TLFWI-HiEKF) to predict time-lapse images of subsurface spatiotemporal velocity changes and simultaneously quantify the uncertainty of the inverted velocity changes over time. We demonstrate the validity and applicability of TLFWI–HiEKF with two realistic CO2 monitoring models derived from Frio-II and Cranfield CO2 injection sites, respectively. In both 2-D and 3-D examples, the inverted high-resolution time-lapse velocity results clearly reveal a continuous velocity reduction due to the injection of CO2. Moreover, the accuracy of the model is increasing over time by assimilating more time-lapse data while the standard deviation is decreasing over lapsed time. We expect TLFWI-HiEKF to be equipped with real-time seismic monitoring systems for continuously imaging the distribution of subsurface gas and fluids in the future large-scale CO2 sequestration experiments and reservoir management.

58 GEOSCIENCES↗

LuGo: An enhanced quantum phase estimation implementation

Quantum Phase Estimation (QPE) is a cardinal algorithm in quantum computing that plays a crucial role in various applications, including cryptography, molecular simulation, and solving systems of linear equations. However, the standard implementation of QPE faces challenges related to time complexity and circuit depth, which limit its practicality for large-scale computations. We introduce LuGo, a novel framework designed to enhance the performance of QPE by reducing circuit duplication, as well as using parallelization techniques to achieve faster generation of the QPE circuit and gate reduction. We validate the effectiveness of our framework by generating quantum linear solver circuits, which require both QPE and inverse QPE, to solve linear systems of equations. LuGo achieves significant improvements in both computational efficiency and hardware requirements without compromising on accuracy. Compared to a standard QPE implementation, LuGo reduces time consumption to generate a circuit that solves a 2 6 × 2 6 system matrix by a factor of 50.68 and over 31× reduction of quantum gates and circuit depth, with no fidelity loss on an ideal quantum simulator. Furthermore, we demonstrated the versatility and scalability of LuGo enabled HHL algorithm by simulating a canonical Hele-Shaw fluid problem using a quantum simulator. With these advantages, LuGo paves the way for more efficient implementations of QPE, enabling broader applications across several quantum computing domains.

Quantum algorithm↗

Density matrix renormalization group description of the island of inversion isotopes F 28 – 33

Recent experiments have confirmed that the neutron-rich isotopes 28,29 F belong to the so-called island of inversion (IOI), a region of the nuclear chart around Z = 10 and N = 20 where nuclear structure deviates from the standard shell model predictions due to deformation and continuum effects. However, while the general principles leading to the IOI are relatively well understood, the details of the low-lying structure of the exotic fluorine isotopes 28–33 F are basically unknown. In this study, we perform large-scale shell model calculations including continuum states to investigate the properties of the neutron-rich isotopes 25–33 F, from a core of 24 O and using an effective two-body interaction with a small number of adjustable parameters in the central and tensor channels. We develop two models adjusted on experimentally confirmed states in 25,26 O and 25–27 F based on different assumptions concerning the positions of the neutron 0d 3/2 and 1p 3/2 shells, and solve the many-body problem using the density matrix renormalization group (DMRG) method for open quantum systems in an sd–fp model space. We obtain the first detailed spectroscopy of 25–33F in the continuum and show how the interplay between continuum effects and deformation explains the recent data on 28,29 F. Several deformed one- and two-neutron halo states are predicted in 29,31 F, and we provide some information about the possible structure of the heaviest fluorine isotopes. We also suggest several experimental studies of interest to constraint models and test the present predictions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Inferring colloidal interaction from scattering by machine learning

A machine learning solution for the potential inversion problem in elastic scattering is outlined. The inversion scheme consists of two major components, a generative network featuring a variational autoencoder which extracts the targeted static two-point correlation functions from experimentally measured scattering cross sections, and a Gaussian process framework which probabilistically infers the relevant structural parameters from the inverted correlation functions. Via a case study of charged colloidal suspensions, the feasibility of this approach for quantitative study of molecular interaction is critically benchmarked and its merit over existing deterministic approaches, in terms of numerical accuracy and computationally efficiency, is demonstrated.

36 MATERIALS SCIENCE↗

Optimal Power Management for Large-Scale Battery Energy Storage Systems via Bayesian Inference

Large-scale battery energy storage systems (BESS) have found ever-increasing use across industry and society to accelerate clean energy transition and improve energy supply reliability and resilience. However, their optimal power management poses significant challenges: the underlying high-dimensional nonlinear nonconvex optimization lacks computational tractability in real-world implementation, and the uncertainty of the exogenous power demand makes exact optimization difficult. This paper presents a new solution framework to address these bottlenecks. The solution pivots on introducing power-sharing ratios to specify each cell’s power quota from the output power demand. To find the optimal power-sharing ratios, we formulate a nonlinear model predictive control (NMPC) problem to achieve power-loss-minimizing BESS operation while complying with safety, cell balancing, and power supply-demand constraints. We then propose a parameterized control policy for the power-sharing ratios, which utilizes only three parameters, to reduce the computational demand in solving the NMPC problem. This policy parameterization allows us to translate the NMPC problem into a Bayesian inference problem for the sake of 1) computational tractability, and 2) overcoming the nonconvexity of the optimization problem. We leverage the ensemble Kalman inversion technique to solve the parameter estimation problem. Concurrently, a low-level control loop is developed to seamlessly integrate our proposed approach with the BESS to ensure practical implementation. This low-level controller receives the optimal power-sharing ratios, generates output power references for the cells, and maintains a balance between power supply and demand despite uncertainty in output power. We conduct extensive simulations and experiments on a 20-cell prototype to validate the proposed approach.

Battery energy storage systems (BESSs)↗

Manifold Learning-Based Polynomial Chaos Expansions for High-Dimensional Surrogate Models

In this work we introduce a manifold learning-based method for uncertainty quantification (UQ) in systems describing complex spatiotemporal processes. Our first objective is to identify the embedding of a set of high-dimensional data representing quantities of interest of the computational or analytical model. For this purpose, we employ Grassmannian diffusion maps, a two-step nonlinear dimension reduction technique which allows us to reduce the dimensionality of the data and identify meaningful geometric descriptions in a parsimonious and inexpensive manner. Polynomial chaos expansion is then used to construct a mapping between the stochastic input parameters and the diffusion coordinates of the reduced space. An adaptive clustering technique is proposed to identify an optimal number of clusters of points in the latent space. The similarity of points allows us to construct a number of geometric harmonic emulators which are finally utilized as a set of inexpensive pretrained models to perform an inverse map of realizations of latent features to the ambient space and thus perform accurate out-of-sample predictions. Thus, the proposed method acts as an encoder-decoder system which is able to automatically handle very high-dimensional data while simultaneously operating successfully in the small-data regime. The method is demonstrated on two benchmark problems and on a system of advection-diffusion-reaction equations which model a first-order chemical reaction between two species. In all test cases, the proposed method is able to achieve highly accurate approximations which ultimately lead to the significant acceleration of UQ tasks.

42 ENGINEERING↗

Scalable deep learning for watershed model calibration

Watershed models such as the Soil and Water Assessment Tool (SWAT) consist of high-dimensional physical and empirical parameters. These parameters often need to be estimated/calibrated through inverse modeling to produce reliable predictions on hydrological fluxes and states. Existing parameter estimation methods can be time consuming, inefficient, and computationally expensive for high-dimensional problems. In this paper, we present an accurate and robust method to calibrate the SWAT model (i.e., 20 parameters) using scalable deep learning (DL). We developed inverse models based on convolutional neural networks (CNN) to assimilate observed streamflow data and estimate the SWAT model parameters. Scalable hyperparameter tuning is performed using high-performance computing resources to identify the top 50 optimal neural network architectures. We used ensemble SWAT simulations to train, validate, and test the CNN models. We estimated the parameters of the SWAT model using observed streamflow data and assessed the impact of measurement errors on SWAT model calibration. We tested and validated the proposed scalable DL methodology on the American River Watershed, located in the Pacific Northwest-based Yakima River basin. Our results show that the CNN-based calibration is better than two popular parameter estimation methods (i.e., the generalized likelihood uncertainty estimation [GLUE] and the dynamically dimensioned search [DDS], which is a global optimization algorithm). For the set of parameters that are sensitive to the observations, our proposed method yields narrower ranges than the GLUE method but broader ranges than values produced using the DDS method within the sampling range even under high relative observational errors. The SWAT model calibration performance using the CNNs, GLUE, and DDS methods are compared using R 2 and a set of efficiency metrics, including Nash-Sutcliffe, logarithmic Nash-Sutcliffe, Kling-Gupta, modified Kling-Gupta, and non-parametric Kling-Gupta scores, computed on the observed and simulated watershed responses. The best CNN-based calibrated set has scores of 0.71, 0.75, 0.85, 0.85, 0.86, and 0.91. The best DDS-based calibrated set has scores of 0.62, 0.69, 0.8, 0.77, 0.79, and 0.82. The best GLUE-based calibrated set has scores of 0.56, 0.58, 0.71, 0.7, 0.71, and 0.8. The scores above show that the CNN-based calibration leads to more accurate low and high streamflow predictions than the GLUE and DDS sets. Our research demonstrates that the proposed method has high potential to improve our current practice in calibrating large-scale integrated hydrologic models.

54 ENVIRONMENTAL SCIENCES↗

Scalable algorithms for physics-informed neural and graph networks

Physics-informed machine learning (PIML) has emerged as a promising new approach for simulating complex physical and biological systems that are governed by complex multiscale processes for which some data are also available. In some instances, the objective is to discover part of the hidden physics from the available data, and PIML has been shown to be particularly effective for such problems for which conventional methods may fail. Unlike commercial machine learning where training of deep neural networks requires big data, in PIML big data are not available. Instead, we can train such networks from additional information obtained by employing the physical laws and evaluating them at random points in the space–time domain. Such PIML integrates multimodality and multifidelity data with mathematical models, and implements them using neural networks or graph networks. Here, we review some of the prevailing trends in embedding physics into machine learning, using physics-informed neural networks (PINNs) based primarily on feed-forward neural networks and automatic differentiation. For more complex systems or systems of systems and unstructured data, graph neural networks (GNNs) present some distinct advantages, and here we review how physics-informed learning can be accomplished with GNNs based on graph exterior calculus to construct differential operators; we refer to these architectures as physics-informed graph networks (PIGNs). We present representative examples for both forward and inverse problems and discuss what advances are needed to scale up PINNs, PIGNs and more broadly GNNs for large-scale engineering problems.

42 ENGINEERING↗

Data-Driven Strategies for Accelerated Materials Design

The ongoing revolution of the natural sciences by the advent of machine learning and artificial intelligence sparked significant interest in the material science community in recent years. The intrinsically high dimensionality of the space of realizable materials makes traditional approaches ineffective for large-scale explorations. Modern data science and machine learning tools developed for increasingly complicated problems are an attractive alternative. An imminent climate catastrophe calls for a clean energy transformation by overhauling current technologies within only several years of possible action available. Tackling this crisis requires the development of new materials at an unprecedented pace and scale. For example, organic photovoltaics have the potential to replace existing silicon-based materials to a large extent and open up new fields of application. In recent years, organic light-emitting diodes have emerged as state-of-the-art technology for digital screens and portable devices and are enabling new applications with flexible displays. Reticular frameworks allow the atom-precise synthesis of nanomaterials and promise to revolutionize the field by the potential to realize multifunctional nanoparticles with applications from gas storage, gas separation, and electrochemical energy storage to nanomedicine. In the recent decade, significant advances in all these fields have been facilitated by the comprehensive application of simulation and machine learning for property prediction, property optimization, and chemical space exploration enabled by considerable advances in computing power and algorithmic efficiency. In this Account, we review the most recent contributions of our group in this thriving field of machine learning for material science. We start with a summary of the most important material classes our group has been involved in, focusing on small molecules as organic electronic materials and crystalline materials. Specifically, we highlight the data-driven approaches we employed to speed up discovery and derive material design strategies. Subsequently, our focus lies on the data-driven methodologies our group has developed and employed, elaborating on high-throughput virtual screening, inverse molecular design, Bayesian optimization, and supervised learning. We discuss the general ideas, their working principles, and their use cases with examples of successful implementations in data-driven material discovery and design efforts. Furthermore, we elaborate on potential pitfalls and remaining challenges of these methods. Finally, we provide a brief outlook for the field as we foresee increasing adaptation and implementation of large scale data-driven approaches in material discovery and design campaigns.

36 MATERIALS SCIENCE↗

Sparse Bayesian mass mapping with uncertainties: peak statistics and feature locations

ABSTRACT Weak lensing convergence maps – upon which higher order statistics can be calculated – can be recovered from observations of the shear field by solving the lensing inverse problem. For typical surveys this inverse problem is ill-posed (often seriously) leading to substantial uncertainty on the recovered convergence maps. In this paper we propose novel methods for quantifying the Bayesian uncertainty in the location of recovered features and the uncertainty in the cumulative peak statistic – the peak count as a function of signal-to-noise ratio (SNR). We adopt the sparse hierarchical Bayesian mass-mapping framework developed in previous work, which provides robust reconstructions and principled statistical interpretation of reconstructed convergence maps without the need to assume or impose Gaussianity. We demonstrate our uncertainty quantification techniques on both Bolshoi N-body (cluster scale) and Buzzard V-1.6 (large-scale structure) N-body simulations. For the first time, this methodology allows one to recover approximate Bayesian upper and lower limits on the cumulative peak statistic at well-defined confidence levels.

79 ASTRONOMY AND ASTROPHYSICS↗

Isospin symmetry breaking in the mirror pair 73 Sr- 73 Br

The recent experimental observation of isospin symmetry breaking (ISB) in the ground states of the T=3/2 mirror pair 73 Sr- 73 Br is theoretically studied using large-scale shell-model calculations. The large valence space and the successful PFSDG-U effective interaction used for the nuclear part of the problem capture possible structural changes and provide a robust basis to treat the ISB effects of both electromagnetic and nonelectromagnetic origin. The calculated shifts and mirror-energy differences are consistent with the inversion of the I π =1/2 - ,5/2 - states between 73 Sr- 73 Br and suggest that the role played by the Coulomb interaction is dominant. Finally, an isospin breaking contribution of nuclear origin is estimated to be ≈25keV.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

The Artificial Scientist: in-Transit Machine Learning of Plasma Simulations

Large-scale simulations or scientific experiments produce petabytes of data per run. This poses massive challenges for I/O and storage when scientific analysis workflows are run manually offline. Unsupervised deep learning-based techniques to extract patterns and non-linear relations from these large amounts of data provide a way to build scientific understanding from raw data, reducing the need for manual pre-selection of analysis steps, but require exascale compute and memory to process the full dataset available. In this paper, we demonstrate a heterogeneous streaming workflow in which plasma simulation data is streamed directly to a Machine Learning (ML) application training a model on the simulation data in-transit, completely circumventing the capacity-constrained filesystem bottleneck. This workflow employs openPMD to provide a high level interface to describe scientific data and also uses ADIOS2, to transfer volumes of data that exceed the capabilities of the filesystem. We employ experience replay to avoid catastrophic forgetting in learning from this non-steady state process in a continual manner and adapt it to improve model convergence while learning in-transit. As a proof-of-concept, we approach the ill-posed inverse problem of predicting particle dynamics from radiation in a particle-incell (PIConGPU) simulation of the Kelvin-Helmholtz instability (KHI). We detail hardware-software co-design challenges as we scale PIConGPU to full Frontier, the Top-1 system as of June 2024 Top500 list.

Kelling, Jeffrey [Helmholtz-Zentrum Dresden Rossen↗

Electromagnetic Transient (EMT) Simulation Algorithms for Evaluation of Large-Scale Extreme Fast Charging Systems (T&D Models)

Simulation of high-fidelity models of extreme fast charging (XFC) systems and large-area power grids with many XFCs can be time consuming in traditional simulators. Traditional simulators use a single method of discretization for all the components that results in imposing a large computational burden of inverting a large matrix as well as increased computations related to single method of discretization (that is typically a trapezoidal method). To overcome the problem of simulating large-area power grids with many XFCs, in this paper, advanced numerical simulation algorithms are applied for the first time together to reduce the dimension of matrix inversion. Here, the algorithms include numerical stiffness-based segregation, time constant-based segregation, clustering and aggregation on differential algebraic equations (DAEs), and multi-order integration approaches. These algorithms apply multiple discretization algorithms rather than a single discretization algorithm that further reduces the computational burden. The approaches mentioned here have resulted in speed-up of up to 18x in the simulation of a single distribution system with 15 XFCs and of up to 271x in the simulation of a transmission-distribution system with 300 XFCs in multiple distribution feeders with respect to conventional simulators (like power systems computer aided design [PSCAD]).

42 ENGINEERING↗

Large-Scale Materials Modeling at Quantum Accuracy: Ab Initio Simulations of Quasicrystals and Interacting Extended Defects in Metallic Alloys

Ab initio electronic-structure has remained dichotomous between achievable accuracy and length-scale. Quantum many-body (QMB) methods realize quantum accuracy but fail to scale. Density functional theory (DFT) scales favorably but remains far from quantum accuracy. We present a framework that breaks this dichotomy by use of three interconnected modules: (i) invDFT: a methodological advance in inverse DFT linking QMB methods to DFT; (ii) MLXC: a machine-learned density functional trained with invDFT data, commensurate with quantum accuracy; (iii) DFT-FE-MLXC: an adaptive higher-order spectral finite-element (FE) based DFT implementation that integrates MLXC with efficient solver strategies and HPC innovations in FE-specific dense linear algebra, mixed-precision algorithms, and asynchronous compute-communication. Furthermore, we demonstrate a paradigm shift in DFT that not only provides an accuracy commensurate with QMB methods in ground-state energies, but also attains an unprecedented performance of 659.7 PFLOPS (43.1% peak FP64 performance) on 619,124 electrons using 8,000 GPU nodes of Frontier supercomputer.

density functional theory↗

Quantitative Characterization of Impacts of Coupled Geomechanics and Flow on Safe and Permanent Geological Storage of CO 2 in Fractured Aquifers

This is the final scientific report for the award DE-FE0023305, entitled “Quantitative Characterization of Impacts of Coupled Geomechanics and Flow on Safe and Permanent Geological Storage of CO 2 in Fractured Aquifers.” The work has been divided into six tasks. In Task 2, we characterized rock properties, which are important when developing a quantitative approach for understanding and predicting geomechanical effects on large-scale CO 2 injection and long-term storage in the subsurface. Rock properties of interest for this characterization include porosity, permeability, elastic constants, strength, and heat capacity. We measured rock properties for three different rock types: concrete, sandstone and shale. These properties were obtained from the acoustic measurement, permeability and porosity measurement, Brazilian test, the uniaxial compressive test, and heat capacity measurement. Then, using the Brazilian technique to fracture samples, we measured their permeability under brine injection for varying effective stresses, followed by similar measurements of permeability under sc-CO 2 injection. Permeability to brine and sc-CO 2 decreased as the effective stress increased. The apparent permeability to sc-CO 2 was an order of magnitude less than that for brine, a consequence of CO 2 being the non-wetting fluid. In Task 3, we developed understandings and correlations for CO 2 injection pressure induced fracturing. We initially fractured four eight inch cubes of concrete with brine to establish a standard test procedure. Then, additional concrete samples were fractured using brine. Confining stresses were exerted on the samples and brine was injected at a constant rate into a borehole that was previously drilled into the concrete cube. The borehole pressure was measured and fracture initiation was identified as a peak in the pressure profile. That was followed by fracture propagation, which was identified as a plateau in the pressure profile, and finally the fracture reached the sample surface. The sample surfaces were photographed before and after fracturing, which enabled determination of where the fracture reached the sample surface. Afterwards, aqueous food dye solution was injected into the fractured concrete samples to color the fracture planes. The borehole was filled with food dye solution in advance, a gas pulse was transmitted to the borehole, and the borehole was then shut in to allow the food dye to be driven into existing fractures. Then, the concrete samples were broken down by high-pressure gas injection to reveal the geometry and morphology of the hydraulic fractures. In addition, acoustic measurements were conducted on multiple locations on each face of the concrete samples before and after injection for comparison. This comparison was used to confirm changes in the internal structure of the concrete sample, namely the formation of fractures. There were six concrete samples fractured with brine. Afterwards, concrete samples were fracture using sc-CO 2 . The main change in experimental procedure was the inclusion of temperature control, since in the field, we mostly encounter temperature and pressure conditions above the supercritical point of CO 2 . The concrete samples were pre-heated in an oven to elevate their temperature, and thermal tape was used to heat the CO 2 to a temperature above its supercritical temperature. Two concrete samples were fractured using liquid CO 2 , four concrete samples were fractured using gaseous CO 2 , and twenty two concrete samples were fractured using sc-CO 2 . For the concrete samples fractured using sc-CO 2 , three had pre-existing hydraulic fractures on the surface near the wellbore, and two were water saturated (as opposed to containing only air). Two concrete samples were composite samples, with a high permeability ball in the center, in order to represent CO 2 injection into a high permeability zone surrounded by a low permeability sealing formation. Later, we fractured five shale samples, obtained from the Niobrara shale outcrop, with sc-CO 2 . The injection rate was constant, confining stresses were applied, the shale samples were pre-heated in an oven to the desired temperature, and the sample was then fractured. An analysis of the failure type was done for many of the concrete samples, based on the predicted break down pressure for tensile and shear failure, and it was found that most samples fractured due to shear failure. Also, the presence of induced fractures has little to no effect on the fracture initiation pressure, but significantly changes the fracture morphology. The CO 2 fracturing in water saturated samples behaved much similar to brine induced fracturing. There was a significant difference of breakdown pressure for injection of CO 2 and brine, where the breakdown pressure for CO 2 induced fracturing is generally around the minimum horizontal stress and that for brine is much higher. Finally, the fracture orientation for most samples with large confining stress differential was perpendicular to the minimum horizontal stress. When the stress differential, especially the difference between the two directions with the smallest stresses was small, the fracture orientation was determined not only by the minimum stress direction, but also the magnitude of the confining stress, the tensile strength of the rock, and the breakdown pressure. In Task 4, we modified our coupled flow-geomechanical models to model fracture growth and propagation in storage formations and caprocks. These flow-geomechanical models are TOUGH2-CSM and TOUGH2-FLAC. The TOUGH2-CSM fluid and heat flow formulation is based on the TOUGH2 formulation of mass and energy conservation equations that govern fluid and heat flow in general multiphase, multicomponent, multi-porosity systems. The TOUGH2-CSM simplified geomechanical formulation is based on the linear theory of elasticity applied to multi-porosity non-isothermal (thermo-multi-poroelastic) media. We previously derived, from the fundamental relations of the linear theory of elasticity, an equation relating mean stress, pore pressures, temperatures, and body force (the Mean Stress equation) that we added to the TOUGH2-CSM fluid and heat flow equations. We extended the TOUGH2-CSM simplified geomechanical formulation to calculate the entire stress tensor by deriving equations for stress tensor components from derivatives of the Cartesian thermo-multi-poroelastic Navier equation components. In addition, we derived equations for stress tensor components in rz-coordinates. We solved these geomechanical equations using the integral finite difference method. This method utilizes momentum fluxes obtained by “factoring” a divergence operator out of each geomechanical equation. In Task 5, we did a literature survey to determine suitable correlations for fracture initiation, growth, and propagation. After surveying numerous papers, we modified TOUGH2-CSM to model stress dependent fracture initiation and growth using the Mohr-Coulomb criterion for shear failure in faults and caprock and the condition of negative minimum effective stress for tensile failure. Tensile failure results in a fracture whose width depends on the difference between fracture pressure and minimum stress and whose propagation is based on the difference between the stress intensity factor at the fracture tip and the rock toughness. The TOUGH-FLAC model is based on the FLAC3D ubiquitous joint model, extended for the modeling of permeability changes induced by changes in effective normal stress across fractures as well permeability changes caused by shearing along fractures. In Task 6, we validated our coupled flow-geomechanical models using analytical solutions and problems from the literature. We validated the TOUGH2-CSM modifications using analytical solutions to the displacement from a uniform load on a semi-infinite elastic medium and the two-dimensional Mandel-Cryer effect. Those were followed by a sample problem to test our stress tensor calculations, which was for injection into a single-phase reservoir with constant properties. Comparisons of our simulator to published results were done on for the depletion of a single-phase reservoir with stress dependent porosity, two examples from the In Salah gas project, an axisymmetric baserock–reservoir–caprock system with a normal faulting stress regime that had thermally induced shear stresses, simulations of CO 2 leakage through caprock in a two-dimensional reservoir in Cartesian coordinates, and predictions of caprock failure from CO 2 injection into an axisymmetric reservoir. Finally, we simulated CO 2 injection pressure-induced fracturing from one of our laboratory studies, Sample 39, and obtained a fracture profile. We also validated the TOUGH2-FLAC model for fluid driven fracture growth against solutions based on the KGD model and a case with an inclined fracture that was loaded from the boundaries to achieve wing-crack propagation. The strain softening tensile behavior and softening of modulus considering a simple damage approach was verified by a simulation representing hydraulic fracturing stress measurement around a vertical well. Finally, the model was validated against deep fracture zone opening and surface uplift at In Salah with inverses analysis used to improve the match of simulation with field data. In Task 7, we developed a scheme based on inverse modeling that can be used to identify caprock leakage. Injection of fluid into a reservoir results in a time varying pressure profile that depends on the properties of the rock and the fluid. The presence of significant caprock leakage would effect this pressure profile and we used inverse modeling, namely the Levenberg-Marquardt method, to determine the leakage location from the pressure profile. A two-dimensional and a three-dimensional example based on published data were presented. The published simulation was run and its results were considered as “field” data. In the two-dimensional example, a simulation of CO 2 injection into a two-dimensional aquifer-caprock system, the caprock had a vertical fault through which leakage could occur. The location of this leakage was obtained from our inverse modeling scheme. In the three-dimensional example, the simulation domain had four geological layers, one of which was the injection zone that contained a horizontal injection well. We simulated fractures as high permeability gridblocks, introduced a fracture that spanned the caprock depth, and ran a simulation to obtain “field” data. Then, using our inverse modeling scheme, we obtained the location of that fracture.

58 GEOSCIENCES↗