Search NASA⌕ Search

SEARCH · Search NASA

Results for “model performance”

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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Prediction of laser beam spatial profiles in a high-energy laser facility by use of deep learning

We adapt the significant advances achieved recently in the field of generative artificial intelligence/machine-learning to laser performance modeling in multipass, high-energy laser systems with application to high-shot-rate facilities relevant to inertial fusion energy. Advantages of neural-network architectures include rapid prediction capability, data-driven processing, and the possibility to implement such architectures within future low-latency, low-power consumption photonic networks. Four models were investigated that differed in their generator loss functions and utilized the U-Net encoder/decoder architecture with either a reconstruction loss alone or combined with an adversarial network loss. We achieved inference times of 1.3 ms for a 256 × 256 pixel near-field beam with errors in predicted energy of the order of 1% over most of the energy range. It is shown that prediction errors are significantly reduced by ensemble averaging the models with different weight initializations. These results suggest that including the temporal dimension in such models may provide accurate, real-time spatiotemporal predictions of laser performance in high-shot-rate laser systems.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

User's Manual for the FE/NETL Onshore CO 2 EOR Cost Model, Version 1

This user's manual describes the conceptual and mathematical basis for the FE/NETL Onshore CO 2 EOR Cost Model (a Fortran program). The model performs a cash flow analysis to estimate the cost of implementing CO 2 EOR using supercritical CO 2 by incorporating oil field performance outputs for a pattern from the FE/NETL CO 2 Prophet Model (available on NETL's website along with its associated user's manuals under the Collection Name: FE/NETL CO 2 Prophet Model) and implementing patterns to develop an oil field for CO 2 EOR. The model calculates capital costs, operation and maintenance costs, and financing costs. The user’s manual also describes how to run the FE/NETL Onshore CO 2 EOR Cost Model, along with the model’s file structure, inputs and outputs. The FE/NETL Onshore CO 2 EOR Cost Model is available on NETL's website under the Collection Name: FE/NETL Onshore CO 2 EOR Cost Model.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Recovering the Star Formation Histories of Recently Quenched Galaxies: The Impact of Model and Prior Choices

Accurate models of the star formation histories (SFHs) of recently quenched galaxies can provide constraints on when and how galaxies shut down their star formation. The recent development of nonparametric SFH models promises the flexibility required to make these measurements. However, model and prior choices significantly affect derived SFHs, particularly for post-starburst galaxies (PSBs), which have sharp changes in their recent SFH. In this paper, we create mock PSBs, then use the Prospector SED fitting software to test how well four different SFH models recover key properties. We find that a two-component parametric model performs well for our simple mock galaxies, but is sensitive to model mismatches. The fixed- and flexible-bin nonparametric models included in Prospector are able to rapidly quench a major burst of star formation, but systematically underestimate the post-burst age by up to 200 Myr. We develop a custom SFH model that allows for additional flexibility in the recent SFH. Our flexible nonparametric model is able to constrain post-burst ages with no significant offset and just ~90 Myr of scatter. Our results suggest that while standard nonparametric models are able to recover first-order quantities of the SFH (mass, SFR, average age), accurately recovering higher-order quantities (burst fraction, quenching time) requires careful consideration of model flexibility. These mock recovery tests are a critical part of future SFH studies. Finally, we show that our new, public SFH model is able to accurately recover the properties of mock star-forming and quiescent galaxies and is suitable for broader use in the SED fitting community. https://github.com/bd-j/prospector

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Machine learning models of intermittent operation of RO wellhead water treatment for salinity reduction and nitrate removal

Machine learning models were developed for intermittent multi-mode operation of a wellhead reverse osmosis water purification and desalination system to predict salt passage, nitrate passage, and permeate flux. The models, based on long short-term memory (LSTM) recurrent neural network (RNN) architecture, included an attention mechanism to increase model performance in proximity of the regulatory limit for nitrate. Training and testing of the models for the Startup, Production, Shutdown and Flushing operational modes were based on operational data (consisting of 22 process variables per data sample) acquired every 2–5 s over a six-month period. The significant sets of model input attributes for the different operational modes were assessed via Spearman ranking correlation, Self-Organizing Map (SOM) analysis and feed forward feature selection (FFFS). Although the variability of nitrate passage, salt passage and permeate flux was significant over the four operational modes, prediction performance for the three outcomes were with R2 and Average Absolute Relative Error (AARE) of 0.78–0.95 and 2.96–6.16 %, respectively. Model updates post membrane elements replacement demonstrated similar levels of prediction accuracy. The study results suggest that there is merit in exploring the utility of multi-mode models for sensor fault detection, data imputation, and for potential use in model-predictive control.

Intermittent RO operation↗
DOE OSTI2026

Battery Performance and Cost Model (BatPaC) Version 6.0

SF-26-016 The Battery Performance and Cost model (BatPaC) is a calculation method based on Microsoft Excel spreadsheets that has been developed at Argonne for estimating the performance and manufacturing cost of lithium-ion batteries for electric-drive vehicles including hybrid-electrics (HEV), plug-in hybrids (PHEVs) and pure electrics. BatPaC was first developed in 2007, was subsequently peer reviewed, and it has served Argonne researchers and the greater battery community in studying the impact of material properties on performance at the pack level. BatPaC has been updated and re-released multiple times since its original public release in 2011. This current version is BatPaC 6.0, which contains additional functionality needed to handle advances in automotive batteries, like the use of lithium metal and silicon anodes and the need to accommodate cell expansion and apply high levels of pressure.

KNEHR, KEVIN [Argonne National Laboratory (ANL), A↗
DOE OSTI2026

Outdoor Deployment Data for a Four-Terminal GaAs//Si Tandem Solar Mini-Module

This dataset contains the complete outdoor measurement and analysis data for a mechanically stacked, four-terminal (4T) gallium arsenide (GaAs)//silicon (Si) tandem solar mini-module deployed from October 2019 to January 2021 at the Solar Radiation Research Laboratory (SRRL) in Golden, Colorado, USA. The data support a performance modeling and degradation analysis framework for tandem photovoltaic devices, as described in the accompanying publication. The dataset includes: (1) current–voltage (J–V) characteristics of each sub-cell measured approximately every five minutes, with extracted performance parameters; (2) spectral irradiance from an EKO MS-710 WISER spectroradiometer, along with derived spectral mismatch ratios (SMR) and average photon energy (APE); (3) one-minute resolution meteorological data from the co-located SRRL weather station and GPS-derived precipitable water vapor (PWV); (4) pre-deployment laboratory characterization (external quantum efficiency, J–V curves, standard test conditions parameters); (5) outdoor-extracted temperature and PWV correction coefficients; and (6) PVcircuit equivalent-circuit simulation outputs used for model validation. Degradation rates of −4.1 ± 0.2 %/year (GaAs) and −2.5 ± 0.9 %/year (Si) were determined using a filtering and normalization methodology adapted for fixed-tilt tandem modules. All data are provided in open, portable formats (Apache Parquet, CSV, JSON) to enable full reproducibility of the published analysis.

14 SOLAR ENERGY↗
DOE OSTI2022

Battery Performance and Cost Modeling for Electric-Drive Vehicles (A Manual for BatPaC v5.0)

This manual details the fifth version of the Battery Performance and Cost (BatPaC v5.0) model developed at Argonne National Laboratory for lithium-ion battery packs used in transportation (file “BatPaC 5.0 2022-07-22.xlsm”). BatPaC is a publicly available model that performs a bottom-up lithium-ion battery design and cost calculation. The model designs the battery for a specified power, energy, and vehicle type (i.e., hybrid, plug-in hybrid, or full-electric). The cost of the designed battery is calculated by accounting for every step in the lithium-ion battery manufacturing process. The original model and manual were publicly peer-reviewed by battery experts assembled by the U.S. Environmental Protection Agency. This revised model and manual include changes made in response to comments received from users and the observed trajectory of the industry.

25 ENERGY STORAGE↗
DOE OSTI2026

VRN3P: Variational Recurrent Neural Network Based Net-Load Prediction under High Solar Penetration

This is the final technical report for the SETO-funded VRN3P project (PNNL# 76914). The goal of this project, led by Pacific Northwest National Laboratory (PNNL), in collaboration with Lawrence Livermore National Laboratory (LLNL) and Portland General Electric (PGE), was to develop and validate a deep variational recurrent neural network-based net-load prediction (VRN3P) framework for probabilistic time-series forecasting of day-ahead net-load under high solar penetration scenarios. The project team reports successful design of a novel probabilistic net-load forecasting architecture, comprising of a variational autoencoder and a recurrent neural network, which demonstrates 30% improvement in forecast performance, 60% improvement in training time, and consumes 44% less memory, when compared with conventional baseline models. The team tested the VRN3P model performance on GridLAB-D test-cases representing varying BTM solar penetration levels of 20%, 30%, and 50%, with integrated time-series net-load profiles provided by the utility partner (PGE). The VRN3P model demonstrate <2% hourly MAPE (averaged over the year) for day- ahead net-load forecast on the test scenario with 20% BTM solar. Transfer learning extension of the VRN3P model has demonstrated 8.33× speed-up in training, while still achieving acceptable forecast performance of 2.24% hourly MAPE on the 30% BTM solar penetration test-scenario. A preliminary version of the VRN3P GridAPPS-D™has been developed, along with a web-based interactive user-interface (named ‘Forte’) which has made available on GitHub for public use.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Data and scripts associated with the manuscript "Encoding Diel Hysteresis and the Birch Effect in Dryland Soil Respiration Models through Knowledge-Guided Deep Learning"

This package contains the data and scripts used in "Encoding Diel Hysteresis and the Birch Effect in Dryland Soil Respiration Models through Knowledge-Guided Deep Learning" (Jiang et al., 2022). The data.zip file contains the flux tower and automated chamber observations used for developing the deep learning model for modeling soil respiration. The scripts.zip file contains the Jupyter notebooks and python scripts for preprocessing the data, training the deep learning models, and postprocessing the results. The src.zip contains the source code for training the deep learning model, performing mutual information analysis, and plotting functions. The trained_models.zip contains multiple folders used for hosting the trained deep-learning models and the associated soil respiration predictions. The whole process is performed using python. We include the REAMD.md to document the python package requirements.Soil respiration in dryland ecosystems is challenging to model due to its complex interactions with environmental drivers. Knowledge-guided deep learning provides a much more effective means of accurately representing these complex interactions than traditional Q10-based models. Mutual information analysis revealed that future soil temperature shares more information with soil respiration than past soil temperature, consistent with their clockwise diel hysteresis. We explicitly encoded diel hysteresis, soil drying, and soil rewetting effects on soil respiration dynamics in a newly designed Long Short Term Memory (LSTM) model. The model takes both past and future environmental drivers as inputs to predict soil respiration. The new LSTM model substantially outperformed three Q10-based models and the Community Land Model when reproducing the observed soil respiration dynamics in a semi-arid ecosystem. The new LSTM model clearly demonstrated its superiority for temporally extrapolating soil respiration dynamics, such that the resulting correlation with observational data is up to 0.7 while the correlations of both Q10-based models and the Community Land Model (CLM) are less than 0.4. Our results underscore the high potential for knowledge-guided deep learning to replace Q10-based soil respiration modules in Earth system models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Improving the accessibility and transferability of machine learning algorithms for identification of animals in camera trap images: MLWIC2

Motion-activated wildlife cameras (or “camera traps”) are frequently used to remotely and noninvasively observe animals. The vast number of images collected from camera trap projects has prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter-out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. We used 3 million camera trap images from 18 studies in 10 states across the United States of America to train two deep neural networks, one that recognizes 58 species, the “species model,” and one that determines if an image is empty or if it contains an animal, the “empty-animal model.” Our species model and empty-animal model had accuracies of 96.8% and 97.3%, respectively. Furthermore, the models performed well on some out-of-sample datasets, as the species model had 91% accuracy on species from Canada (accuracy range 36%–91% across all out-of-sample datasets) and the empty-animal model achieved an accuracy of 91%–94% on out-of-sample datasets from different continents. Our software addresses some of the limitations of using machine learning to classify images from camera traps. By including many species from several locations, our species model is potentially applicable to many camera trap studies in North America. We also found that our empty-animal model can facilitate removal of images without animals globally. We provide the trained models in an R package (MLWIC2: Machine Learning for Wildlife Image Classification in R), which contains Shiny Applications that allow scientists with minimal programming experience to use trained models and train new models in six neural network architectures with varying depths.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Improving E3SM Land Model Photosynthesis Parameterization via Satellite SIF, Machine Learning, and Surrogate Modeling

The parameterization of key photosynthesis parameters is one of the key uncertain sources in modeling ecosystem gross primary productivity (GPP). Solar-induced chlorophyll fluorescence (SIF) offers a good proxy for GPP since it marks the actual process of photosynthesis; while machine learning (ML) provides a robust approach to model the GPP-SIF relationship. Here, we trained the boosted regressing tree (BRT) and the Random Forest ML models with Greenhouse Gases Observing Satellite SIF data and in situ GPP observations from 49 eddy covariance towers. These trained ML GPP-SIF models were fed into the Energy Exascale Earth System Model (E3SM) Land Model (ELM) to generate ELM-simulated global SIF estimates, which were then benchmarked against satellite SIF observations with a surrogate modeling approach. Our results indicated good modeling performance of the ML-based GPP-SIF relationship. The ELM model when fed with the ML GPP-SIF models also can well predict the spatial-temporal variations in SIF. We also found high model accuracy for the surrogate modeling. Model parameter sensitivity analysis suggested that the fraction of leaf nitrogen in RuBisCO (flnr) is the most sensitive parameter to the SIF; other sensitive parameters include the Ball-Berry stomatal conductance slope (mbbopt) and the vcmax entropy (vcmaxse). The posterior uncertainty in simulated GPP was greatly reduced after benchmarking, and the model produced improved spatial patterns of mean GPP relative to FLUXCOM GPP. Our integrated approach provides a new avenue for improving land models and using remote-sensing SIF, which can be further improved in the future with more ground- and satellite-based observations.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

CovTransformer: A transformer model for SARS-CoV-2 lineage frequency forecasting

With hundreds of SARS-CoV-2 lineages circulating in the global population, there is an ongoing need for predicting and forecasting lineage frequencies and thus identifying rapidly expanding lineages. Accurate prediction would allow for more focused experimental efforts to understand pathogenicity of future dominating lineages and characterize the extent of their immune escape. Here, we first show that the inherent noise and biases in lineage frequency data make a commonly-used regression-based approach unreliable. To address this weakness, we constructed a machine learning model for SARS-CoV-2 lineage frequency forecasting, called CovTransformer, based on the transformer architecture. We designed our model to navigate challenges such as a limited amount of data with high levels of noise and bias. We first trained and tested the model using data from the UK and the USA, and then tested the generalization ability of the model to many other countries and US states. Remarkably, the trained model makes accurate predictions two months into the future with high levels of accuracy both globally (in 31 countries with high levels of sequencing effort) and at the US-state level. Our model performed substantially better than a widely used forecasting tool, the multinomial regression model implemented in Nextstrain, demonstrating its utility in SARS-CoV-2 monitoring. Assuming a newly emerged lineage is identified and assigned, our test using retrospective data shows that our model is able to identify the dominating lineages 7 weeks in advance on average before they became dominant. Overall, our work demonstrates that transformer models represent a promising approach for SARS-CoV-2 forecasting and pandemic monitoring.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Importance of different parameterization changes for the updated dust cycle modeling in the Community Atmosphere Model (version 6.1)

Abstract. The Community Atmosphere Model (CAM6.1), the atmospheric component of the Community Earth System Model (CESM; version 2.1), simulates the life cycle (emission, transport, and deposition) of mineral dust and its interactions with physio-chemical components to quantify the impacts of dust on climate and the Earth system. The accuracy of such quantifications relies on how well dust-related processes are represented in the model. Here we update the parameterizations for the dust module, including those on the dust emission scheme, the aerosol dry deposition scheme, the size distribution of transported dust, and the treatment of dust particle shape. Multiple simulations were undertaken to evaluate the model performance against diverse observations, and to understand how each update alters the modeled dust cycle and the simulated dust direct radiative effect. The model–observation comparisons suggest that substantially improved model representations of the dust cycle are achieved primarily through the new more physically-based dust emission scheme. In comparison, the other modifications induced small changes to the modeled dust cycle and model–observation comparisons, except the size distribution of dust in the coarse mode, which can be even more influential than that of replacing the dust emission scheme. We highlight which changes introduced here are important for which regions, shedding light on further dust model developments required for more accurately estimating interactions between dust and climate.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Measurement of Wind Loading on Heliostats at the Crescent Dunes Power Plant: An Overview

The cost of solar collectors constitutes almost one third of the total cost of a CSP plant. One of the ongoing challenges in the design of these collectors is wind loading on mirrors, support structures, and drives. A particular challenge is dynamic wind loading, caused by the turbulent wind flow. To date, the design of solar collector structures has relied on wind tunnel experiments and numerical simulations that do not entirely capture the dynamic effects observed at scale. The CSP industry has shown increased interest in validating the idealized assumptions with measurements obtained in operational settings to improve wind load assumptions and increase reliability and cost-efficiency of the collector design. Further, performance models need realistic assumptions about wind loading and its impact on optical performance. In a parabolic trough field campaign, NREL successfully collected a wealth of long-term, high-resolution wind and loads data [2], that are publicly available and that can be used for the above-mentioned purposes. Heliostats are impacted differently by wind than parabolic troughs, due to their different shape, size, and field layout. To study the impact of wind and turbulence on heliostats, we initiated another field campaign in an operational power- tower plant, Crescent Dunes, in Nevada, USA. In this work, we present an overview of the measurements, first results, and potential implications of wind driven loads on Heliostats.

concentrated solar power↗
DOE OSTI2020

Evaluating three evapotranspiration estimates from model of different complexity over China using the ILAMB benchmarking system

The land surface models range in complexity of terrestrial evapotranspiration, yet it is unknown how model complexity translates to accuracy of modeled evapotranspiration estimates. Here, we use the International Land Model Benchmarking system to assess ET estimates from three models of varying complexity driven by the same forcing datasets: an earth system model, a terrestrial biosphere model, and a stand-alone ET model. The performance assessment includes both temporal and spatial evaluation, and different plant functional types across China. Our results indicate that the most complex model, an earth system model, performed best against the benchmarking datasets and metrics. Terrestrial biosphere model performed best in simulating inter-annual variability of ET, while earth system model performed best in simulating the seasonal cycle. The more complex models (earth system model and terrestrial biosphere model) perform better in forest, shrub and crop ecosystems, while the simpler model (stand-alone ET model) perform better in grass ecosystems. Our study demonstrates the impact of model complexity on ET estimates and highlights directions for future ET model improvements.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Dry in-pile fracture test (DRIFT) for separate-effects validation of ceramic fuel fracture models

Fracture is an important component of nuclear fuel behavior, and significant efforts have been invested into developing fuel performance models that are capable of accurately representing fracture. Usable data on the process of fracture propagation in nuclear fuel under realistic service conditions are very limited. To address this need, a series of separate-effects experiments were developed and performed at Idaho National Laboratory's Transient Reactor Test (TREAT) facility. These experiments employ a heat sink to radially remove heat from the fuel in a manner that approximates the effect of coolant in an operating light-water reactor (LWR). The test holder for these experiments is known as the Dry In-pile Fracture Test (DRIFT). A series of experiments employing DRIFT and TREAT were performed to provide data on the extent and nature of fracture in fresh fuel at various points during a ramp to full power. Novel aspects of these experiments include the way they employ a heat sink to replicate steady-state LWR conditions, as well as the use of fiber optic sensors for in-reactor thermal instrumentation. Details on the development of this experiment, experimental conditions, and resulting data (including in situ thermal measurements and post-irradiation imaging of fracture) are provided in this work. LWR-equivalent powers ranging from 10 to 25 kW/m were tested using this apparatus. Cracking was visible at all power levels, with increasing cracking extent as the power level increased, although there was little difference in the cracking between the two highest-power tests, which had LWR-equivalent powers of 20 and 25 kW/m.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Deep-learning-based domain adaptation for cavity fault prediction at Jefferson Laboratory

Superconducting radio-frequency (SRF) cavities are the core components of the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab, providing high-power electron beams for nuclear physics experiments. The facility comprises 418 SRF cavities, and any fault in these cavities can lead to interruptions in the electron beam supply. Cavity faults are the leading cause of beam trips in CEBAF. Predicting and mitigating those faults before onset can help maintain normal operation. Existing models face challenges in distinguishing between normal and fault signals when changes occur in the underlying time-series data, from changes in control software, operational parameters, or the environment. This work proposes a deep learning domain adaptation model that leverages transfer learning to address fault prediction challenges by improving accuracy. The model is trained and fine-tuned using a dataset collected for faulty and normal operation using a data acquisition system in CEBAF. Our deep learning-based domain adaptation model achieves a prediction accuracy of 89.61% of the fault and normal signals. The developed model effectively predicts normal running signals compared to the baseline approach without domain adaptation. This capacity is essential for the fault prediction task in the CEBAF because of heavily imbalanced data containing vast amounts of normal signals. The model performs well for predicting faults several hundred milliseconds before the fault onset compared to other models where no adaptation is applied. Incorporating deep learning-based domain adaptation techniques will significantly improve the fault prediction performance.

Rahman, Md Monibor [Old Dominion Univ., Norfolk, V↗
DOE OSTI2021

PreMevE Update: Forecasting Ultra-Relativistic Electrons Inside Earth's Outer Radiation Belt

Energetic electrons inside Earth's Van Allen belts pose a major radiation threat to spaceborne electronics that often play vital roles in modern society. Ultra-relativistic electrons with energies greater than or equal to two megaelectron-volt (MeV) are of particular interest, and thus forecasting these ≥2 MeV electrons has a significant meaning to all space sectors. Here, we update the latest development of the predictive model for MeV electrons in the outer radiation belt. The new version, called PREdictive MEV Electron (PreMevE)-2E, forecasts ultra-relativistic electron flux distributions across the outer belt, with no need for in situ measurements of the trapped MeV electron population except at the geosynchronous orbit (GEO). Model inputs include precipitating electrons observed in low-Earthorbits by NOAA satellites, upstream solar wind speeds and densities from solar wind monitors, as well as ultra-relativistic electrons measured by one Los Alamos GEO satellite. We evaluated 32 supervised machine learning models that fall into four different classes of linear and neural network architectures, and successfully tested ensemble forecasting by using groups of top-performing models. All models are individually trained, validated, and tested by in situ electron data from NASA's Van Allen Probes mission. It is shown that the final ensemble model outperforms individual models at most L-shells, and this PreMevE-2E model can provide 25-h (~1-day) and 50-h (~2-day) forecasts with high mean performance efficiency and correlation values. Our results also suggest that this new model is dominated by nonlinear components at L-shells <~4 for ultra-relativistic electrons, different from the dominance of linear components for 1 MeV electrons as previously discovered.

79 ASTRONOMY AND ASTROPHYSICS↗