Search NASA⌕ Search

SEARCH · Search NASA

Results for “Domain knowledge”

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 451 records · Page 25

Reducing acquisition times in multidimensional NMR with a time-optimized Fourier encoding algorithm

Speeding up the acquisition of multidimensional nuclear magnetic resonance (NMR) spectra is an important topic in contemporary NMR, with central roles in high-throughput investigations and analyses of marginally stable samples. A variety of fast NMR techniques have been developed, including methods based on non-uniform sampling and Hadamard encoding, that overcome the long sampling times inherent to schemes based on fast-Fourier-transform (FFT) methods. Here, we explore the potential of an alternative fast acquisition method that leverages a priori knowledge, to tailor polychromatic pulses and customized time delays for an efficient Fourier encoding of the indirect domain of an NMR experiment. By porting the encoding of the indirect-domain to the excitation process, this strategy avoids potential artifacts associated with non-uniform sampling schemes and uses a minimum number of scans equal to the number of resonances present in the indirect dimension. An added convenience is afforded by the fact that a usual 2D FFT can be used to process the generated data. Acquisitions of 2D heteronuclear correlation NMR spectra on quinine and on the anti-inflammatory drug isobutyl propionic phenolic acid illustrate the new method's performance. This method can be readily automated to deal with complex samples such as those occurring in metabolomics, in in-cell as well as in in vivo NMR applications, where speed and temporal stability are often primary concerns.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Bimodal Visualization of Industrial X-Ray and Neutron Computed Tomography Data

Advanced manufacturing creates increasingly complex objects with material compositions that are often difficult to characterize by a single modality. Our collaborating domain scientists are going beyond traditional methods by employing both X-ray and neutron computed tomography to obtain complementary representations expected to better resolve material boundaries. However, the use of two modalities creates its own challenges for visualization, requiring either complex adjustments of bimodal transfer functions or the need for multiple views. Together with experts in nondestructive evaluation, we designed a novel interactive bimodal visualization approach to create a combined view of the co-registered X-ray and neutron acquisitions of industrial objects. Using an automatic topological segmentation of the bivariate histogram of X-ray and neutron values as a starting point, the system provides a simple yet effective interface to easily create, explore, and adjust a bimodal visualization. Here, we propose a widget with simple brushing interactions that enables the user to quickly correct the segmented histogram results. Our semiautomated system enables domain experts to intuitively explore large bimodal datasets without the need for either advanced segmentation algorithms or knowledge of visualization techniques. We demonstrate our approach using synthetic examples, industrial phantom objects created to stress bimodal scanning techniques, and real-world objects, and we discuss expert feedback.

image segmentation↗

Probabilistic Approach to Enable Extreme-Scale Simulations under Uncertainty and System Faults. Final Technical Report

The current project develops a novel approach that uses a probabilistic description to capture the current state of knowledge about the computational solution. To effectively spread the computational effort over multiple nodes, the global computational domain is split into many subdomains. Computational uncertainty in the solution translates into uncertain boundary conditions for the equation system to be solved on those subdomains, and many independent, concurrent subdomain simulations are used to account for this bound- ary condition uncertainty. By relying on the fact that solutions on neighboring subdomains must agree with each other, a more accurate estimate for the global solution can be achieved. Statistical approaches in this update process make it possible to account for the effect of system faults in the probabilistic description of the computational solution, and the associated uncertainty is reduced through successive iterations. By combining all of these elements, the probabilistic reformulation allows splitting the computational work over very many independent tasks for good scalability, while being robust to system faults.

97 MATHEMATICS AND COMPUTING↗

Multi-agent AI collaboration for digital twin development and assessment

Developing a digital twin (DT) model involves different steps that encompass formulating requirements, model development, implementation, and assessment with respect to real applications. Human expertise is required to coordinate and implement different steps in the DT development and assessment process. However, certain parts of this process can be automated using artificial intelligence (AI) agents for efficient workflow development. In this work, we test and analyze a multiagent AI collaboration with humans in the loop to automate different elements of the DT development and assessment process. To implement the workflow for multiagent AI DT development and assessment, we use Autogen, a multiagent framework developed by Microsoft. Autogen offers a modular and flexible framework for configuring and designing task-specific multiagent workflows. In this framework, large language models (LLMs) form the core intelligence of the AI agents where the quality and performance of the automated element is governed by the inherent capabilities and knowledge base of the LLM. We use retrieval augmented generation to supplement the LLM with relevant domain-specific information for DT requirement formulation. We illustrate this multiagent workflow using a case study on a thermal energy storage system, focusing on how AI agents can collaborate with humans to expedite and optimize different elements of DT development and assessment process.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

Parallel processing of filtered queries in attributed semantic graphs

Execution of complex analytic queries on massive semantic graphs is a challenging problem in big-data analytics that requires high-performance parallel computing. In a semantic graph, vertices and edges carry attributes of various types and the analytic queries typically depend on the values of these attributes. Thus, the computation must view the graph through a filter that passes only those individual vertices and edges of interest. Previous investigations have developed Knowledge Discovery Toolbox (KDT), a sophisticated Python library for parallel graph computations. In KDT, the user can write custom graph algorithms by specifying operations between edges and vertices ( semiring operations ). The user can also customize existing graph algorithms by writing filters. While the high-level language for this customization enables domain scientists to productively express their graph analytics requirements, the customized queries perform poorly due to the overhead of having to call into the Python virtual machine for each vertex and edge. In this work, we use the Selective Embedded Just-In-Time Specialization (SEJITS) approach to automatically translate semiring operations and filters defined by programmers into a lower-level efficiency language, bypassing the upcall into Python. We evaluate our approach by comparing it with the high-performance Combinatorial BLAS engine and show that our approach combines the benefits of programming in a high-level language with executing in a low-level parallel environment. We increase the system's flexibility by developing techniques that provide users with the ability to define new vertex and edge types from Python. We also present a new Roofline model for graph traversals and show that we achieve performance that is significantly closer to the bounds suggested by the Roofline. Finally, to further understand the complex interaction with the underlying architecture, we present an analysis using performance counters that quantifies the improvement in hardware behavior in the context our SEJITS methodology. Comprehensively, we demonstrate the first known solution to the problem of obtaining high performance from a productivity language when applying graph algorithms selectively on semantic graphs with hundreds of millions of edges and scaling to thousands of processors for graphs.

97 MATHEMATICS AND COMPUTING↗

Role of a Transcriptional Regulator in Programmed Cell Death and Plant Development

The long-term goal of this research is to understand the role(s) and molecular mechanisms of programmed cell death (PCD) in the controlling plant growth, development and responses to biotic and abiotic stress. We developed a genetic selection scheme to identify A. thaliana FB1-resistant (fbr) mutants as a way to find genes involved in PCD (Stone et al., 2000; Stone et al., 2005; Khan and Stone, 2008). The disrupted gene in fbr6 (AtSPL14) responsible for the FB1-insensitivity and plant architecture phenotypes encodes a plant-specific SBP DNA-binding domain transcriptional regulator (Stone et al., 2005; Liang et al., 2008). This research plan is designed to fill gaps in the knowledge about the role of SPL14 in plant growth and development. The work is being guided by three objectives aimed at determining the pathways in which SPL14 functions to modulate PCD and/or plant development: (1) determine how SPL14 functions in plant development, (2) identify target genes that are directly regulated by SPL14, and (3) identify SPL14 modifications and interacting proteins. We made significant progress during the funding period. Briefly, some major accomplishments are highlighted below: (1) To identify potential AtSPL14 target genes, we identified a consensus DNA binding site for the AtSPL14 SBP DNA-binding domain using systematic evolution of ligands by exponential selection (SELEX) and site-directed mutagenesis (Liang et al., 2008). This consensus binding site was used to analyze Affymetrix microarray gene expression data obtained from wild-type and fbr6 mutant plants to find possible AtSPL14-regulated genes. These candidate AtSPL14-regulated genes are providing new information on the molecular mechanisms linking plant PCD and plant development through modulation of the 26S proteasome. (2) Transgenic plants expressing epitope-tagged versions of AtSPL14 are being used to confirm the AtSPL14 targets (by ChIP-PCR) and further dissect the molecular interactions (Nazarenus, Liang and Stone, in preparation) (3) Double mutants generated between fbr6 and various accelerated cell death (acd) mutants indicate that sphingolipid metabolism is influenced by AtSPL14 and sphingolipidomics profiling supports this conclusion (Lin, Markham and Stone, in preparation). (4) A new set of phenotypes have been uncovered in the original fbr6-1 mutant, including a short-root phenotype related to auxin signaling and altered photosynthetic parameters related to stomatal density and conductance (Lin and Stone, in preparation; Lin, Madhavan and Stone, in preparation). Additional AtSPL14-related mutants and transgenic plants have been generated to effectively dissect the functions of AtSPL14, including a dominant negative fbr6-2 allele and transgenic plants overexpressing FBR6/AtSPL14 that display an accelerated cell death (acd) phenotype.

59 BASIC BIOLOGICAL SCIENCES↗

Federation of UML models for cyber physical use cases

This method employs the concept of federation, which is defined as the use of existing models that represent aspects of a system in specific domains (such as physical and cyber security domains) and building interfaces to link all of domain models. Federation seeks to build on existing bodies of work. Some examples include the Common Information Models (CIM) maintained by the International Electrotechnical Commission Technical Committee 57 (IEC TC 57) for the electric power industry. Another relevant model is the CIM maintained by the Distributed Management Task Force (DMTF)? this CIM defines a representation of the managed elements in an Information Technology (IT) environment. The power system is an example of a cyber-physical system, where the cyber systems, consisting of computing infrastructure such as networks and devices, play a critical role in the operation of the underlying physical electricity delivery system. Measurements from remote field devices are relayed to control centers through computer networks, and the data is processed to determine suitable control actions. Control decisions are then relayed back to field devices. It has been observed that threat actors may be able to successfully compromise this cyber layer in order to impact power system operation. Therefore, future control center applications must be wary of potentially compromised measurements coming from field devices. In order to ensure the integrity of the field measurements, these applications could make use of compromise indicators from alternate sources of information such as cyber security. Thus, modern control applications may require access to data from sources that are not defined in the local information model. In such cases, software application interfaces will require integration of data objects from cross-domain data models. When incorporating or federating different domains, it is important to have subject matter experts work together, recognizing that not everyone has the same knowledge, responsibilities, focus, or skill set.

Source record↗

A thermoelectric materials database auto-generated from the scientific literature using ChemDataExtractor

An auto-generated thermoelectric-materials database is presented, containing 22,805 data records, automatically generated from the scientific literature, spanning 10,641 unique extracted chemical names. Each record contains a chemical entity and one of the seminal thermoelectric properties: thermoelectric figure of merit, ZT; thermal conductivity, κ; Seebeck coefficient, S; electrical conductivity, σ; power factor, PF; each linked to their corresponding recorded temperature, T. The database was auto-generated using the automatic sentence-parsing capabilities of the chemistry-aware, natural language processing toolkit, ChemDataExtractor 2.0, adapted for application in the thermoelectric-materials domain, following a rule-based sentence-simplification step. Data were mined from the text of 60,843 scientific papers that were sourced from three scientific publishers: Elsevier, the Royal Society of Chemistry, and Springer. To the best of our knowledge, this is the first automatically-generated database of thermoelectric materials and their properties from existing literature. The database was evaluated to have a precision of 82.25% and has been made publicly available to facilitate the application of data science in the thermoelectric-materials domain, for analysis, design, and prediction.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Ontologizing health systems data at scale: making translational discovery a reality

Common data models solve many challenges of standardizing electronic health record (EHR) data but are unable to semantically integrate all of the resources needed for deep phenotyping. Open Biological and Biomedical Ontology (OBO) Foundry ontologies provide computable representations of biological knowledge and enable the integration of heterogeneous data. However, mapping EHR data to OBO ontologies requires significant manual curation and domain expertise. We introduce OMOP2OBO, an algorithm for mapping Observational Medical Outcomes Partnership (OMOP) vocabularies to OBO ontologies. Using OMOP2OBO, we produced mappings for 92,367 conditions, 8611 drug ingredients, and 10,673 measurement results, which covered 68–99% of concepts used in clinical practice when examined across 24 hospitals. When used to phenotype rare disease patients, the mappings helped systematically identify undiagnosed patients who might benefit from genetic testing. By aligning OMOP vocabularies to OBO ontologies our algorithm presents new opportunities to advance EHR-based deep phenotyping.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Study of the two pion final state photoproduction on deuterium

Understanding the structure of baryons in terms of the fundamental interaction of the constituent quarks and gluons is one of the challenges in strong interaction physics. This interaction is governed by Quantum Chromodynamics (QCD). However, solutions of this theory in the non-perturbative domain of the interaction are extremely difficult to achieve. In inelastic electron scattering, very little is known about exclusive hadron production purely contributed to a lack of knowledge. The {gamma}N interaction is recognized for being a powerful method for investigating hadrons and the mysteries that still exist within the strong interaction. From reactions with the nucleon, the strong interaction can be tested through the amplitudes of the N and {Delta} resonances. More specifically, if an electromagnetic interaction is well known then the intermediate resonance states may be evaluated through pion photoproduction. To gain more detailed insight into this interaction, we look to probe the baryon structure of {Delta} and the meson structure of the pion through photon scattering off a deuteron producing two pions in the final state. The photoproduction processes on the deuteron will be used to investigate known baryon resonances in the proton-pion channel. The two pion final state will be investigated for unraveling new information in to the rho decay at threshold. We want to explore both final states interactions to search for ''missing'' states that are predicted by quark models but have not yet been found experimentally. Using the CEBAF Large Acceptance Spectrometer (CLAS), the hadronic products are detected in coincidence with the scattered photon. This makes it possible to measure the differential cross section and the decay angular distribution for the production of two and three pion final states. The measured cross sections will contribute significantly and push the knowledge of the strong interaction to the next level. We propose to use the CEBAF Large Acceptance Spectrometer (CLAS) to study the two pion channel from the EG3 data set, for {Delta}{sup ++}(1232) production. We look to investigate the exclusive reaction of {gamma}d{yields}p{pi}{sup +{pi}-}(n), extracting the relevant cross sections to comparable data sets. This reaction is produced using a high intensity photon beam incident on a deuterium target. These measurements provide unique and coherent results from tagged photons over a broad range of energy, and represent the only pion production data above 5 GeV at this present time.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Artificial Judgement Assistance from teXt (AJAX): Applying Open Domain Question Answering to Nuclear Non-proliferation Analysis

Nuclear non-proliferation analysis is complex and subjective, as the data is sparse, and examples are rare and diverse. While analysing non-proliferation data, it is often desired that the findings be completely auditable such that any claim or assertion can be sourced directly to the reference material from which it was derived. Currently this is accomplished by analysts thoroughly documenting underlying assumptions and clearly referencing details to source documents. This is a labour-intensive and time-consuming process that can be difficult to scale with geometrically increasing quantities of data. In this work, we describe an approach to leverage bi-directional language models for nuclear non-proliferation analysis. It has been shown recently that these models not only capture language syntax but also some of the relational knowledge present in the training data. We have devised a unique Salt and Pepper strategy for testing the knowledge present in the language models, while also introducing auditability function in our pipeline. We demonstrate that fine-tuning the bi-directional language models on domain specific corpus improves their ability to answer domain-specific factoid questions. Our hope is that the results presented in this paper will further the natural language processing (NLP) field by introducing the ability to audit the answers provided by the language models to bring forward the source of said knowledge.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

KG-Hub—building and exchanging biological knowledge graphs

Knowledge graphs (KGs) are a powerful approach for integrating heterogeneous data and making inferences in biology and many other domains, but a coherent solution for constructing, exchanging, and facilitating the downstream use of KGs is lacking. Here we present KG-Hub, a platform that enables standardized construction, exchange, and reuse of KGs. Features include a simple, modular extract–transform–load pattern for producing graphs compliant with Biolink Model (a high-level data model for standardizing biological data), easy integration of any OBO (Open Biological and Biomedical Ontologies) ontology, cached downloads of upstream data sources, versioned and automatically updated builds with stable URLs, web-browsable storage of KG artifacts on cloud infrastructure, and easy reuse of transformed subgraphs across projects. Current KG-Hub projects span use cases including COVID-19 research, drug repurposing, microbial–environmental interactions, and rare disease research. KG-Hub is equipped with tooling to easily analyze and manipulate KGs. KG-Hub is also tightly integrated with graph machine learning (ML) tools which allow automated graph ML, including node embeddings and training of models for link prediction and node classification.

59 BASIC BIOLOGICAL SCIENCES↗

Metal artifact reduction in x-ray computed tomography (CT) by constrained optimization

Purpose: The streak artifacts caused by metal implants have long been recognized as a problem that limits various applications of CT imaging. In this work, the authors propose an iterative metal artifact reduction algorithm based on constrained optimization. Methods: After the shape and location of metal objects in the image domain is determined automatically by the binary metal identification algorithm and the segmentation of ''metal shadows'' in projection domain is done, constrained optimization is used for image reconstruction. It minimizes a predefined function that reflects a priori knowledge of the image, subject to the constraint that the estimated projection data are within a specified tolerance of the available metal-shadow-excluded projection data, with image non-negativity enforced. The minimization problem is solved through the alternation of projection-onto-convex-sets and the steepest gradient descent of the objective function. The constrained optimization algorithm is evaluated with a penalized smoothness objective. Results: The study shows that the proposed method is capable of significantly reducing metal artifacts, suppressing noise, and improving soft-tissue visibility. It outperforms the FBP-type methods and ART and EM methods and yields artifacts-free images. Conclusions: Constrained optimization is an effective way to deal with CT reconstruction with embedded metal objects. Although the method is presented in the context of metal artifacts, it is applicable to general ''missing data'' image reconstruction problems.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Density characterization of discharged gas-filled capillaries through common-path two-color spectral-domain interferometry

Electrically discharged plasma structures, typically several centimeters in length and sub-millimeter in diameter, have been applied to guide laser pulses in laser plasma accelerators and to focus ion and relativistic electron beams in compact, radially symmetric transport configurations. Knowledge of the on-axis plasma density is critical. Traditional density interferometry has been ineffective for these laser-machined structures, while group velocity delay (GVD) techniques involve combining two laser paths with corresponding alignment complexities and stability sensitivities. Here the GVD technique is advanced to a common-path two-color interferometer configuration performed in the spectral domain of a broad-bandwidth femtosecond laser. Multi-shot tracking of the phase is not required, and the common path assures improved stability. Here, this in situ technique was validated on 15 mm long plasma structures, measuring electron densities of 10 17 -10 18 cm -3 for various fill pressures.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Text Mining for Process–Structure–Properties Relationships in Metals

With the advent of large language models (LLMs), the vast unstructured text within millions of academic papers is increasingly accessible for materials discovery—although significant challenges remain. While LLMs offer promising few- and zero-shot learning capabilities, particularly valuable in the materials domain where expert annotations are scarce, general-purpose LLMs often fail to address key materials-specific queries without further adaptation. To bridge this gap, fine-tuning LLMs on human-labeled data is essential for effective structured knowledge extraction (Liu in The Importance of Human-Labeled Data in the Era of LLMs, 2023). Here, in this study, we introduce a novel annotation schema designed to extract generic process–structure–properties relationships from scientific literature. We demonstrate the utility of this approach using a dataset of 128 abstracts, with annotations drawn from two distinct domains: high-temperature materials (Domain I) and uncertainty quantification in simulating materials microstructure (Domain II). Initially, we developed a conditional random field (CRF) model based on MatBERT—a domain-specific BERT variant—and evaluated its performance on Domain I. Subsequently, we compared this model with a fine-tuned LLM (GPT-4o from OpenAI) under identical conditions. Our results indicate that fine-tuning LLMs can significantly improve entity extraction performance over the BERT-CRF baseline on Domain I. However, when additional examples from Domain II were incorporated, the performance of the BERT-CRF model became comparable to that of the GPT-4o model. These findings underscore the potential of our schema for structured knowledge extraction and highlight the complementary strengths of both modeling approaches.

Materials science↗

Implementations of a Flexible Framework for Managing Geologic Sequestration Modeling Projects

Numerical simulation is a standard practice used to support designing, operating, and monitoring CO 2 injection projects. Although a variety of computational tools have been developed that support the numerical simulation process, many are single-purpose or platform specific and have a prescribed workflow that may or may not be suitable for a particular project. We are developing an open-source, flexible framework named Velo that provides a knowledge management infrastructure and tools to support modeling and simulation for various types of projects in a number of scientific domains. The Geologic Sequestration Software Suite (GS3) is a version of this framework with features and tools specifically tailored for geologic sequestration studies. Because of its general nature, GS3 is being employed in a variety of ways on projects with differing goals. GS3 is being used to support the Sim-SEQ international model comparison study, by providing a collaborative framework for the modeling teams and providing tools for model comparison. Another customized deployment of GS3 has been made to support the permit application process. In this case, GS3 is being used to manage data in support of conceptual model development and provide documentation and provenance for numerical simulations. An additional customized deployment of GS3 is being created for use by the United States Environmental Protection Agency (US-EPA) to aid in the CO2 injection permit application review process in one of its regions. These use cases demonstrate GS3’s flexibility, utility, and broad applicability

58 GEOSCIENCES↗

Multi-head physics-informed neural networks for learning functional priors and uncertainty quantification

In numerous applications, the integration of prior knowledge and historical information is essential, particularly for tasks requiring the solution of ordinary or partial differential equations (ODEs/PDEs) in data-sparse or noisy environments. For instance, achieving accurate solutions to time-dependent PDEs with limited initial condition measurements necessitates an effective strategy for embedding prior knowledge. Hard-parameter sharing architectures in neural networks (NNs) have demonstrated success in both traditional and scientific machine learning domains, facilitating the learning of informative representations. Here, in this study, we introduce a novel, yet efficient, method to enhance physics-informed neural networks (PINNs) by incorporating a multi-head structure that enables the learning of functional priors from both empirical data and governing physical laws. This prior information can then be used to address data sparsity and high-level noise in solving ODE/PDE problems with uncertainty quantification (UQ). The approach, termed Multi-Head PINN (MH-PINN), consists of a shared body NN and multiple head NNs, each corresponding to an individual PINN instance. Our framework for functional prior learning is carried out in two stages: (1) training the MH-PINNs to develop a shared body NN alongside multiple head NNs, and (2) employing these trained head NNs to estimate a prior distribution through a normalizing flow-based density estimator. The learned functional prior can then be applied as a regularization mechanism in deterministic contexts or as an informative prior within a Bayesian inference framework, aiding in the resolution of subsequent ODE/PDE tasks. We evaluate the efficacy of MH-PINNs across five benchmark problems, including a high-dimensional parametric PDE, all characterized by data sparsity or substantial noise levels. Our findings reveal that MH-PINNs deliver accurate solutions and robust UQ, demonstrating adaptability across a range of complex and challenging scenarios.

Bayesian inference↗

Predicting CO 2 Plume Migration in Heterogeneous Formations Using Conditional Deep Convolutional Generative Adversarial Network

Numerical simulation of flow and transport in heterogeneous formations has long been studied, especially for uncertainty quantification and risk assessment. The high computational cost associated with running large-scale numerical simulations in a Monte Carlo sense has motivated the development of surrogate models, which aim to capture the important input-output relations of physics-based models but require only a fraction of the cost of full model runs. In this work, we formulate a conditional deep convolutional generative adversarial network (cDC-GAN) surrogate model to learn the dynamic functional mappings in multiphase models. The cDC-GAN belongs to a class of semisupervised learning methods that can be used to learn the data generation processes. Like the original GAN, a main strength of the cDC-GAN is that it includes a self-training scheme for improving the quality of generative modeling in a game theoretic framework, without requiring extensive statistical knowledge and assumptions on input data distributions. In particular, our cDC-GAN model is designed to learn cross-domain mappings between high-dimensional input (e.g., permeability) and output (e.g., phase saturations) pairs, with the ability to incorporate conditioning information (e.g., prediction time). As a use case, we demonstrate the performance of cDC-GAN for predicting the migration of carbon dioxide (CO 2 ) plume in heterogeneous carbon storage reservoirs, which has both numerical and practical significance because of the safe storage requirements now mandated in many countries. Results show that cDC-GAN achieves high accuracy in predicting the spatial and temporal evolution patterns of the injected CO 2 plume, as compared to the original results obtained using a compositional reservoir simulator. The performance of cDC-GAN models, trained using the same number of training samples, stays relatively robust when the level of spatial heterogeneity is increased. Our cDC-GAN is pattern based and is not limited by the underlying physics. Thus, it provides a general framework for developing surrogate models, and for conducting uncertainty analyses for a wide range of physics-based models used in both groundwater and subsurface energy exploration applications.

54 ENVIRONMENTAL SCIENCES↗