Search NASA⌕ Search

SEARCH · Search NASA

Results for “generative modeling”

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 307 records · Page 17
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Using ARM Observations to Evaluate Process-Interactions in MCS Simulations Across Scales (Final Progress Report)

This project, funded by DOE Atmospheric System Research (DE-SC0020050), focused on improving the representation of mesoscale convective systems (MCSs) in numerical weather and climate models by leveraging high-resolution observations from the DOE Atmospheric Radiation Measurement (ARM) program. The research aimed to evaluate model sensitivities to grid spacing, microphysics, and planetary boundary layer (PBL) schemes, with a particular emphasis on improving convection parameterization for high-resolution modeling. Findings from this work highlight several key advancements. Model validation against ARM radar wind profiler data from the Southern Great Plains (SGP) and Manaus (MAO) sites revealed systematic biases in simulated convective mass flux profiles, leading to the development of an observationally constrained evaluation framework for diagnosing and improving model performance. Sensitivity analyses demonstrated that the representation of Amazonian MCSs was highly dependent on PBL scheme selection, while mid-latitude MCSs were more strongly influenced by microphysics parameterizations. A series of high-resolution WRF simulations, ranging from 4 km to 125 m grid spacing, provided insight into the behavior of convective drafts across scales. While updraft properties converged at sub-kilometer resolutions, biases in downdraft intensity persisted even at the finest resolution tested, emphasizing the need for further refinements in model physics. Additionally, comparisons of MCS vertical structures between mid-latitude and tropical environments revealed stronger updrafts and larger mass flux in mid-latitude MCSs, providing critical insights for improving climate model representations of storm-scale dynamics. The project’s findings have already contributed to advancing numerical modeling capabilities, particularly in WRF, MPAS, ICON, and DOE’s SCREAM model, by refining how convective processes are represented in high-resolution climate simulations. Results were disseminated through peer-reviewed publications, conference presentations, and ARM/ASR Research Highlights, engaging the broader scientific community. The project also provided valuable training opportunities for two postdoctoral researchers, who played central roles in model development, analysis, and dissemination of results. Their work contributed to several publications and conference presentations, helping prepare them for careers in atmospheric modeling. By improving the simulation of MCSs, this research directly supports the development of next-generation climate models capable of more accurately representing extreme precipitation and convective processes. The insights gained will inform future improvements in convective parameterization and guide the design of high-resolution weather and climate simulations, ultimately enhancing the reliability of climate projections and weather forecasts.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

HARMONY: Large-Scale Architecture Search for Efficient Hybrid Language Models

As large language models scale to trillions of parameters, their computational and memory requirements present critical challenges for efficient training and deployment. While Mixture of Experts (MoE) architectures enable efficient scaling through sparse parameter activation, and state-space models like Mamba offer linear-time complexity, principled methods for combining these paradigms remain undeveloped. We introduce HARMONY (Hybrid Architecture Research for Mamba, Optimized with Neural efficiencY), a multi-objective evolutionary neural architecture search framework for discovering efficient hybrid language models that integrate Transformer attention mechanisms, Mixture-of-Experts routing, and Mamba state-space components. Through large-scale distributed search using 16,384 MI250X GPUs on the Frontier supercomputer, HARMONY explores a comprehensive design space encompassing six attention variants (MHA, MQA, GQA, MLA, SWA, and Mamba-2), variable MoE configurations with both routed and shared experts, and extensive Mamba hyperparameters. Our framework discovers heterogeneous architectures that balance training performance with computational efficiency through multi-objective optimization incorporating latency penalties and fitness-based selection. Analysis of discovered architectures reveals that optimal hybrid designs favor heterogeneous component mixing rather than homogeneous patterns, with Mamba-2 and Multi-Head Latent Attention (MLA) emerging as preferred mechanisms. Discovered architectures demonstrate superior training efficiency: our best configuration achieves a final perplexity of 1.0874 with 2.38B parameters while processing 4,320 tokens/second, outperforming significantly larger manually designed models. Full-scale evaluation shows HARMONY's top architectures achieve better loss trajectories than equivalently-sized models using state-of-the-art configurations including Mixtral, Jamba, and Samba. Additionally, we demonstrate 91% weak scaling efficiency when training discovered 36B-parameter models across 1,024 GPUs. HARMONY is released as an open framework with comprehensive tools for building and training hybrid models using expert-data-pipeline parallelism, democratizing access to automated architecture design for next-generation language models.

Herron, Emily [ORNL] (ORCID:0000000273008172)↗
DOE OSTI2025

ESAC (EQ-SANS Assisting Chatbot): Application of large language models and retrieval-augmented generation for enhanced user experience at EQ-SANS

Neutron scattering experiments have played vital roles in exploring materials properties in the past decades. While user interfaces have been improved over time, neutron scattering experiments still require specific knowledge or training by an expert due to the complexity of such advanced instrumentation and the limited number of experiments each person may perform each year. This paper introduces an innovative chatbot application that leverages Large Language Models(LLM) and Retrieval-Augmented Generation (RAG) technologies to significantly enhance the user experience at the EQ-SANS, a small-angle neutron scattering instrument at the Spallation Neutron Source of Oak Ridge National Laboratory. Through a user-centric design approach, the EQ-SANS Assisting Chatbot (ESAC) serves as an interactive reference for users, thereby facilitating the use of the instrument by visiting scientists. By bridging the gap between the users of EQ-SANS and the control systems required to perform their experiments, the ESAC sets a new standard for interactive learning and support for the scientific community using large-scale scientific facilities.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Dataset for Top Model Decision Tree: Selecting Segmentation Models for Reliable Quantitative Analysis in Low- and Ultralow-Dose CryoEM

Motivation Multiple deep learning model architectures can be used to segment bacterial membranes in cryoEM images. However, an AI-based tool advancement is often presented with only a single segmentation model for broad use, and this single model may show inconsistent results across datasets from different users. Here, we present the Top Model Decision Tree, a model screening framework to screen for the best model to generate bacterial inner and outer membrane masks based on user priorities. We use pre-trained segmentation models from YOLOv11, YOLO26, U-Net, Detectron2 and SAM3 fine-tuned on bacterial inner and outer membranes imaged with cryoEM. Run the Framework This notebook must be opened in Google Colab. Mount Google Drive and run with a GPU-based runtime. Open the notebook and follow steps to git clone in folders and files within this repository. There will be a repeating top_model_decision_tree.ipynb (notebook clone) that will not be used. Save your .png binary mask files and .csv table outputs within your Google Drive or download before closing the notebook. The models and all analysis/training scripts are available at [GitHub: https://github.com/Lynnicia/CryoEM_membranes_top_model_decision_tree and https://github.com/Sireesiru/Semantic-Segmentation-of-bacterial-cell-envelope-using-U-Nets.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Deep Image Prior Enabled Full Waveform Inversion (Final Technical Report)

MS Student Naveen Gupta worked on the problem of full waveform inversion (FWI) using neural networks as shown in Figure 1. Our goal was to learn a neural network to represent the subsurface velocity model, which when fed into the FWI module (implemented using a numerical forward model of wave equations) produces amplitude estimates that match with ground-truth observations of amplitude. We used neural networks to solve the inverse problem of estimating velocity distributions for a given seismic amplitude data such that, once trained, our neural network model can generate a distribution of velocity profiles for different random vectors fed as inputs to the neural network model.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

PFLOTRAN modeling data and scripts associated with “Refining the Hydrogeologic Framework of a Large River Corridor Model Using Waterborne Transient Electromagnetics”

NOTE: The manuscript associated with this data package is currently in review. The data may be revised based on reviewer feedback. Upon manuscript acceptance, this data package will be updated with the final dataset and additional metadata. This data package is associated with the publication “Refining the Hydrogeologic Framework of a Large River Corridor Model Using Waterborne Transient Electromagnetics” submitted to Water Resources Research (Terry et al. 2025). The data package contains the groundwater modeling dataset from PFLOTRAN software. It includes the python script for mesh generation, boundary condition setting, PFLOTRAN input deck formation and postprocessing. It couples groundwater flow and species transport for Hanford Reach river corridor and pipelines the model generation and processing. This model can be used to easily generate the model and analysis for Hanford site. It can also be adjusted to other hydrologic area with ease. For details on how to navigate data packages generated by this project, see https://data.ess-dive.lbl.gov/portals/PNNLRiverCorridorSFA/About. The data package consists of 6 folders: (1) “data” contains all necessary data as input and intermediate data for processing; (2) “mesh” contains all mesh related files to generate mesh in Hanford Reach river corridor; (3) “model_run” contains the generated script for PFLOTRAN modeling; (4) “notebooks” contains all the Python script to generate the model; (5) “output” contains all the output from the computation; (6) “postprocessing” contains the Python script to generate scientific figure for manuscript. All files are .csv (comma-separated values), .h5 (HDF5 format), .in (input files), .ipynb (Jupyter notebooks), .p (Python pickle), .png (images), .PNG (images), .py (Python scripts), .pyc (Python bytecode), .r (R scripts), .sh (shell scripts), .txt (text files), .vtu (3D mesh/visualization format), .xz (compressed archive), or .zip (compressed archive).

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Flavor constraints in a generational three-Higgs-doublet model

We propose a three-Higgs-doublet model (3HDM) that goes beyond natural flavor conservation and in which each of the three Higgs doublets couples mainly to a single generation of fermions via nonstandard Yukawa structures. A hierarchy in the vacuum expectation values of the three Higgs doublets can partially address the Standard Model flavor puzzle. In light of the experimentally observed 125 GeV Higgs boson, we primarily work within a 3HDM alignment limit such that a Standard Model-like Higgs is recovered. In order to reproduce the observed Cabibbo–Kobayashi–Maskawa mixing among quarks, the neutral Higgs bosons of the theory necessarily mediate flavor changing neutral currents at the tree level. We consider constraints from neutral kaon, B meson, and D meson mixing as well as from the rare leptonic decays B s / B 0 / K L → μ + μ − / e + e − . We identify regions of parameter space in which the new physics Higgs bosons can be as light as a TeV or even lighter. Published by the American Physical Society 2025

Altmannshofer, Wolfgang (ORCID:0000000316212561)↗
DOE OSTI2024

Battery Degradation Modeling in Hybrid Power Plants: An Island System Unit Commitment Study: Preprint

As hybrid power plants (HPPs), such as photovoltaic (PV) and battery combinations, become increasingly important in power systems with high renewable energy penetration to address PV variability and ensure grid stability. This paper focuses on the urgent need to model the coordination between PV and battery systems in HPPs while accounting for battery degradation. We present a generation scheduling model that explicitly incorporates PV-battery hybridization in the unit commitment problem. Moreover, the cost function of the HPP scheduling problem endogenously considers battery degradation with adjustable weights to strike a balance between minimizing production costs and prolonging battery life, particularly when providing energy arbitrage and ancillary services. Using a realistic island system simulation, we demonstrate that accounting for battery degradation in the scheduling problem can significantly extend battery life with only minor additional production costs.

battery degradation↗
DOE OSTI2025

Oklo Sponsored Testing using PELICAN: System, Subchannel, and High-Fidelity Software Validation (Final CRADA Report)

Argonne National Laboratory (the Contractor), located in Lemont IL, and Oklo, Inc., (the Participant), headquartered in Santa Clara, CA, propose to enter into a Cooperative Research and Development Agreement (CRADA) to perform a gap analysis of thermal hydraulic data, perform prototypical fuel assembly pressure drop and cavitation model validation, generate the experimental data as well as the corresponding uncertainties for this matrix and, finally, develop the validation models with the Argonne system level code SAS4A/SASSYS-1, the Argonne subchannel analysis code DASSH, the Argonne high fidelity code Nek5000, and/or the Idaho National Laboratory code Pronghorn Subchannel. The work outlined below will significantly improve the experimental and validation database currently available for liquid metal fast reactors, thus making it a viable part of a comprehensive reactor design and licensing suite to be used by the participant.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

STILGAR: Subsurface Models for Graymont Pleasant Gap Mine

The detection, location, and monitoring of underground structures are of great importance to national and global security. Tunnels and voids generate seismic signatures detectable at the surface, but using non-invasive seismic data to image near-surface presents several challenges in real-world applications. In this report, we describe the use of a dense surface seismic deployment to generate subsurface models of the Graymont Pleasant Gap mine - a single-layer mine with a complex structure embedded in a high-velocity P-wave limestone bedrock. Our approach consists of three key methods. We use P-wave arrival times from local blast events to perform a tomography inversion with the tomoTD method, constructing a P-wave velocity model of the subsurface. We model the layer above the mine using Rayleigh wave ellipticity and inversion techniques. We leverage ongoing anthropogenic activities to identify and locate noise sources both on the surface and within the subsurface. With this integrated approach we aim to overcome the challenges and enhance our ability to non-invasively characterize underground structures, contributing to improved seismic monitoring techniques.

58 GEOSCIENCES↗
DOE OSTI2024

Spectroscopy-guided discovery of three-dimensional structures of disordered materials with diffusion models

Spectroscopy techniques such as x-ray absorption near edge structure (XANES) provide valuable insights into the atomic structures of materials, yet the inverse prediction of precise structures from spectroscopic data remains a formidable challenge. In this study, we introduce a framework that combines generative artificial intelligence models with XANES spectroscopy to predict three-dimensional atomic structures of disordered systems, using amorphous carbon (a-C) as a model system. In this work, we introduce a new framework based on the diffusion model, a recent generative machine learning method, to predict 3D structures of disordered materials from a target property. For demonstration, we apply the model to identify the atomic structures of a-C as a representative material system from the target XANES spectra. We show that conditional generation guided by XANES spectra reproduces key features of the target structures. Furthermore, we show that our model can steer the generative process to tailor atomic arrangements for a specific XANES spectrum. Finally, our generative model exhibits a remarkable scale-agnostic property, thereby enabling generation of realistic, large-scale structures through learning from a small-scale dataset (i.e. with small unit cells). Our work represents a significant stride in bridging the gap between materials characterization and atomic structure determination; in addition, it can be leveraged for materials discovery in exploring various material properties as targeted.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Verification, Validation, and Calibration Through a Causal Lens

While typical validation and verification approaches focus on identifying the associations between data elements using statistical and machine learning methods, the novel methods in this paper focus instead on identifying causal relationships between data elements. Statistical and machine-learning-based approaches are strictly data-driven, meaning that they provide quantitative comparison measures between data sets without explicitly considering the hypotheses behind them. This can lead to the erroneous conclusion that, if two data sets are close enough, the models that generated them are similar. In addition, when experimental and simulated data differ to an extent that fails to meet the acceptance criteria, calibration techniques are used to tweak simulation model parameters to reduce the gap between the two types of data. This produces the false expectation that a simulation model will match reality. The methods presented in this paper move away from these strictly data-driven methods for validation and calibration toward more robust, model-driven methods based on causal inference. Causal inference aims to identify the possible mechanisms that might have generated data. Thus, this analysis targets the prediction of the effects when one (or more) of the identified mechanisms are altered. There are many approaches to identify, quantify, and illustrate causal relationships. For the scope of this paper, directed graphs are employed as causal models. If the directed graph lacks cycles, it is known as a directed acyclic graph. A node in such a graph represents an observed data element while a directed edge connecting two nodes represents a causal relationship between two variables. The developed causal methods are designed to extract causal models from simulation models and experimental data. Causal models capture the causal relationships between data elements (e.g., simulated and experimental data). In this context, validation and verification are performed by comparing causal models. The proposed approach does not only inform system analysts on how a simulation model matches real-world data, but also identifies elements of the simulation model that should be revised when discrepancies between simulation and experimental data are observed. Through these causal methods, analysts can identify the portion of the model equation(s) that are behind an edge connecting two variables. Hence, once the structural differences between causal models have been determined, model calibration can occur by changing only those model parameters that impact the identified causal relationships.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Integrating a Water Tracer Model Into WRF‐Hydro for Characterizing the Effect of Lateral Flow in Hydrologic Simulations

Abstract Most current land models approximate terrestrial hydrological processes as one‐dimensional vertical flow, neglecting lateral water movement from ridges to valleys. Such lateral flow is fundamental at catchment scales and becomes crucial for finer‐scale land models. To test the effect of incorporating lateral flow toward three‐dimensional representations of hydrological processes in the next generation land models, we integrate a water tracer model into the WRF‐Hydro framework to track water movement from precipitation to discharge and evapotranspiration. This hydrologic‐tracer integrated system allows us to identify the key mechanisms by which lateral flow affects the flow paths and transit times in WRF‐Hydro. By comparing modeling experiments with and without lateral routing in two contrasting catchments, we determine the impacts of lateral flow on the transit times of precipitation event‐water. Results show that with limited hydrologic connectivity, lateral flow extends the transit times by reducing (increasing) event‐water drainage loss (accumulation) in ridges (valleys) and allowing reinfiltration of infiltration‐excess flow, which is missing in most land models. On the contrary with high hydrologic connectivity, lateral flow can effectively accelerate the water release to streams and reduce the transit time. However, the transit times are substantially underestimated by the model compared with isotope‐derived estimates, indicating model limitations in representing flow paths and transit times. This study provides some insights on the fundamental differences in terrestrial hydrology simulated by land models with and without lateral flow representation.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Constraining nuclear mass models using 𝑟-process observables with multiobjective optimization

Modeling nuclear masses, particularly for nuclei far from stability, remains a key objective in nuclear physics. One contemporary approach is machine learning (ML), which trains on experimental data, but can suffer large errors when extrapolating toward neutron-rich species. In nature, such masses shape observables for the rapid neutron capture process (𝑟 process), which in principle could inform ML models. Here, we introduce a multiobjective optimization approach using the Pareto front algorithm. We show that this technique, capable of identifying models that generate 𝑟-process abundances aligning with both solar and stellar data, is a promising method to select ML models with reliable extrapolation power.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Demonstrate new plasticity models for doped UO 2 that capture dislocation mechanisms

In light water reactors, fuel vendors are investigating the use of dopants to modify the properties of UO 2 pellets, with the goal of improving pellet-cladding mechanical interactions during operation. Dopants are expected to ‘soften’ the pellets; that is, the doped pellets have higher plastic deformation than conventional UO 2 . This leads to a reduction in the severity of mechanical pellet-cladding interactions, helping to reduce the hoop strain on the cladding. By minimizing the strain exerted by the pellet on the cladding, it is anticipated that cladding performance under accident conditions can be enhanced (i.e., lowering the risk of burst during a LOCA). Dopants such as chromium (Cr) promote grain growth during pellet fabrication, leading to larger grains; therefore, understanding the link between chemistry, microstructure and mechanical deformation (enhanced creep rates) behavior of UO 2 is critical to helping operators further substantiate the benefits of doping UO 2 . Historically, the nuclear energy industry has relied on empirical models to make assessments of performance. Compared to empirical models, mechanistic physics-based models provide benefits, such as, fewer data points for validation and better extrapolation where experimental data is scarce or non-existent. In this report, Bayesian inference techniques have been applied to a previously developed lower length-scale-informed diffusional creep model. The objective is to i) infer lower-length-scale parameter distributions from available experiment and then ii) determine the uncertainties in the measurable quantity (in this case creep rates) after propagating the inferred lower length scale parameter uncertainties. The approach requires many evaluations of the model, which becomes computationally insurmountable; therefore, a neural-network model is trained to data obtained by sampling the full model over the most important parameters. This neural-network is then used in the Bayesian inference approach to determine probability distributions in the parameter values that represent the uncertainty in the model given what is known from the experiments (posterior). A significant reduction compared to conservative initial (prior) uncertainties is achieved through inference against the experimental data, demonstrating the efficacy of this approach. Furthermore, by accounting for uncertainties in the experimental conditions and sample non-stoichiometry, it is possible to resolve apparent discrepancies in experimental measurements within a self-consistent grain boundary (Coble) creep model that is sensitive to chemistry. This work has been written up and submitted to Nuclear Technology for a special issue on accelerated fuel qualification (AFQ). This uncertainty quantification (UQ) work not only improves the diffusional model, while accounting for uncertainty, but also establishes a framework which can readily be applied to the mechanistic models of dislocation deformation developed in this study. The most likely values from the Bayesian analysis are incorporated into our UO 2 diffusional creep model and a lower length scale-informed irradiation UO 2 creep mechanistic model to generate a dataset. This dataset has been provided to our INL collaborators for training an artificial neural network surrogate model, which will be implemented in the BISON fuel performance code to assess how the results differ from those currently obtained using a fully empirical model and that of using the nominal (uncalibrated) atomic scale parameters in our mechanistic model. Plastic deformation (creep and glide) in UO 2 is a complex phenomenon, governed by multiple underlying processes such as local defect concentrations, applied stresses, and microstructural characteristics. Consequently, there is a need for a meso-scale model with polycrystalline resolution capable of extrapolating to large grain sizes applicable to doped UO 2 , where data is limited and the model can help bridge the knowledge gap. By integrating atomistic data into the polycrystal LApx code, it becomes possible to predict dislocation climb and glide plasticity that simple analytical models cannot accurately represent. The application of atomic-scale data within LApx demonstrated the importance of climb and glide mechanisms in reproducing high-stress UO 2 behavior. Behaviors such as this are crucial to capture and implement in BISON, as parts of the fuel pellet can reach temperatures where glide can occur before pellet cracking. This model which captures dislocation based mechanisms for UO 2 is then used to stand up the doped model accounting for larger grain sizes. It was found that larger grain sizes can lead to enhanced deformation rates in the glide regime, and therefore can help with the pellet cladding mechanical interaction. Therefore if the fuel pellet reaches conditions (stress/temperature) where glide is active, the enhanced creep rates for larger grains in the glide regime (doped UO 2 ) can help with pellet cladding mechanical interactions. Plastic deformation in UO 2 involves multiple mechanisms, including diffusional creep, dislocation climb, and glide. This milestone contains two parts: (1) UQ of a pre-existing lower length scale informed mechanistic diffusional creep model, and (2) development of a new LApx based model for dislocation-mediated creep mechanisms in UO 2 , with application to large-grain doped UO 2 .

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2026

Neural Posterior Estimation for Scalable and Accurate Inverse Parameter Inference in Li-Ion Batteries

Diagnosing the internal state of Li-ion batteries is critical for battery research, operation of real-world systems, and prognostic evaluation of remaining lifetime. By using physics-based models to perform probabilistic parameter estimation via Bayesian calibration, diagnostics can account for the uncertainty due to model fitness, data noise, and the observability of any given parameter. However, Bayesian calibration in Li-ion batteries using electrochemical data is computationally intensive even when using a fast surrogate in place of physics-based models, requiring many thousands of model evaluations. A fully amortized alternative is neural posterior estimation (NPE). NPE shifts the computational burden from the parameter estimation step to data generation and model training, reducing the parameter estimation time from minutes to milliseconds, enabling real-time applications. The present work shows that NPE can infer parameters equally or more accurately than Bayesian calibration, even if it leads to higher voltage reconstruction errors. We also demonstrate that the higher computational costs for data generation are tractable even in high-dimensional cases (ranging from 6 to 27 estimated parameters). The NPE method also offers several interpretability advantages over Bayesian calibration, such as local parameter sensitivity to specific regions of the voltage curve. The NPE method is demonstrated using an experimental fast charge dataset, with parameter estimates validated against measurements of loss of lithium inventory and loss of active material. The implementation is made available in a companion repository (https://github.com/NatLabRockies/BatFIT).

25 ENERGY STORAGE↗
DOE OSTI2024

Hydrogen Leak Modeling for Development of Smart Distributed Monitoring Under Unintended Releases

Hydrogen is a versatile and clean energy carrier that can be produced from various renewable sources such as wind, solar, and hydropower. Hydrogen has the potential to play a crucial role in decarbonizing industrial processes that are currently reliant on fossil fuels and provide long-duration and/or seasonal energy storage to enable electricity decarbonization. Hydrogen can also be used as a fuel for fuel cell vehicles, providing a zero-emission alternative to traditional internal combustion engines. DOE launched the Hydrogen Energy Earthshot (Hydrogen Shot) in June 2021 to reduce the cost of clean hydrogen by 80% to $1 per 1 kilogram in 1 decade ("1 1 1"). While promising, Hydrogen is highly-flammable, and in the presence of oxygen, it can form explosive mixtures. . Therefore, understanding leak scenarios is essential to evaluate and mitigate the safety risks associated with potential hydrogen leaks. An increased understanding of leak behavior, and having tools to model leaks, can help assess how hydrogen would disperse in different environments, influencing emergency response plans and safety measures, and identify potential issues with materials and design systems that can withstand the challenges posed by hydrogen. Recently, researchers have attempted to study hydrogen leaks for development of risk management strategies. However, the focus has been on closed or semi-closed spaces like storage rooms, vehicles, garages, and fueling stations - all promising locations for future hydrogen infrastructure. In this presentation, the modeling environment extends the span of research further by modeling hydrogen leak in an outdoor, open space. We will present the key challenges with modeling hydrogen leaks in an uncontrollable environment, how they were handled, and how modeling results informed sensor selection and placement. A Hydrogen research facility at the National Renewable Energy Laboratory (NREL) was used as a case study to model hydrogen leaks. In the future, Hydrogen wide area detection methodologies will be developed and tested at this site to monitor for unintended and operational hydrogen releases. The data generated from modeling will be used to develop a predictive model to detect hydrogen leak location based on concentration measured by sensors in this open space. Furthermore, the facility was also chosen because controlled hydrogen releases can be performed. A computational fluid dynamics (CFD) based modeling approach was taken to model hydrogen leak. The full-scale hydrogen facility was modeled with a large ambient domain. The electrolyzer at the facility can produce a controlled release rate of 27 kg-H2/hr. Site-specific atmospheric and weather condition data such as wind direction, wind speed at various altitudes, and temperature were used as inputs to the model. To capture the variability of weather conditions, a subset of the weather conditions experienced during daytime hours without precipitation over the course of three months was generated; using established data clustering techniques, a total of 100 condition sets were chosen. The results show statistical distributions and ranges of hydrogen concentrations at locations throughout the domain. These distributions are compared to experimental data from a constant mass flow, controlled hydrogen release at the facility. The stochastic wind conditions of the release make direct validation difficult, therefore, statistical comparison approaches were used. Wind conditions are found to significantly impact the release behavior, including direction and concentration. Sensor selection and placement is proposed for the facility and is now based on release behavior predicted for the facility given its weather patterns; this is much more informed than without the modeling results. The methodology and analysis procedure can be translated to other facilities using modified geometries and site-specific weather conditions. Hydrogen holds great promise as a renewable energy fuel, but ensuring safety in its production, storage, and use is paramount. Studying potential leak scenarios in an open space will help develop sensors to detect hydrogen on a large spectrum of concentration and eventually build a smart distributed monitoring system.

CFD↗
DOE OSTI2025

A Variational Autoencoder Model Toward Molecular Structure Representation Learning of Fuels

Here, in this work, a Variational Autoencoder (VAE)-based data-driven modeling framework is developed with the overarching goal of enabling fuel design. The VAE model is trained on a large dataset with several chemical species to learn a compressed latent space molecular representation. Chemical structure in the form of Simplified Molecular Input Line Entry System (SMILES) string is fed as input, encoded into the VAE latent space, and decoded back to the SMILES string using Long Short-Term Memory (LSTM) networks. Complexities of the VAE training loss function are thoroughly examined by varying the weightage (beta (𝜷) parameter) of the latent space regularization term, thereby assessing the balance between reconstruction accuracy and validity, and focusing on both accurate molecular structure reconstruction and latent space consistency. Two different strategies for 𝜷 variation are evaluated: linear annealing and cyclic annealing. In addition, the impact of total correlation adjustment and hierarchical priors is also studied with regard to the balance between reconstruction fidelity and latent space regularization, and potential issues such as posterior collapse, over-regularization, and poor disentanglement of latent variables. Overall, the best performance of the model is achieved with hierarchical priors and incrementally increasing 𝜷 from 0 to a threshold value of 0.25 over 75 epochs. The generative VAE model can be readily coupled with Quantitative Structure–Property Relationship (QSPR) analysis to develop an integrated end-to-end framework for fuel-property prediction and molecular design of novel promising fuels.

fuel design↗