Search NASA⌕ Search

SEARCH · Search NASA

Results for “Short-term prediction”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

$\mathrm{SageNet}$: Fast Neural Network Emulation of the Stiff-amplified Gravitational Waves from Inflation

Accurate modeling of the inflationary gravitational waves (GWs) requires time-consuming, iterative numerical integrations of differential equations to take into account their backreaction on the expansion history. To improve computational efficiency while preserving accuracy, we present the Stiff-amplified Gravitational-wave Emulator Network (SageNet), a deep learning framework designed to replace conventional numerical solvers (code available at https://github.com/YifangLuo/SageNet). SageNet employs a long short-term memory architecture to emulate the present-day energy density spectrum of the inflationary GWs with possible stiff amplification, Ω GW (f). Trained on a data set of 25,689 numerically generated solutions, SageNet allows accurate reconstructions of Ω GW (f) and generalizes well to a wide range of cosmological parameters; 90.9% of the test emulations with randomly distributed parameters exhibit errors of under 4%. In addition, SageNet demonstrates its ability to learn and reproduce the artificial, adaptive sampling patterns in numerical calculations, which implement denser sampling of frequencies around changes in spectral indices in Ω GW (f). The dual capability of learning both physical and artificial features of the numerical GW spectra establishes SageNet as a robust alternative to exact numerical methods. Finally, our benchmark tests show that SageNet reduces the computation time from tens of seconds to milliseconds, achieving a speedup of ∼10 4 times over standard CPU-based numerical solvers with the potential for further acceleration on GPU hardware. These capabilities make SageNet a powerful tool for accelerating Bayesian inference procedures for extended cosmological models. In a broad sense, the SageNet framework offers a fast, accurate, and generalizable solution to modeling cosmological observables whose theoretical predictions demand costly differential equation solvers.

Astronomy data modeling↗
DOE OSTI2024

Automated Framework for Groundwater Monitoring Using DWT with LSTM and Transformers

Environmental monitoring is critical for safeguarding public health and ecological well-being. Traditional data structuring and workflow monitoring methods consume significant time and effort, hindering timely insights and effective decision-making. Our study addresses this challenge by presenting an AI framework that automates data cleaning, structuring, and modeling processes, specifically targeting applications in groundwater monitoring. By leveraging automation for data processing and model training, our framework establishes a novel and efficient paradigm for environmental monitoring, with its potential application to the vast network of over a hundred Department of Energy Environmental Management (DoE-EM) cleanup sites across the country. It analyzes data streams from a network of groundwater Internet-of-Things (IoT) sensors deployed at the Savannah River Site (SRS) for prediction modeling. This allows human experts to focus on analysis and decision-making, ultimately leading to better environmental outcomes.The framework employs multivariate time-series forecasting methods to study and model the behavior of varying chemical analytes. The continuous learning process is enabled by utilizing deep learning techniques. It allows the framework to become more nuanced in its analysis over time, adapting to the specific characteristics of the environmental site and the evolving nature of contaminant behavior. Deep learning models known for sequence modeling, LSTM, and Transformers are employed for time series forecasting. Data processing and structuring are essential components significantly impacting the final model's performance. This hypothesis was proven by presenting a comparative analysis of model performance with processed and unprocessed data. The feature engineering approach utilized was the Discrete Wavelet Transform, which works well with time series data.

Discrete Wavelet Transform (DWT)↗
DOE OSTI2025

A cross-dimensional analysis of data-driven short-term load forecasting methods with large-scale smart meter data

Electricity load forecasting is essential to utility operation and power grid stability. A wide spectrum of data-driven methods, ranging from linear regression models to more recent deep learning models have been adopted to forecast electric load over the years. However, there still lacks a holistic evaluation of the applicability of conventional statistical and machine learning based algorithms with respect to different temporal and spatial scopes, computational requirements, and sensitivity of model-tuning. Enabled by a large-scale electricity load profile dataset of over 40,000 residential customers in a utility region, we conducted a cross-dimensional analysis of data-driven load forecasting methods. Three regression-based and seven deep learning algorithms with different model configurations were evaluated in terms of their overall and peak load prediction accuracy, and training burdens, across spatial aggregation levels ranging from the transformer, feeder, substation, to neighborhood. We found, first, the load forecasting accuracy is constrained by a predictability boundary, influenced by the forecasting horizon and spatial aggregation level. Specifically, RandomForest, XGBoost, TFT, TSMixer, and TiDE models achieved less than 10 % prediction error for up to 96-h ahead forecasting for district, substation, and feeder levels, while other models struggle at long-horizon predictions; Second, for winter and summer peak load dates, most models were able to predict the peak demand timing within ± 1 h, but the prediction percentage error varied by models, with TFT and TiDE models being the top performers; Third, models with similar prediction accuracy can differ in training burden by an order of magnitude. Therefore, choosing model configurations that balance prediction performance and computational resource is an important practical consideration for large-scale deployment of the machine learning based load forecasting. The outcome of this study can guide researchers and practitioners to choose the proper load forecasting algorithms based on their problem scope, required accuracy, and available resources. The predictability boundary can serve as a benchmark for electricity load forecasting problems with new algorithms and datasets.

Li, Han↗
DOE OSTI2026

Physics vs structure: A systematic benchmark of learning strategies for multi-zone building thermal dynamics

Recent advances in physics-informed and data-driven machine learning promise improved thermal models for advanced building control, yet there is limited quantitative evidence on when added physics structure and architectural complexity are beneficial. Here, this work presents a systematic benchmark of five representative system identification methods for modeling multi-zone building thermal dynamics: linear state-space models, multi-layer perceptrons, neural state-space models, neural ordinary differential equations, and physically-consistent neural networks. The methods are evaluated across multiple data regimes and zone coupling strategies. Using a high-fidelity multi-zone commercial building emulator, we examine short-term and long-term prediction accuracy, computational efficiency, and ease of development. Our results reveal critical trade-offs between prediction performance, model complexity, and physical consistency. We demonstrate that decoupled, nonlinear black-box models consistently outperform coupled physics-constrained architectures in both predictive accuracy and out-of-distribution robustness in majority of the test cases for the building type considered in the study. Our findings quantify the cost of complexity in building thermal modeling and provide concrete, actionable, scenario-based guidelines for selecting model classes for control-oriented applications.

Building thermal modeling↗
DOE OSTI2024

eCoral: How Electrolysis Could Restore Seawater Conditions Ideal for Coral Reefs

Coral reefs suffer from climate change, including long-term ocean acidification (OA) and warming and short-term bleaching, tropical storms, and pollution events, all of which are increasing in frequency and severity. It is urgent yet unclear how to intervene to save coral reefs. Reversal of the ocean pH to preindustrial levels could restore coral reefs to their preindustrial growth rates; however, strategies to reverse OA on environmentally relevant scales have not been established. Anecdotally, electrolysis seems to help coral reefs recover from acidification and short-term events, but few uncontrolled studies support such claims. Here, using two independent continuum simulation approaches (COMSOL and CrunchFlow), we show the effect of electrolysis on seawater chemistry relevant to coral reef survival and growth. We conclude that near the negative electrodes, the cathodes, seawater pH, supersaturation, and carbonate concentration all increase significantly. Electrolysis of seawater, therefore, can be used to restore preindustrial ocean conditions locally to save coral reefs, an approach termed eCoral here. We anticipate these simulation results to be the starting point for controlled experiments to test whether seawater electrolysis promotes coral reef growth and restoration, as these simulations predict.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Weak-form inference for hybrid dynamical systems in ecology

Species subject to predation and environmental threats commonly exhibit variable periods of population boom and bust over long timescales. Understanding and predicting such behaviour, especially given the inherent heterogeneity and stochasticity of exogenous driving factors over short timescales, is an ongoing challenge. A modelling paradigm gaining popularity in the ecological sciences for such multi-scale effects is to couple short-term continuous dynamics to long-term discrete updates. We develop a data-driven method utilizing weak-form equation learning to extract such hybrid governing equations for population dynamics and to estimate the requisite parameters using sparse intermittent measurements of the discrete and continuous variables. The method produces a set of short-term continuous dynamical system equations parametrized by long-term variables, and long-term discrete equations parametrized by short-term variables, allowing direct assessment of interdependencies between the two timescales. We demonstrate the utility of the method on a variety of ecological scenarios and provide extensive tests using models previously derived for epizootics experienced by the North American spongy moth ( Lymantria dispar dispar ).

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Efficient Simulation of Cascading Outages Using an Energy Function-Embedded Quasi-Steady-State Model

Here, this paper proposed an energy function-embedded quasi-steady-state model for efficient simulation of cascading outages on a power grid while addressing transient stability concerns. Compared to quasi-steady-state models, the proposed model incorporates short-term dynamic simulation and an energy function method to efficiently evaluate the transient stability of a power grid together with outage propagation without transient stability simulation. Cascading outage simulation using the proposed model conducts three steps for each disturbance such as a line outage. First, it performs time-domain simulation for a short term to obtain a post-disturbance trajectory. Second, along the trajectory, the system state with the local maximum potential energy is found and used as the initial point to search for a relevant unstable equilibrium by Newton's method. Third, the transient energy margin is estimated based on this unstable equilibrium to predict an out-of-step condition with generators. The proposed energy function-embedded quasi-steady-state model is tested in terms of its accuracy and time performance on an NPCC 140-bus power system and compared to a quasi-steady-state model embedding transient stability simulation.

Guo, Zhenping [Univ. of Tennessee, Knoxville, TN (↗
DOE OSTI2025

S AP F LOWER : an automated tool for sap flow data preprocessing, gap-filling, and analysis using deep learning

Sap flow, a critical process in plant water use and ecosystem water cycles, is often measured using thermal dissipation probes (TDP) due to their ease of installation and continuous data collection. However, sap flow data frequently include noise, outliers, and gaps, creating challenges for analysis and requiring substantial manual processing. We developed S AP F LOWER , a tool that automates data preprocessing, model training, gap-filling, sapwood area scaling and modeling, and water use analysis. It integrates autocleaning, machine learning and deep learning models (e.g. random forest, Gaussian process regression, long short-term memory (LSTM), bidirectional LSTM (BiLSTM)), and efficient workflows to process sap flow data. S AP F LOWER can remove over 90% of noisy data while preserving legitimate variations and achieve high accuracy in gap-filling based on user-determined parameters. Random forest, LSTM, and BiLSTM models reduced root mean square error to 10% or less for long-term gaps. Model training and prediction can be performed efficiently within seconds. S AP F LOWER significantly enhances the efficiency and accessibility of TDP data analysis by automating complex tasks, enabling researchers without programming expertise to employ advanced techniques. Future improvements will focus on species-specific corrections for TDP and support for additional measurement methods. S AP F LOWER is openly available on GitHub (https://github.com/JiaxinWang123/SapFlower) and Zenodo (doi: 10.5281/zenodo.13665919).

ecosystem water balance↗
DOE OSTI2025

A Variational Autoencoder Model Toward Molecular Structure Representation Learning of Fuels

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

fuel design↗
DOE OSTI2025

Convective potential and fuel availability complement near-surface weather in regulating global wildfire activity

Wildfires are favored by hot, dry, windy, rainless conditions—this knowledge about fire weather informs both short-term forecast and long-term prediction of wildfire activity. Yet, wildfires rely on the availability of ignition and fuel, which are underrepresented in fire forecast and prediction practices. By analyzing satellite measurements and atmospheric reanalysis, here we show that near-surface weather only partially captures wildfire occurrence and intensity across the daily to seasonal timescales. Beyond near-surface weather, convection and fuel abundance play a complementary role in regulating burning processes. Specifically, enhanced atmospheric convection is identified for over 40% of the low-human-impact regions and 61% of global burnable areas during wildfire ignition and spreading periods. Meanwhile, 56% of shrublands and 54% of grasslands see higher fuel load with actual occurrence of fire. Our results highlight the role of convection and fuel in wildfire forecast, prompting a revisit of wildfire prediction under intertwined atmospheric and terrestrial changes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Microbial community data from throughfall exclusion experiment: Metadata, SI, community composition, LefSe, and FunGuilR data tables from PARCHED Panama tropical forest soils, 2024-2025

Soil contains more carbon (C) than terrestrial vegetation and the atmosphere combined, with some of the largest terrestrial C stocks in tropical rainforests. Soil microbes decompose organic matter, playing a vital role in the storage or loss of soil C. With climate change, drought conditions are predicted to increase in many tropical regions, including both chronic drying and extended drought, potentially influencing these processes. This project explored the effects of chronic and seasonal drying on soil microbial communities across four distinct tropical forests in a long-term drying experiment. We investigated the effects of a chronic drying manipulation on soil microbial community abundance and variation across different forests and seasons. We also compared findings with previously published data from these forests after short-term drying. This project used soils from a long-term drying experiment established in 2018 across four seasonal lowland forests in Panama. Soils were collected from 0 – 10 cm depths during three seasonal periods in control and drying plots in 2024 and 2025 from a total of 32 plots (n = 4 per forest per treatment). The forests varied in baseline rainfall and soil fertility. We calculated alpha and beta diversity indices and compared taxonomic community composition. We found significant biogeographic variation in microbial diversity and taxonomy, with significant differences across the forests and significant effects of the drying treatment. Metadata and sample IDs are within Metadata_16S.csv and Metadata_ITS.csv. Relative abundance tables of every sample at every season are shown in the Excel workbooks 16S Relative Abundance.xlsx and ITS Relative Abundance.xlsx. They are then also shown in CSV files by each taxonomic level. Linear discriminant analysis effect size (LefSe) tables are shown for the full 16S and ITS datasets (n = 96), subsets for every site at every season (n = 8), and then for the forests with each plot merged by season (n = 8). FunGuildR data table of ITS data is uploaded.

Bacteria↗
NASA NTRS1973

Utilization of remotely-sensed data in the management of inland wetlands

The author has identified the following significant results. ERTS-1 data and aerial photography are proving to be a useful tool for the inventory and management of inland wetlands. Two examples of the application of remotely-sensed data to specific wetland management needs or requirements are discussed. Studies of the Great Dismal Swamp are utilizing ERTS-1 imagery and color IR photography in: (1) study area selection; (2) field inspection; (3) vegetation mapping; (4) identification of drainage characteristics and moisture regime; (5) location of intensive study areas; and (6) detection of change. Thematic extractions of ERTS-1 data made using the United States Geological Survey's Autographic Theme Extraction System are aiding analyses of swamp hydrologic regime and providing information pertinent to quick recognition and inventory of wetlands from ERTS-1. DCP'S in south Florida wetlands provide near-real time data for water resources managers. Data relayed by satellite can be entered into models to provide predictive data and water storage information for long-term and short-term decision making.

Carter, V.↗
NASA NTRS1981

Optical monitoring of 15 quasars

The data of fifteen quasars are analyzed to determine how well a short-term program of observation can characterize the general optical behavior as compared to a long-term program. Three major research objectives were pursued, namely, to measure archival brightness of the quasars, to compare the brightness observed (mid-1970) to that of the Harvard historical plate collection 1900-1950, and to compare the general behavior during a short-time (two years) to that found over 50 years. Results show that in eight of the 15 quasars the mean magnitude derived from the present short-epoch program differed by 0.11 mag or less from the mean determined in the long-epoch program. The mean difference for all 15 quasars was 0.20 mag. However, in two cases, the short program failed by 0.83 and 0.66 mag, respectively, in predicting the mean brightness. The largest rate of increase in brightness observed was 0.9 mag/day (3C263) and of decrease was 0.44 (3C345), and it is concluded that the rates of decrease and increase in brightness are essentially the same.

Angione, R. J.↗
NASA NTRS1985

Accelerated viscoelastic characterization of T300-5208 graphite-epoxy laminates

A viscoelastic response scheme for the accelerated characterization of polymer-based composite laminates in applied to T300/5208 graphite/epoxy. The response of uni-directional specimens is modeled. The transient component of the viscoelastic creep compliance is assumed to follow a power law approximation. A recursive relationship is developed, based upon the Schapery single-integral equation, which allows approximation of a continuous time-varying uniaxial load using discrete steps in stress. The viscoelastic response of T300/5208 to transverse normal and shear stresses is determined unsing 90 deg and 10 deg off-axis tensile specimens. In each case the seven viscoelastic material parameters required in the analysis are determined experimentally using short-term creep and creep recovery tests. It is shown that an accurate measure of the power law exponent is crucial for accurate long-term prediction. A short term test cycle selection procedure is proposed, which should provide useful guidelines for the evaluation of other viscoelastic materials.

Tuttle, M. E.↗
NASA NTRS1987

Solar ultraviolet radiation induced variations in the stratosphere and mesosphere

The detectability and interpretation of short-term solar UV induced responses of middle atmospheric ozone, temperature, and dynamics are reviewed. The detectability of solar UV induced perturbations in the middle atmosphere is studied in terms of seasonal and endogenic dynamical variations. The interpretation of low-latitude ozone and possible temperature responses on the solar rotation time scale is examined. The use of these data to constrain or test photochemical model predictions is discussed.

Hood, L. L.↗
NASA NTRS1993

Viscoelastic characterization of thin-film polymers exposed to low Earth orbit

The materials made available through the Long Duration Exposure Facility (LDEF) satellite provide a set of specimens that can be well characterized and have a known exposure history with reference to atomic oxygen and ultraviolet radiation exposure. Mechanical characteristics measured from control samples and exposed samples provide a data base for predicting the behavior of polymers in low earth orbit. Samples of 1.0 mil thick low density polyethylene were exposed to the low earth orbit environment for a period of six years. These materials were not directly exposed to ram atomic oxygen and offer a unique opportunity for measuring the effect of atomic oxygen and UV radiation on mechanical properties with little concern to the effect of erosion. The viscoelastic characteristics of these materials were measured and compared to the viscoelastic characteristics of control samples. To aid in differentiating the effects of changes in crystallinity resulting from thermal cycling, from the effects of changes in chemical structure resulting from atomic oxygen/UV attack to the polymer, a second set of control specimens, annealed to increase crystallinity, were measured as well. The resulting characterization of these materials will offer insight into the impact of atomic oxygen/UV on the mechanical properties of polymeric materials. The viscoelastic properties measured for the control, annealed, and exposed specimens were the storage and loss modulus as a function of frequency and temperature. From these datum is calculated the viscoelastic master curve derived using the principle of time/temperature superposition. Using the master curve, the relaxation modulus is calculated using the method of Ninomiya and Ferry. The viscoelastic master curve and the stress relaxation modulus provide a direct measure of the changes in the chemical or morphological structure. In addition, the effect of these changes on long-term and short-term mechanical properties is known directly. It should be noted that the dependence on directionality for the polymer films was considered since these films were manufactured by a blown-film process.

Letton, Alan↗
NASA NTRS1986

Resilience: Concepts and Measures

Inertia, the resistance of an ecosystem property to change under stress, is distinguished from resilience, which refers to the degree, manner. and pace of change or recovery in ecosystem properties following disturbance. In turn, these two terms are differentiated from 'stability'. which is used here to refer to the pattern of natural fluctuation in ecosystem properties in the absence of major exogenous disturbance. Four component attributes of resilience are reviewed in the context of Mediterranean-climate examples. The elasticity component concerns the rate of recovery of an ecosystem property following disturbance; amplitude, the threshold of stress beyond which recovery to the initial state does not occur; hysteresis, the degree to which the pattern of recovery after stress differs from that of deterioration under chronic stress, and malleability the ease with which the ecosystem can become permanently altered. Each ecosystem property will typically reveal a different level of resilience to a given stress and stressor. The degree of recovery should not be expected to be complete in any event, due to sample variability and stochastic events. In cyclicallystable ecosystems, the pattern of recovery should be measured in light of this periodicity, and short-term (within-cycle) recovery distinguished from long-term (between-cycle) recovery. The prediction of resilience properties of ecosystems can be approached through a knowledge of the modular structure of foodwebs, through knowledge of the autecological adaptations of key species to the stressor, or through cumulative experience of the response to disturbance at the community level. At present there is much room for investigation of each of these approaches in Mediterranean-climate ecosystems.

Westman, Walter E.↗
NASA NTRS1993

Conventional heart rate variability analysis of ambulatory electrocardiographic recordings fails to predict imminent ventricular fibrillation

OBJECTIVES. The purpose of this report was to study heart rate variability in Holter recordings of patients who experienced ventricular fibrillation during the recording. BACKGROUND. Decreased heart rate variability is recognized as a long-term predictor of overall and arrhythmic death after myocardial infarction. It was therefore postulated that heart rate variability would be lowest when measured immediately before ventricular fibrillation. METHODS. Conventional indexes of heart rate variability were calculated from Holter recordings of 24 patients with structural heart disease who had ventricular fibrillation during monitoring. The control group consisted of 19 patients with coronary artery disease, of comparable age and left ventricular ejection fraction, who had nonsustained ventricular tachycardia but no ventricular fibrillation. RESULTS. Heart rate variability did not differ between the two groups, and no consistent trends in heart rate variability were observed before ventricular fibrillation occurred. CONCLUSIONS. Although conventional heart rate variability is an independent long-term predictor of adverse outcome after myocardial infarction, its clinical utility as a short-term predictor of life-threatening arrhythmias remains to be elucidated.

Non-NASA Center↗