SEARCH · Engineering Papers

Results for “performance model”

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.

SourceResearch connectionFrom yearThrough year

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

At least 451 records · Page 25

All sources NASA NTRS DOE OSTI

DOE OSTI2026

Beyond interpolation: Physics-inspired gating transformers for extrapolating irradiation conditions to novel nuclear fuels

The qualification of advanced nuclear fuels relies on irradiation experiments in test reactors that emulate commercial conditions. Designing these tests requires accurate prediction of key irradiation quantities, particularly heat generation rate and burnup, yet obtaining them typically involves computationally expensive multi-step simulation workflows. We propose a physics-inspired gating transformer (PIGT) that integrates an inverse-square, distance-based attenuation into the encoder representation to bias attention toward physically relevant spatial relationships while retaining data-driven flexibility. Using MiniFuel irradiation data from the High Flux Isotope Reactor at Oak Ridge National Laboratory, we benchmark against ensemble methods, feedforward and recurrent networks, convolutional models, and standard transformers. While baseline models perform well under interpolation, they exhibit a pronounced generalization gap when evaluated on fuels not included in the training set. The proposed model consistently improves extrapolative accuracy and stability, yielding the strongest performance on unseen fuel configurations. These results indicate that a lightweight physics structure embedded within attention mechanisms can substantially improve robustness, enabling more reliable surrogate predictions to accelerate the design of nuclear fuel irradiation experiments.

Fuel qualification↗

DOE OSTI2025

On the effectiveness of neural operators at zero-shot weather downscaling

Machine-learning (ML) methods have shown great potential for weather downscaling. These data-driven approaches provide a more efficient alternative for producing high-resolution weather datasets and forecasts compared to physics-based numerical simulations. Neural operators, which learn solution operators for a family of partial differential equations, have shown great success in scientific ML applications involving physics-driven datasets. Neural operators are grid-resolution-invariant and are often evaluated on higher grid resolutions than they are trained on, i.e., zero-shot super-resolution. Given their promising zero-shot super-resolution performance on dynamical systems emulation, we present a critical investigation of their zero-shot weather downscaling capabilities, which is when models are tasked with producing high-resolution outputs using higher upsampling factors than are seen during training. To this end, we create two realistic downscaling experiments with challenging upsampling factors (e.g., 8x and 15x) across data from different simulations: the European Centre for Medium-Range Weather Forecasts Reanalysis version 5 (ERA5) and the Wind Integration National Dataset Toolkit. While neural operator-based downscaling models perform better than interpolation and a simple convolutional baseline, we show the surprising performance of an approach that combines a powerful transformer-based model with parameter-free interpolation at zero-shot weather downscaling. We find that this Swin-Transformer-based approach mostly outperforms models with neural operator layers in terms of average error metrics, whereas an Enhanced Super-Resolution Generative Adversarial Network-based approach is better than most models in terms of capturing the physics of the ground truth data. We suggest their use in future work as strong baselines.

17 WIND ENERGY↗

DOE OSTI2026

BISON analyses of TRISO fuel performance, its dependence on time-at-temperature, and possible implications for fuel design and qualification

The Advanced Gas Reactor Fuel Development and Qualification (AGR) program has established a substantial technical foundation to support private entry into the U.S. high-temperature gas-cooled reactor market. However, emerging tristructural isotropic (TRISO)-fueled reactor applications include small modular reactors and microreactors with longer fuel residence times, which may expose fuels to higher time-at-temperature (TAT) values than were explored by the AGR program. Increased TAT could affect diffusive and thermomechanical behaviors such as Pd penetration, fission gas release, creep, and fission product transport. In this work, we applied multiscale best-estimate BISON fuel performance modeling to assess these effects within a representative design space based on the AGR-5/6/7 experiment and analyzed trends in predicted particle and compact fuel performance metrics with possible implications for near-term fuel design and qualification. BISON unambiguously predicted that TRISO fuel performance is sensitive to TAT. Increasing TAT was not predicted to increase the magnitude of failure-inducing tangential stresses in particle coating layers. Predictions obtained using a mechanistic model for Pd penetration indicated that penetration depth does not depend strongly on TAT. While these observations suggest that AGR testing provides a conservative upper bound for the steady-state operation of TRISO particles at lower powers and higher residence times, BISON also predicted that the release of poorly retained Ag would increase with TAT. Because these analyses applied models to extrapolate beyond the available experimental data, the authors recommend performing targeted experiments to confirm these predictions. Nevertheless, these predictions may provide reactor developers with enough confidence to make near-term design decisions associated with the potential fuel performance trade-offs of increasing TAT.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2024

3D modeling of the Staged Z-pinch with the FLASH code

The Staged Z-pinch (SZP) fusion concept is a magneto-inertial compression scheme developed by Magneto-Inertial Fusion Technologies, Inc. (MIFTI), in which small amounts of fusion fuel are brought to fusion-relevant conditions by passing multi-million amperes strong current through a cylindrical shell of high atomic number material. One- and two-dimensional modeling performed by MIFTI with the MACH2 code, suggests that net fusion energy gain can be achieved when currents in the 10 million amperes range compress a 50%-50% mixture of deuterium and tritium gas. In this project we enhance and use the FLASH code to execute high-fidelity, simulations of various SZP configurations, a collaboration between the Flash Center for Computational Science at the University of Rochester and the Laboratory for Laser Energetics, and MIFTI. FLASH is a high-performance computing, multi-physics, radiation magnetohydrodynamic (MHD) code with extended physics capabilities, which is developed at the Flash Center. The goal of this project is to assess the shell/fuel stability of the pinch to two- and three-dimensional MHD instabilities and to utilize FLASH’s extended physics capabilities to understand how extended-MHD effects impact implosion dynamics and plasma conditions at stagnation. The project is a natural extension of an ongoing collaborative effort between MIFTI and the Flash Center through the U.S. DOE Advanced Research Projects Agency-Energy (ARPA-E) BETHE program and can provide MIFTI with simulation capabilities that are currently beyond their reach with MACH2. Ultimately, MIFTI wants FLASH to become one of their simulation workhorses for reliable, high-fidelity SZP platform design. During the project, the Flash Center team developed a suite of in one- and two-dimensional FLASH simulations to model different variants of the SZP platform, performed code-to-code comparisons with MACH2, and used the FLASH code to design and model SZP experiments like the Double Eagle experiments that MIFTI recently performed. Also, the Flash Center team developed the capabilities of FLASH to be able to do three-dimensional simulations of pulsed-power experiments with the code for the first time. Two publications from this effort are currently under review and two more are in preparation. MIFTI has also committed to the use of FLASH for future SZP simulation efforts.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2024

Brain‐age prediction: Systematic evaluation of site effects, and sample age range and size

Abstract Structural neuroimaging data have been used to compute an estimate of the biological age of the brain (brain‐age) which has been associated with other biologically and behaviorally meaningful measures of brain development and aging. The ongoing research interest in brain‐age has highlighted the need for robust and publicly available brain‐age models pre‐trained on data from large samples of healthy individuals. To address this need we have previously released a developmental brain‐age model. Here we expand this work to develop, empirically validate, and disseminate a pre‐trained brain‐age model to cover most of the human lifespan. To achieve this, we selected the best‐performing model after systematically examining the impact of seven site harmonization strategies, age range, and sample size on brain‐age prediction in a discovery sample of brain morphometric measures from 35,683 healthy individuals (age range: 5–90 years; 53.59% female). The pre‐trained models were tested for cross‐dataset generalizability in an independent sample comprising 2101 healthy individuals (age range: 8–80 years; 55.35% female) and for longitudinal consistency in a further sample comprising 377 healthy individuals (age range: 9–25 years; 49.87% female). This empirical examination yielded the following findings: (1) the accuracy of age prediction from morphometry data was higher when no site harmonization was applied; (2) dividing the discovery sample into two age‐bins (5–40 and 40–90 years) provided a better balance between model accuracy and explained age variance than other alternatives; (3) model accuracy for brain‐age prediction plateaued at a sample size exceeding 1600 participants. These findings have been incorporated into CentileBrain ( https://centilebrain.org/#/brainAGE2 ), an open‐science, web‐based platform for individualized neuroimaging metrics.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2025

Machine learning prediction of enzyme optimum pH

The relationship between pH and enzyme catalytic activity, especially the optimal pH (pH opt ) at which enzymes function, is critical for biotechnological applications. Hence, computational methods to predict pH opt will enhance enzyme discovery and design by facilitating accurate identification of enzymes that function optimally at specific pH levels, and by elucidating sequence-function relationships. Here, in this study, we proposed and evaluated various machine learning methods for predicting pH opt , conducting extensive hyperparameter optimization and training over 11,000 model instances. Our results demonstrate that models utilizing language model embeddings markedly outperform other methods in predicting pHopt. We present EpHod, the best-performing model, to predict pHopt, making it publicly available to researchers. From sequence data, EpHod directly learns structural and biophysical features that relate to pH opt , including proximity of residues to the catalytic centre and the accessibility of solvent molecules. Overall, EpHod presents a promising advancement in pH opt prediction and will potentially speed up the development of enzyme technologies.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

Separation of Hydrogen Using Pd/Ag Membranes: Experimental and Modeling Results with Potential Application to Direct Internal Recycle

Implementation of fusion energy requires processing the deuterium-tritium (D-T) mixture used to fuel the reaction, and separation of hydrogen isotopes from other gases is imperative. Specifically, the separation of hydrogen isotopes from helium is a matter of importance to the fusion fuel cycle community. Initial testing with a palladium-silver (Pd-Ag) membrane indicates that even moderate vacuum (~100 torr permeate pressure) can provide a high degree of separation (>90%) at a high ratio of H 2 to He. Given the presence of He in many fusion systems, a high technology readiness level (TRL) for Q 2 /He (where Q represents any isotope of hydrogen) separations is needed. This study demonstrates the efficacy of H 2 removal from He via permeation and potential applications for direct internal recycle. Modeling will accompany the experimental campaign to generate a predictive capability and quantify the separation performance. Modeling from previous hydrogen permeation studies has demonstrated that the typical Sieverts’ law fails to predict the measured permeation rates at high hydrogen fluxes. Existing models are being refined to integrate the effects of surface phenomena into permeation predictions, which have been expanded to account for mixtures with large ranges of Q 2 concentrations. These data will improve the TRL of permeators as a separation technology for the fusion fuel cycle.

08 HYDROGEN↗

DOE OSTI2020

WRF Modelling of Deep Convection and Hail for Wind Power Applications

Deep convection and the related occurrence of hail, intense precipitation and wind gusts represent a hazard to a range of energy infrastructure including wind turbine blades. Wind turbine blade leading edge erosion (LEE) is caused by the impact of falling hydrometeors onto rotating wind turbine blades. It is a major source of wind turbine maintenance costs and energy losses from wind farms. In the United States Southern Great Plains (SGP), where there is widespread wind energy development, deep convection and hail events are common, increasing the potential for precipitation-driven LEE. A 25-day Weather Research and Forecasting (WRF) model simulation conducted at convection-permitting resolution and using a detailed microphysics scheme is carried out for the SGP to evaluate the effectiveness in modelling the wind and precipitation conditions relevant to LEE potential. WRF output for these properties is evaluated using radar observations of precipitation (including hail) and reflectivity, in situ wind speed measurements and wind power generation. This research demonstrates some skill for the primary drivers of LEE. Wind speeds, rainfall rates and precipitation totals show good agreement with observations. Here, the occurrence of precipitation during power producing wind speeds is also shown to exhibit fidelity. Hail events frequently occur during periods when wind turbines are rotating and are especially important to LEE in the SGP. The presence of hail is modelled with a mean proportion correct of 0.77 and an odds ratio 4.55. Further research is needed to demonstrate sufficient model performance to be actionable for the wind energy industry and there is evidence for positive model bias in cloud reflectivity.

17 WIND ENERGY↗

DOE OSTI2024

Understanding early HIV-1 rebound dynamics following antiretroviral therapy interruption: The importance of effector cell expansion

Most people living with HIV-1 experience rapid viral rebound once antiretroviral therapy is interrupted; however, a small fraction remain in viral remission for an extended duration. Understanding the factors that determine whether viral rebound is likely after treatment interruption can enable the development of optimal treatment regimens and therapeutic interventions to potentially achieve a functional cure for HIV-1. We built upon the theoretical framework proposed by Conway and Perelson to construct dynamic models of virus-immune interactions to study factors that influence viral rebound dynamics. We evaluated these models using viral load data from 24 individuals following antiretroviral therapy interruption. The best-performing model accurately captures the heterogeneity of viral dynamics and highlights the importance of the effector cell expansion rate. Our results show that post-treatment controllers and non-controllers can be distinguished based on the effector cell expansion rate in our models. Furthermore, these results demonstrate the potential of using dynamic models incorporating an effector cell response to understand early viral rebound dynamics post-antiretroviral therapy interruption.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2023

Knowledge Distillation for Anomaly Detection

Unsupervised deep learning techniques are widely used to identify anomalous behaviour. The performance of such methods is a product of the amount of training data and the model size. However, the size is often a limiting factor for the deployment on resource-constrained devices. Here, we present a novel procedure based on knowledge distillation for compressing an unsupervised anomaly detection model into a supervised deployable one and we suggest a set of techniques to improve the detection sensitivity. Compressed models perform comparably to their larger counterparts while significantly reducing the size and memory footprint.

Pol, Adrian Alan↗

DOE OSTI2023

Flash Diffusivity Measurement of Semi-Porous Insulation Material Intended for Radioisotope Thermoelectric Generators

This research involves the analysis of flash-heating data pursuant to find the thermal diffusivity of a semi-porous insulation material (Min-K) that is being considered for use in a radioisotope thermoelectric generator (RTG). An RTG uses radioactive nuclear fuel to produce electricity through a temperature difference imposed on a bimetallic thermocouple. Insulation is required to protect sensitive equipment from high temperatures and to conserve heat in the fuel. Using flash diffusivity temperature data, various simulations were fitted in order to find the most appropriate heat transfer model for the experiments. Four models were allowed to compete and the standard deviation of the residuals were compared for each model in evaluating model performance. The residuals are simply the difference in temperatures between the measurements and the mathematical models at each measurement point.

McMasters, Robert L.↗

DOE OSTI2026

Generalizing synthetic data-trained acoustic predictive models to real-world measurements

Acoustic Resonance Spectroscopy (ARS) is highly sensitive to structural properties such as material, geometry, and environmental conditions; as a consequence, it can noninvasively measure internal properties that are unobservable by most other methods. Because of its sensing capabilities and low implementation cost and complexity, ARS has potential as a paradigm shift in noninvasive sensing, characterization, and monitoring applications. However, extracting specific properties from ARS measurements, comprising the vibration spectrum of a test object, is challenging due to the sensitivity of the spectra to other structural changes not being measured, e.g. manufacturing tolerances, component coupling, environmental variation, etc. Neural Networks are promising tools for identifying trends in ARS measurements, but their training typically requires large datasets, which are often impractical to obtain for real-world systems. Synthetic data can be simulated efficiently, but discrepancies between synthetic and real-world data frequently lead to poor generalization when testing on the real-world data. We propose a novel ARS model training framework that enables networks trained exclusively on synthetic ARS data to generalize effectively to real-world measurements. Our approach leverages the Correlation Alignment (CORAL) technique to enforce the extraction of features common to both synthetic and real-world domains. As a case study, we demonstrate noninvasive ARS-based pressure measurements in sealed systems. Finite element method (FEM) simulations were used to generate synthetic training data across diverse vessel configurations and pressure conditions, and model performance was then tested on real-world measurements. We demonstrate that robust machine learning models for ARS can be developed without large real-world datasets, significantly broadening the applicability of ARS for noninvasive sensing. Moreover, the approach is extensible to other sensing modalities where synthetic data are abundant but real-world data are limited.

36 MATERIALS SCIENCE↗

DOE OSTI2024

Recent Progress on Surface Water Quality Models Utilizing Machine Learning Techniques

Surface waterbodies are heavily exposed to pollutants caused by natural disasters and human activities. Empowering sensor technologies in water quality monitoring, sufficient measurements have become available to develop machine learning (ML) models. Numerous ML models have quickly been adopted to predict water quality indicators in various surface waterbodies. This paper reviews 78 recent articles from 2022 to October 2024, categorizing water quality models utilizing ML into three groups: Point-to-Point (P2P), which estimates the current target value based on other measurements at the same time point; Sequence-to-Point (S2P), which utilizes previous time series data to predict the target value at one time point ahead; and Sequence-to-Sequence (S2S), which uses previous time series data to forecast sequential target values in the future. The ML models used in each group are classified and compared according to water quality indicators, data availability, and model performance. Widely used strategies for improving performance, including feature engineering, hyperparameter tuning, and transfer learning, are recognized and described to enhance model effectiveness. The interpretability limitations of ML applications are discussed. This review provides a perspective on emerging ML for surface water quality models.

machine learning (ML)↗

DOE OSTI2024

Thermal Management of Wide-Bandgap Semiconductor Amplifiers Used for Plasma Heating and Control

Princeton Fusion Systems (PFS) has designed, built, and tested a Load Switch printed circuit board (PCB) to demonstrate the capabilities of 2 kV silicon carbide (SiC) cascodes in development by Qorvo towards plasma heating and control applications. Initial tests have been conducted at low power (~100 W) for validation with thermal finite element analysis (FEA) modeling performed by the National Renewable Energy Laboratory (NREL). Comparisons of experimental data with the thermal modeling results, along with considerations for operating in plasma systems, will be discussed.

CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND SU↗

DOE OSTI2026

Benchmarking soil moisture and its relationship to ecohydrologic variables in Earth System Models

Soil moisture (SM) is a key regulator of ecosystem biogeophysics, influencing plant water relations and land-atmosphere energy exchanges. We evaluate the representation of SM in 16 Earth System Models from the Coupled Model Intercomparison Project Phase 6 (CMIP6) using the International Land Model Benchmarking (ILAMB) framework, focusing on surface (0–5, 0–10 cm) and rootzone (0–100 cm) depths, as well as key ecohydrological variables like gross primary productivity (GPP), leaf area index (LAI), and evapotranspiration (ET), and their coupling. Models are benchmarked against multiple observational and assimilated datasets to assess both state variables and cross-variable relationships. Surface SM is generally well represented (r > 0.87), while rootzone SM variability is systematically overestimated (normalized standard deviation > 1). ET shows strong agreement with observations (r > 0.9), whereas GPP and LAI exhibit larger inter-model spread. Skill in individual variables does not guarantee realistic SM–ecohydrology coupling, which varies strongly across models and depends on the reference dataset. Köppen-based regional analyses reveal strong regime dependence, with several models performing well in Tropical and Temperate regions but degrading in Continental (high-latitude) zones. Across both global and regional benchmarks, models cluster by land surface framework, indicating that structural choices in soil hydrology and soil–plant coupling exert a first-order control on performance. These results provide process-relevant benchmarks and suggest that improving the representation of vertical soil structure, rooting depth distributions, and soil–plant hydraulic coupling will be central to advancing soil moisture realism in next-generation Earth system models.

CMIP6↗

DOE OSTI2023

A Comparative Study of Deep Learning Models for Fracture and Pore Space Segmentation in Synthetic Fractured Digital Rocks

This study focuses on the comparative study of deep learning (DL) models for pore space and discrete fracture networks (DFNs) segmentation in synthetic fractured digital rocks, specifically targeting low-permeability rock formations, such as shale and tight sandstones. Accurate characterization of pore space and DFNs is critical for subsequent property analysis and fluid flow modeling. Four DL models, SegNet, U-Net, U-Net-wide, and nested U-Net (i.e., U-Net++), were trained, validated, and tested using synthetic datasets, including input and label image pairs with varying properties. The model performance was assessed regarding pixel-wise metrics, including the F1 score and pixel-wise difference maps. In addition, the physics-based metrics were considered for further analysis, including sample porosity and absolute permeability. Particularly, We first simulated the permeability of porous media containing only pore space and then simulated the permeability of porous media with DFNs added. The difference between these two values is used to quantify the connectivity of segmented DFNs, which is an important parameter for low-permeability rocks. The pixel-wise metrics showed that the nested U-Net model outperformed the rest of the DL models in pore space and DFNs segmentation, with the SegNet model exhibiting the second-best performance. Particularly, nested U-Net enhanced segmentation accuracy for challenging boundary pixels affected by partial volume effects. The U-Net-wide model achieved improved accuracy compared to the U-Net model, which indicated the influence of parameter numbers. Similarly, nested U-Net has the closest match to the ground truth of physics-based metrics, including the porosity of pore space and DFNs, and the permeability difference quantifying the connectivity of DFNs. The findings highlight the effectiveness of DL models, especially the U-Net++ model with nested architecture and redesigned skip connections, in accurately segmenting pore spaces and DFNs, which are crucial for pore-scale fluid flow and transport simulation in low-permeability rocks.

Wang, Hongsheng↗

DOE OSTI2025

Bayesian learning with Gaussian processes for low-dimensional representations of time-dependent nonlinear systems

This work presents a data-driven method for learning low-dimensional time-dependent physics-based surrogate models whose predictions are endowed with uncertainty estimates. We use the operator inference approach to model reduction that poses the problem of learning low-dimensional model terms as a regression of state space data and corresponding time derivatives by minimizing the residual of reduced system equations. Standard operator inference models perform well with accurate training data that are dense in time, but producing stable and accurate models when the state data are noisy and/or sparse in time remains a challenge. Another challenge is the lack of uncertainty estimation for the predictions from the operator inference models. Our approach addresses these challenges by incorporating Gaussian process surrogates into the operator inference framework to (1) probabilistically describe uncertainties in the state predictions and (2) procure analytical time derivative estimates with quantified uncertainties. The formulation leads to a generalized least-squares regression and, ultimately, reduced-order models that are described probabilistically with a closed-form expression for the posterior distribution of the operators. The resulting probabilistic surrogate model propagates uncertainties from the observed state data to reduced-order predictions. Furthermore, we demonstrate the method is effective for constructing low-dimensional models of two nonlinear partial differential equations representing a compressible flow and a nonlinear diffusion–reaction process, as well as for estimating the parameters of a low-dimensional system of nonlinear ordinary differential equations representing compartmental models in epidemiology.

Data-driven model reduction↗

DOE OSTI2024

A new data-driven map predicts substantial undocumented peatland areas in Amazonia

Tropical peatlands are among the most carbon-dense terrestrial ecosystems yet recorded. Collectively, they comprise a large but highly uncertain reservoir of the global carbon cycle, with wide-ranging estimates of their global area (441 025–1700 000 km 2 ) and below-ground carbon storage (105–288 Pg C). Substantial gaps remain in our understanding of peatland distribution in some key regions, including most of tropical South America. Here we compile 2413 ground reference points in and around Amazonian peatlands and use them alongside a stack of remote sensing products in a random forest model to generate the first field-data-driven model of peatland distribution across the Amazon basin. Our model predicts a total Amazonian peatland extent of 251 015 km 2 (95th percentile confidence interval: 128 671–373 359), greater than that of the Congo basin, but around 30% smaller than a recent model-derived estimate of peatland area across Amazonia. The model performs relatively well against point observations but spatial gaps in the ground reference dataset mean that model uncertainty remains high, particularly in parts of Brazil and Bolivia. For example, we predict significant peatland areas in northern Peru with relatively high confidence, while peatland areas in the Rio Negro basin and adjacent south-western Orinoco basin which have previously been predicted to hold Campinarana or white sand forests, are predicted with greater uncertainty. Similarly, we predict large areas of peatlands in Bolivia, surprisingly given the strong climatic seasonality found over most of the country. Very little field data exists with which to quantitatively assess the accuracy of our map in these regions. Data gaps such as these should be a high priority for new field sampling. This new map can facilitate future research into the vulnerability of peatlands to climate change and anthropogenic impacts, which is likely to vary spatially across the Amazon basin.

54 ENVIRONMENTAL SCIENCES↗