Search NASA⌕ Search

SEARCH · Search NASA

Results for “error budget”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

GeoLoRA: Geometric integration for parameter efficient fine-tuning

Low-Rank Adaptation (LoRA) has become a widely used method for parameter-efficient fine-tuning of large-scale, pre-trained neural networks. However, LoRA and its extensions face several challenges, including the need for rank adaptivity, robustness, and computational efficiency during the fine-tuning process. We introduce GeoLoRA, a novel approach that addresses these limitations by leveraging dynamical low-rank approximation theory. GeoLoRA requires only a single backpropagation pass over the small-rank adapters, significantly reducing computational cost as compared to similar dynamical low-rank training methods and making it faster than popular baselines such as AdaLoRA. This allows GeoLoRA to efficiently adapt the allocated parameter budget across the model, achieving smaller low-rank adapters compared to heuristic methods like AdaLoRA and LoRA, while maintaining critical convergence, descent, and error-bound theoretical guarantees. The resulting method is not only more efficient but also more robust to varying hyperparameter settings. We demonstrate the effectiveness of GeoLoRA on several state-of-the-art benchmarks, showing that it outperforms existing methods in both accuracy and computational efficiency.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗
DOE OSTI2025

Water Mass Transformation Budgets in Finite‐Volume Generalized Vertical Coordinate Ocean Models

Water Mass Transformation (WMT) theory provides conceptual tools that in principle enable innovative analyses of numerical ocean models; in practice, however, these methods can be challenging to implement and interpret, and therefore remain under-utilized. Our aim is to demonstrate the feasibility of diagnosing all terms in the water mass budget and to exemplify their usefulness for scientific inquiry and model development by quantitatively relating water mass changes, overturning circulations, boundary fluxes, and interior mixing. We begin with a pedagogical derivation of key results of classical WMT theory. We then describe best practices for diagnosing each of the water mass budget terms from the output of Finite-Volume Generalized Vertical Coordinate (FV-GVC) ocean models, including the identification of a non-negligible remainder term as the spurious numerical mixing due to advection scheme discretization errors. We illustrate key aspects of the methodology through the analysis of a polygonal region of the Greater Baltic Sea in a regional demonstration simulation using the Modular Ocean Model v6 (MOM6). We verify the convergence of our WMT diagnostics by brute-force, comparing time-averaged (“offline”) diagnostics on various vertical grids to timestep-averaged (“online”) diagnostics on the native model grid. Finally, we briefly describe a stack of xarray-enabled Python packages for evaluating WMT budgets in FV-GVC models (culminating in the new xwmb package), which is intended to be model-agnostic and available for community use and development.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Analyzing and Exploring Training Recipes for Large-Scale Transformer-Based Weather Prediction

Abstract The rapid rise of deep learning (DL) in numerical weather prediction (NWP) has led to a proliferation of models which forecast atmospheric variables with comparable or superior skill than traditional physics-based NWP. However, among these leading DL models, there is a wide variance in both the training settings and architecture used. Further, the lack of thorough ablation studies makes it hard to discern which components are most critical to success. In this work, we show that it is possible to attain high forecast skill even with relatively off-the-shelf architectures, simple training procedures, and moderate compute budgets. Specifically, we train a minimally modified Swin Transformer V2 (SwinV2) on ERA5 data and find that it attains superior skill in terms of mean-square errors of deterministic forecasts when compared against the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS). Almost all DL–NWP systems share a core set of hyperparameters and design decisions. To aid and expedite future DL–NWP research, we present an in-depth, systematic exploration of different loss functions, model sizes and depths, patch sizes, and multistep training objectives. We also examine the model performance with metrics beyond the typical accuracy (ACC) and RMSE and investigate how the performance scales with model size. Through our open-source code, scoring pipelines, and models, we share our findings on key aspects of the training pipeline. These ablations reduce the necessity for expensive hyperparameter tuning and lower the barrier to entry for future DL–NWP research. Significance Statement This study investigates the potential of using large-scale transformer-based models for weather prediction, showing that it is possible to achieve high forecast accuracy with simpler, off-the-shelf architectures. By training a minimally modified SwinV2 transformer on ERA5 data, we show that the model achieves competitive forecast skill in terms of mean-square error for key variables, outperforming the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS) at all lead times. Our findings suggest that effective training strategies, such as multistep fine-tuning and channel-weighted losses, significantly enhance the model’s performance. However, we also highlight that these improvements come with trade-offs in other areas, such as ensemble spread and high-frequency spatial detail. This work highlights the promise of deep learning in improving weather forecasts, which could lead to better preparedness and response to weather events, ultimately benefiting society by providing more reliable weather predictions.

Willard, Jared D. [Lawrence Berkeley National Labo↗
DOE OSTI2026

Measurements of the 239 Pu(n,f)/ 235 U(n,f) and 238 U(n,f)/ 235 U(n,f) Cross-Section Ratios Using Quasi-Monoenergetic Neutron Beams

Neutron-induced fission cross-section ratio measurements were carried out at Triangle Universities Nuclear Laboratory (TUNL) over multiple experiment campaigns from 2021-2023. The total beam time for these measurements across all the experimental campaigns was approximately three weeks. This work was intended to serve as an independent validation of the fissionTPC cross-section ratio measurements. In contrast to the white spectrum neutron source and time-projection chamber utilized in the fissionTPC, these measurements utilized pulsed, quasi-monoenergetic neutron beams and fission ionization chambers. Therefore, this work has different sources of systematic error and can be used as a complementary measurement. In this report we describe our fission cross-section ratio measurements, analysis, results and provide a detailed uncertainty budget.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2022

Energy Exascale Earth System Model v2.0.1

First patch release of v2.0.0 Changes since v2.0.0 [Important change] Fix ocean threading bug seen in debug cases on Chrysalis with Intel 20.0.4. Was introduced around time of v2.0.0 tag. Does not change v2.0.0 answers on Chrysalis because those didn't use threading or debugging. [EAM] Add semi-lagrangian tracer transport for theta-l (F90 and C++), add new algorithm for finding tropopause, add DSCREAM to allow v2 and SCREAM settings in same code such as adjust_ps [EAM-MMF] 60L default, allow C++ back end of RRTMGP (EAM too). [EAMxx] add nu-top functionality, fix forcing functor, add ttype9 and dcmip2012 tests 2.1, 2.2, and 3 HOMME: remove obsolete remap algs, option to specify dynamics alg indep of tracer, new sponge layer, add imex tests [ELM] Add topography-based subgrid (topounits), add FATES-ELM Nitro., Phos. and CH4 coupling, add land-use ts for NARRM, add lulc for SSP3 RCP7, Fix nutrient fertilization exp test and carbon isotope flux, Fix xactive lnd dry deposition, add lake water storage option, fix plant hydraulics 2d params, fix carbon budget calc, fix soil nutrient conc. bug, fix mosart dam bug, add test for new ELM, MOSART features, fix bug in O3 dry dep stomatal resistances, fix plant hydraulics restart BFB error, update mkmapdata. [MOSART] fix bug for reading the latitude from an unstructured input file, fix oversat in bubble test. [MPAS-ocean] Add CFC11, CFC12 tracers, add 2D spherical transport tests, fix del4 tracer mixing, add MARBL ocean tracer mixing, modify harmonic analysis options, add GPU port of vmix routines, fix calc of ML-averaged BV freq. [MPAS-seaice] Change extents of initial polar disks for oRRS18to6v3 grid, fix ice BGC with MARBL, update spherical test cases, fix DON coupling, Remove Cf from sea ice constants. [MPAS-landice] add CRYO1850-4xCO2 compset [CIME] add GCP, ANL GCE, Spock, Perlmutter, deprecate config_compilers.xml, fix and clean-up cmake macros, fix slurm bindings, refactor CIME internal testing, cleanup SCORPIO perf data, allow position independent compset naming, [also] update v2 benchmarking suite, extend e3sm_prod with throughput and memory checks

E3SM Project, DOE↗
DOE OSTI2022

Forest expansion dominates China’s land carbon sink since 1980

Carbon budget accounting relies heavily on Food and Agriculture Organization land-use data reported by governments. Here we develop a new land-use and cover-change database for China, finding that differing historical survey methods biased China’s reported data causing large errors in Food and Agriculture Organization databases. Land ecosystem model simulations driven with the new data reveal a strong carbon sink of 8.9 ± 0.8 Pg carbon from 1980 to 2019 in China, which was not captured in Food and Agriculture Organization data-based estimations due to biased land-use and cover-change signals. The land-use and cover-change in China, characterized by a rapid forest expansion from 1980 to 2019, contributed to nearly 44% of the national terrestrial carbon sink. In contrast, climate changes (22.3%), increasing nitrogen deposition (12.9%), and rising carbon dioxide (8.1%) are less important contributors. This indicates that previous studies have greatly underestimated the impact of land-use and cover-change on the terrestrial carbon balance of China. This study underlines the importance of reliable land-use and cover-change databases in global carbon budget accounting.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Simulation budgeting for hybrid effective field theories

In this work, we forecast the number of, and requirements on, N-body simulations needed to train hybrid effective field theory (HEFT) emulators for a range of use cases, using a hybrid of HMcode and perturbation theory as a surrogate model. Our accuracy goals, determined with careful consideration of statistical and systematic uncertainties, are 1% accurate in the high-likelihood range of cosmological parameters, and 2% accurate over a broader parameter space volume for k < 1 h Mpc -1 and z < 3. Focusing in part on the 8-parameter w 0 w a CDM+ m ν cosmological model, we find that < 225 simulations are required to meet our error goals over our wide parameter space, including models with rapidly evolving dark energy, given our simulation and emulator recommendations. For a more restricted parameter space volume, as few as 80 simulations are sufficient. We additionally present simulation forecasts for example use cases, and make the code used in our analyses publicly available. These results offer practical guidance for efficient emulator design and simulation budgeting in future cosmological analyses.

cosmological parameters from LSS↗
DOE OSTI2021

A novel approach to characterize the full spectrum and radiative properties of aerosols

Organic aerosols play a significant role in the absorption and radiative budget of the atmosphere, but current modeling approaches do not adequately capture the photo-chemical and oxidative bleaching that occurs during their lifetimes. Most climate models inaccurately assume constant optical properties of these particles, therefore leading to errors in their predictions. In this study we present an approach to better parameterize the optical properties of atmospheric aerosols, including brown carbon (BrC). Using thin-film spectrometry, the full spectrum of reflectivity and emissivity was used to calculate the complex refractive index for multiple species of organic aerosols. This is in contrast to previous techniques which only probe single wavelengths. Since BrC is a complex molecular mixture, we initially focused on single species absorption using four molecular surrogates. Dye-doped polyacrylic-acid thin-films were prepared on glass slides and compared to an aerosolized collection method. Verification of the collection method will open the door to characterizing the complex refractive index of smoke and ambient aerosol.

58 GEOSCIENCES↗
DOE OSTI2020

SCOPE: Spectral cloud optical property estimation using real-time GOES-R longwave imagery

The output of ground-based, solar power generation systems is strongly dependent on cloud cover, which is the main contributor to solar power variability and uncertainty. Cloud optical properties are typically over-simplified in forecasting applications due to the lack of realtime, accurate estimates. In this work, we introduce a method, the Spectral Cloud Optical Property Estimation (SCOPE), for estimating cloud optical properties directly from high-resolution (5-min, 2 km) imagery from Geostationary Operational Environmental Satellite (GOES)-R, which is the newest generation of the GOES system. The SCOPE method couples a two-stream, spectrally resolved radiative model with the longwave GOES-R sensor output to simultaneously estimate the cloud optical depth, cloud top height, and cloud thickness during both day and night at 5-min intervals. The accuracy of SCOPE is evaluated using one year (2018) of downwelling longwave (DLW) radiation measurements from the Surface Radiation Budget Network, which consists of seven sites spread across climatically diverse regions of the contiguous United States. During daytime clear-sky periods, SCOPE predicts DLW within instrument uncertainty (10 W m -2 ) for four of the seven locations, with the remaining locations yielding errors of the order of 11.2, 17.7, and 20.2 W m -2 . For daytime cloudy-sky, daytime all-sky (clear or cloudy), and nighttime all-sky periods, SCOPE achieves root mean square error values of 23.0–34.5 W m -2 for all seven locations. These results, together with the low-latency of the method (1 s per sample), show that SCOPE provides a viable solution to real-time, accurate estimation of cloud optical properties for both day and night.

42 ENGINEERING↗
DOE OSTI2024

ZFP: A compressed array representation for numerical computations

HPC trends favor algorithms and implementations that reduce data motion relative to FLOPS. We investigate the use of lossy compressed data arrays in place of traditional IEEE floating point arrays to store the primary data of calculations. Simulation is fundamentally an exercise in controlled approximation, and error introduced by finite-precision arithmetic (or lossy compression) is just one of several sources of error that need to be managed to ensure sufficient accuracy in a computed result. We describe ZFP, a compressed numerical format designed for in-memory storage of multidimensional arrays, and summarize theoretical results that demonstrate that the error of repeated lossy compression can be bounded and controlled. Furthermore, we establish a relationship between grid resolution and compression-induced errors and show that, contrary to conventional floating point, ZFP reduces finite-difference errors with finer grids. We present example calculations that demonstrate data reduction by 4x or more with negligible impact on solution accuracy. Our results further demonstrate several orders-of-magnitude increase in accuracy using ZFP over IEEE floating point and Posits for the same storage budget.

Lindstrom, Peter↗
DOE OSTI2026

Autonomous Hydrogen Fueling Station

This project “Autonomous Hydrogen Fueling Station” covered the autonomous refueling with both gaseous hydrogen and liquid hydrogen. The part on gaseous hydrogen focused on the development of an autonomous robotic fueling arm that would couple to a fuel cell engine for hydrogen refueling without guidance from the forklift operator and budget period. Research was also covered for the robotic fueling with a commercial vehicle. The second phase of the project created the baseline for an autonomous liquid hydrogen transfer system that would minimize boil off losses by operating at thermodynamically efficient state points. For the development of the robotic fueling arm, testing was conducted to establish a baseline measurement of the accuracy and repeatability of a human operator positioning a lift truck in front of a dispenser. The goal was to establish the range of motion required for an autonomous fueling mechanism to mate a hydrogen nozzle with a receptacle on a fuel cell system installed in a forklift. The final design comprised a selective compliance articulated robot arm (SCARA)-type mechanism with two arms for horizontal motion and a ball screw for vertical movement and color and LIDAR cameras were used for marker identification and proximity awareness to guide the robotic arm to its target receptacle. Initial tests resulted in 199 out of 200 successful attempts at autonomous coupling of the dispensing coupler and a fuel cell engine, without hydrogen. The dispenser prototype was modified to include tubing for both hydrogen fuel and air purge lines, but subsequent tests were confounded by the shoulder motor over current errors which limited the robot from getting to the fully inserted position to achieve a positive latch. Robotic hydrogen refueling was successfully demonstrated over 1.5 hours of testing, Plug completed 29 successful latches with an average number of 4 sequential latches before failure. However, a robot capable of placing the nozzle with more force is required for higher reliability. For budget period two, a small scale (10 kg / transfer) automated control system was designed that would operate valves to control pressure and flow of liquid nitrogen between a source and receiving tank with an aim to minimize boil off losses by operating at the most thermodynamically efficient state points. Control system logic flow and a P&ID were developed prior to system safety characterization via HAZOP. A control narrative and system state points were defined. Delays in approval for a change of project objective and procurement issues precluded the construction and test of the final prototype system.

08 HYDROGEN↗
DOE OSTI2021

Active learning strategy for high fidelity short-term data-driven building energy forecasting

The quality of a data-driven model is heavily dependent on the quality of data. Data from building operation often have data bias problems, which means that the data sample is collected in a way that some members of the intended data population are less likely to be included than others. Data-driven energy forecasting models built on such data hence are biased and could lead to large forecasting errors. Active learning—an effective method to defying data bias—is rarely studied or applied in the area of data-driven building energy forecasting modeling. This paper attempts to fill this gap and explores the application of active learning in data-driven building energy forecasting. The developed strategy in this paper efficiently generate informative training data within a time budget and uses block design to passively consider weather disturbances. The developed active learning strategy is applied and evaluated in both virtual and real-building testbeds against traditional data-driven methods. Via these virtual and real-building evaluation cases, we have demonstrated that the data bias problem typically exists in building operation data is resolved by applying the developed active learning strategy. Furthermore, building energy forecasting models trained from data generated from the active learning strategy have shown improved performances in both model accuracy and model extendibility perspectives. The effectiveness of the block design module is also validated to effectively consider the impact of weather conditions on active learning design.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

HydraGNN_Predictive_GFM_2026 - Ensemble of predictive graph foundation models for atomistic materials modeling

This release contains data and parameters of HydraGNN-based graph foundation models trained as a result of the work published in the pre-print "Exascale Multi-Task Graph Foundation Models for Imbalanced, Multi-Fidelity Atomistic Data" by M. Lupo Pasini et al. (https://arxiv.org/abs/2604.15380). We jointly train on 16 open first-principles datasets (544+ million structures covering 85+ elements) using a multi-task architecture with per-dataset heads and a scalable ADIOS2/DDStore data pipeline. On Frontier, we execute six large-scale DeepHyper hyperparameter optimization campaigns in FP64 and promote the top-performing message-passing models to sustained 2,048-node training, yielding a PaiNN-based lead model. The version of HydraGNN used to generate the outputs provided in this release is HydraGNN v5.0 (https://github.com/ORNL/HydraGNN/releases/tag/v5.0) The list of datasets used for the training of the graph foundation model is the following: 1) Alexandria [1] 2) ANI1x [2] 3) MPTrj [3] 4) Open Catalyst 2020 (OC20) [4] 5) Open Catalyst 2022 (OC22) [5] 6) Open Catalyst 2025 (OC25) [6] 7) Open Direct ir Capture 2023 (ODAC23) [7] 8) Open Materials 2024 (OMat24) [8] 9) Open Molecules 2025 (OMol25) [9] 10) OMol25-neutral (subset of OMol25 that contains only molecules with zero total charge) 11) OMol25-non-neutral (subset of OMol25 that contains only molecules with non-zero total charge) 12) Open Polymers 2026 (OPoly2026) [10] 13) Nabla2DFT [11] 14) QCML [12] 15) QM7X [reference 13] 16) transition1x [14] Dataset references: [1] J. Schmidt et al., “A dataset of 175k stable and metastable materials calculated with the PBEsol and SCAN functionals,” Scientific Data, vol. 9, p. 64, 2022. [2] J. S. Smith et al., “The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules,” Scientific Data, vol. 7, p. 134, 2020. [Online]. Available: https: //www.nature.com/articles/s41597-020-0473-z [3] A. Jain et al., “Commentary: The Materials Project: A materials genome approach to accelerating materials innovation,” APL Materials, vol. 1, no. 1, p. 011002, 07 2013. [Online]. Available: https://doi.org/10.1063/1.4812323 [4] L. Chanussot et al., “Open catalyst 2020 (oc20) dataset and community challenges,” ACS Catalysis, vol. 11, no. 10, pp. 6059–6072, 2021. [Online]. Available: https://doi.org/10.1021/acscatal.0c04525 [5] K. Tran et al., “Open catalyst 2022 (oc22) dataset and challenges for oxidation electrocatalysts,” ACS Catalysis, vol. 13, no. 5, pp. 3066–3084, 2023. [Online]. Available: https://doi.org/10.1021/acscatal.2c05426 [6] S. J. Sahoo et al., “The open catalyst 2025 (oc25) dataset and models for solid-liquid interfaces,” arXiv preprint arXiv:2509.17862, 2025. [Online]. Available: https://arxiv.org/abs/2509.17862 [7] A. Sriram et al., “The open DAC 2023 dataset and challenges for sorbent discovery in direct air capture,” ACS Central Science, vol. 10, no. 5, pp. 923–941, 2024. [8] L. Barroso-Luque et al., “Open materials 2024 (omat24) inorganic materials dataset and models,” 2024. [Online]. Available: https://arxiv.org/abs/2410.12771 [9] D. S. Levine et al., “The open molecules 2025 (OMol25) dataset, evaluations, and models,” 2025. [Online]. Available: https://arxiv.org/abs/2505.08762 [10] D. S. Levine et al., The open polymers 2026 (OPoly26) dataset and evaluations,” arXiv preprint arXiv:2512.23117, 2025. [Online]. Available: https://arxiv.org/abs/2512.23117 [11] K. Khrabrov et al., “Nabla2dft: A universal quantum chemistry dataset of drug-like molecules and a benchmark for neural network potentials,” in NeurIPS 2024 Datasets and Benchmarks Track, 2024. [Online]. Available: https://openreview.net/forum?id=ElUrNM9U8c [12] S. Ganscha et al., “The QCML dataset, quantum chemistry reference data from 33.5M DFT and 14.7B semi-empirical calculations,” Scientific Data, vol. 12, p. 406, 2025. [13] J. Hoja et al., “QM7-X, a comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules,” Scientific Data, vol. 8, p. 43, 2021. [Online]. Available: https://www.nature.com/articles/s41597-021-00812-2 [14] M. Schreiner et al., “Transition1x - a dataset for building generalizable reactive machine learning potentials,” Scientific Data, vol. 9, p. 779, 2022. The folder "datasets_ADIOS2_format" contains the set of pre-processed datasets in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used for the development and training of GFMs in this work. The "datasets_ADIOS2_format" directory contains 2 sub-directories, one for the version "v1" of the datasets and one for the version "v2" of the datasets. The version "v1" of the datasets provides values of the total energy as they are extracted from the original data as it was released by the respective institutions. The version "v2" of the datasets provides values of the energy that have been realigned. The realignment was performed by training a linear regression model that predicts the total energy as a function of the chemical composition of the atomistic structure, and then subtract such prediction from the original value of the total energy. Both folders "v1" and "v2" contain 16 sub-directories, each corresponding to an ADIOS2-formatted dataset The folder "DeepHyper-results" contains the configurational files and model's parameters for all the 186 HPO trials that were successfully completed by the scalable hyperparameter optimization (HPO) runs on Frontier. The content of the folder "DeepHyper-results" I structured as follows: 1) task-list.txt: list of mpnn name, jobid, and deephyper task id 2) gfm_${MPNN}_${JOBID}_0.${TASKID}: run directory with checkpoint files 3) gfm_${MPNN}: deephyper summary directory (*.csv) for each specific MPNN type 4) deephyper-experiment-${JOBID}: output and error logs for each job The file "deephyper-sorted.csv" contains the details of each HydraGNN model built and tested by HPO, obtained by merging the (*.csv) filed from each HPO run executed. Out of all the HPO trials, we selected 10 to continue the training of the respective HydraGNN models. Due to limited computational budget available in the LRN070 allocation we could not complete the training till convergence for all these 10 selected models. The folder "models" contains multiple sub-folders, one per each HydraGNN model trained. Each model sub-folder contains the parameters of each HydraGNN model, with multiple checkpoint-restarts. The list of sub-folders are as follows: 1) multidataset_hpo-BEST1-fp64 2) multidataset_hpo-BEST2-fp64 3) multidataset_hpo-BEST3-fp64 4) multidataset_hpo-BEST4-fp64 5) multidataset_hpo-BEST5-fp64 6) multidataset_hpo-BEST6-fp64 7) multidataset_hpo-BEST7-fp64 8) multidataset_hpo-BEST8-fp64 9) multidataset_hpo-BEST9-fp64 10) multidataset_hpo-BEST10-fp64 Within each one of these folders, additional auxiliary log files are provided with descriptions about how the training proceeded. The lead PaiNN-model is contained inside "multidataset_hpo-BEST6-fp64". The file "mlp_branch_weights" contains the parameters of the multi-layer perceptron (MLP) used to reconcile the predictions of the 16 output decoding heads of the HydragNN architectures. The MLP takes in input the chemical composition of the atomistic structure and predicts averaging weights to linearly mix the predictions of each output decoding head toward consolidating them into a single one. The folder "1.1billion-structure-inference" contains 1.1 billion atomistic structures randomly generated. Each structures is associated with energy and forces predicted with the lead-PaiNN model combined with the MLP model for reconciliation of the multi-branch predictions generated by the 16 output decoding heads. The folder "1.1billion-structure-inference" contains 9,300 (*.tar.gz) subdirectories, one per Frontier compute node used to execute the inference at exascale. Once uncompressed, each (*.tar.gz) subdirectory contains an ADIOS2 (*.bp) file container, where each atomistic structure is stored as a PyTorch-Geometric Data object. The file "export_dataset_environment_variables.sh" contains the environment variables that need to be set before running the HydraGNN code to reproduce the results provided in this dataset release. The code that can be used to load the ADIOS2 files, load HydraGNN models, and run inference is available at: https://github.com/ORNL/HydraGNN/releases/tag/v5.0

36 MATERIALS SCIENCE↗
DOE OSTI2021

Surface Atmosphere Integrated Field Laboratory (SAIL) Science Plan

Mountains are the natural water towers of the world, effectively turning water vapor into readily available fresh water through precipitation, snowpack, and runoff. Unfortunately, Earth system models (ESMs) have persistently been unable to predict the timing and availability of water resources from mountains because the source(s) of model error are difficult to isolate in complex terrain with limited atmospheric or land-surface observations. Further complications arise from the gross scale mismatch between ESM grid box sizes and the relevant scales of mountainous hydrological processes. The mountain hydrometeorology community has repeatedly called for integrated atmospheric and land observations of water and energy budgets in complex terrain that span these scales to establish benchmarks against which scale-dependent models can be further developed.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Optimizing the Heisenberg Vortex Tube for Hydrogen Cooling (Final Technical Report)

Hydrogen utilization at Plug Power sites ranges between 70-93% with the remainder vented and lost to the atmosphere. The goal of this project was to improve hydrogen utilization at Plug Power fulfillment centers via the patented Heisenberg vortex tube (HVT). The HVT combines conventional counterflow vortex tubes with para-orthohydrogen conversion catalyst to allow the cooling of hydrogen flows with no moving parts or external energy inputs. In year 1 of this project the HVT was evaluated to improve hydrogen utilization via the following concepts: 1) increasing liquid hydrogen pump volumetric efficiency by 20% through vapor separation and subcooling of the liquid, 2) reducing liquid hydrogen storage tank boil-off losses by 20% through thermal vapor shielding (TVS), and 3) increase isentropic efficiency of supercritical hydrogen expansion at 40-50 K, from 31% to over 40%. The TVS concept was selected as the most favorable application for further development. While the conceptual analysis was underway, the cryocatalysis hydrogen experiment facility (CHEF) was retrofitted with a new cryocooler, higher pressure condenser tanks, and in-situ fiber-optic Raman probes for ortho-parahydrogen composition analysis at the inlets and outlets of the HVT. To our knowledge this is the first in-situ implementation of cryogenic Raman probes for ortho-parahydrogen analysis. Subsequent testing of catalyzed and non-catalyzed HVT identified a low flow rate, high-conversion efficacy regime well suited for TVS development. Computational fluid dynamics (CFD), Reduced order modeling (ROM), and quantum Monte Carlo atomistic simulations were applied to optimize the design of the HVT implemented for TVS. The simulations and analysis identified a ruthenium-based catalyst as the most optimal for high conversion with little pressure losses and was matched to experimental measurements. These results indicated the design could achieve the 20% boil-off reduction target. An HVT field-trial was designed and constructed by Plug Power and implemented at a customer site. However, budget limitations reduced the amount of catalyst that could be applied to the HVT. With the reduced amount of catalyst, CFD analysis estimated about 3% reduction in boil-off. A similar amount of boil-off reduction was observed during preliminary measurements during tank commissioning. However, the results have a large margin for error and were operated at relatively low tank temperatures when the HVT has the lowest potential for improvement. This project demonstrated the use of para-orthohydrogen catalysis for reducing liquid hydrogen tank boil-off losses. The field trial system is anticipated to provide long-term experimental measurements on tank operational performance that will allow Plug Power to instrument additional tanks in the future. Recommendations for future work include the development of higher activity, lower cost para-orthohydrogen conversion catalyst for use in the HVT. Tank process optimizations could maximize the efficacy of the HVT and further improve hydrogen utilization.

08 HYDROGEN↗
DOE OSTI2024

Understanding the Biases in Global Monsoon Simulations from the Perspective of Atmospheric Energy Transport

Understanding global monsoon (GM) variability and projecting its future changes rely heavily on climate models. However, climate models generally show pronounced biases in GM simulations, and the reasons for this remain unclear. Here, in this study, we evaluate the performance of 20 pairs of climate models that participated in both phase 5 of the Coupled Model Intercomparison Project (CMIP5) and phase 6 of CMIP (CMIP6) and identify the sources of their GM simulation biases from an energy transport perspective. The multimodel mean improvement in CMIP6 compared to CMIP5 is demonstrated by the increasing skill scores for various GM metrics from 0.20–0.79 to 0.48–0.83. More specifically, the dry biases in the Northern Hemisphere Summer Monsoon (NHSM) precipitation in CMIP5 [root-mean-square error (RMSE): 1.85 mm day −1 ] are reduced in CMIP6 (RMSE: 1.66 mm day −1 ). This higher simulation skill is associated with higher skill in simulating the precipitation-solstitial mode, monsoon intensity, and monsoon domains. The improvement in the NHSM precipitation simulation results from that in the meridional transport of atmospheric energy. Atmospheric energy budget analysis shows that the negative biases in downward surface longwave radiation and northward energy transport are smaller in CMIP6 than in CMIP5 in the boreal summer, resulting in a more realistic interhemispheric thermal contrast and meridional gradient of moist static energy. However, a major weakness of the CMIP6 models is found in the Southern Hemisphere Summer Monsoon precipitation simulation due to the positive bias in the top-of-the-atmosphere downward longwave radiation. This study shows that reasonably reproducing the meridional global atmospheric energy transportation is necessary for skillful GM simulation.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Adaptive experimental design for multi-fidelity surrogate modeling of multi-disciplinary systems

We present an adaptive algorithm for constructing surrogate models of multi-disciplinary systems composed of a set of coupled components. With this goal we introduce “coupling” variables with a priori unknown distributions that allow surrogates of each component to be built independently. Once built, the surrogates of the components are combined to form an integrated-surrogate that can be used to predict system-level quantities of interest at a fraction of the cost of the original model. The error in the integrated-surrogate is greedily minimized using an experimental design procedure that allocates the amount of training data, used to construct each component-surrogate, based on the contribution of those surrogates to the error of the integrated-surrogate. Here, the multi-fidelity procedure presented is a generalization of multi-index stochastic collocation that can leverage ensembles of models of varying cost and accuracy, for one or more components, to reduce the computational cost of constructing the integrated-surrogate. Extensive numerical results demonstrate that, for a fixed computational budget, our algorithm is able to produce surrogates that are orders of magnitude more accurate than methods that treat the integrated system as a black-box.

42 ENGINEERING↗
DOE OSTI2020

Influence of sun zenith angle on canopy clumping and the resulting impacts on photosynthesis

Addressing the impact of vegetation architecture on shortwave radiation transfer in land surface models is important for accurate weather forecasting, carbon budget estimates, and climate predictions. This paper investigates to what extent it is possible to retrieve structural parameters of two different parameterization schemes from direct transmittance derived from digital hemispherical photography and 3D radiative transfer modeling for two study sites with different vegetation canopy architectures. Neglecting the representation of 3D canopy structure in radiative transfer schemes leads to significant errors in shortwave radiation partitioning (up to 3.5 times more direct transmittance in the 3D model). Structural parameters, referred to as whole canopy ‘clumping indices’, were obtained in order to evaluate the impact of angular variation in clumping on shortwave radiation transfer. Impacts on photosynthesis were evaluated at site level with the UKESM land surface model, JULES. A comparison between flux tower derived and modeled photosynthesis indicates that considering zenith angular variations of structural parameters in the radiative transfer scheme of the UKESM land surface model significantly improves photosynthesis prediction in light limited ecosystems (from RMSE = 2.91 μmol CO 2 .m -2 .s -1 to RMSE = 1.51 μmol CO 2 .m -2 .s -1 , 48% smaller), typically with enhanced photosynthesis from bottom layers.

54 ENVIRONMENTAL SCIENCES↗