Search NASA⌕ Search

SEARCH · Search NASA

Results for “GRU”

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.

33 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Effects of input gradient regularization on neural networks time-series forecasting of thermal power systems

This study proposes using neural networks, specifically gated recurrent unit (GRU), long-short-term memory (LSTM), and transformer networks, to improve control strategies in a 450 MW coal-fired power plant. However, neural networks face issues of becoming overly dependent on just a few variables to make predictions, which negatively impacts control decisions that rely on the model to determine the value of all manipulated variables. The paper introduces regularization techniques, including noise injection and input gradient regularization, during the training phase. Here, the work presents novel contributions in adapting neural networks to control industrial systems and applying regularization techniques from computer vision to industrial process control. Results demonstrate the effectiveness of input gradient regularization in reducing model dependence on subsets of variables, emphasizing the balance between fidelity and controllability. Further exploration is recommended, including the development of recurrent transformers, closed-loop control testing, and a sensitivity analysis on computer models to provide further insight.

20 FOSSIL-FUELED POWER PLANTS↗
DOE OSTI2022

Time series anomaly detection in power electronics signals with recurrent and ConvLSTM autoencoders

The anomalies in the high voltage converter modulator (HVCM) remain a major down time for the spallation neutron source facility, that delivers the most intense neutron beam in the world for scientific materials research. In this work, we propose neural network architectures based on Recurrent AutoEncoders (RAE) to detect anomalies ahead of time in the power signals coming from the HVCM. Bi-directional gated recurrent unit, bi-directional long-short term memory (LSTM), and convolutional LSTM (ConvLSTM) are developed, trained, and tested using real experimental signals from the HVCM module. The results show a good performance of the proposed RAE models, achieving precision up to 91%, recall up to 88%, false omission rate as low as 20% (i.e. 80% of the anomalies were detected), and area under the ROC curve up to 0.9. The three RAE models provide very comparable performance, with LSTM showing slightly better performance than GRU and ConvLSTM. The RAE models are benchmarked against other anomaly detection methods, including isolation forest, support vector machine, local outlier factor, feedforward and convolutional autoencoders, and others; showing a better performance. Here, the results of this study demonstrate the promising potential of RAE in anomaly detection for real-world power systems, and for increasing the reliability of the HVCM modules in the spallation neutron source.

42 ENGINEERING↗
DOE OSTI2026

Predicting river turbidity in Pine Island Bayou using machine learning techniques coupled with variational mode decomposition

Elevated turbidity levels pose significant public health risks by facilitating the transport of harmful pollutants, including metals, organic compounds, and pathogenic microorganisms into the surface water. These conditions create serious challenges for public recreational water use and drinking water treatment, leading to economic losses and health risks. This study utilizes water monitoring data in Pine Island Bayou, Texas, and develops a Sequence-to-Sequence (S2S) model to predict turbidity using Attention-based Gated Recurrent Units with Encoder-Decoder (AT-GRU-ED) and Long Short-Term Memory (LSTM), coupled with Variational Mode Decomposition (VMD). Compared to the model without VMD, the model demonstrates satisfactory 72-hour turbidity prediction performance, achieving MAEs of 2.60 and 3.29 NTU (reductions of 53% and 58%), RMSEs of 21.08 and 31.49 NTU (reductions of 82% and 80%), and R² values of 0.96 and 0.84 on the validation and test sets, respectively. Feature importance analysis reveals that water temperature is the dominant factor influencing seasonal turbidity patterns, while real-time hourly rainfall significantly contributes to short-term variability. Turbidity typically peaks within 48 hours after rainfall events due to lagged effects from surface runoff and upstream flow. Findings suggest suspending recreational water use and water supply pumping for three days after heavy rainfall can benefit public health and improve water treatment processes. Discharges above 100 m3/s are found to accelerate sediment dilution and transport, reducing turbidity levels more quickly after the peak. In conclusion, the proposed model demonstrates reliable 72-hour turbidity prediction, supporting decision-making for water treatment plant operations and providing early warning for public recreational water use.

Deep learning↗
DOE OSTI2023

High-Pressure Structural and Thermodynamic Properties of Cerium Orthosilicates (CeSiO 4 )

Pressure-induced phase transitions from the zircon structure-type (I4 1 /amd) to the scheelite structure type (I4 1 /a) are known for many ternary oxides systems (ABO 4 ). In this work, we present the first high-pressure study on synthetic stetindite (CeSiO 4 ) by a combination of in situ high-pressure synchrotron powder X-ray diffraction up to 36 GPa, implemented with and without dual sided laser heating, and in situ high-pressure Raman spectroscopy up to 43 GPa. Two phase transitions were identified: zircon to a high-pressure low-symmetry (HPLS) phase at 15 GPa and then to a scheelite at 18 GPa. The latter from HPLS scheelite phase was found irreversible; i.e., scheelite is fully quenchable at ambient conditions, as in other zircon-type phases. The bulk moduli (K 0 ) of stetindite, HPLS, and high-pressure scheelite phases were determined, respectively, as 171(5), 105(4), and 221(40) GPa by fitting to a second-order Birch-Murnaghan equation of state. The pressure derivatives of vibrational modes and Gru''neisen parameters of the zircon-structured polymorph are similar to those of other orthosilicate minerals. In conclusion, due to the larger ionic radii of Ce 4+ , with respect to Zr 4+ , stetindite was found to possess a softer bulk modulus and undergo the phase transitions at a lower pressure than zircon (ZrSiO 4 ), such observations are consistent with what were found in coffinite (USiO 4 ).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

APSO-enhanced algebraic derivative estimation approach for real-time traffic flow prediction on critical road sections during wildfire evacuation

In rapid-onset disaster scenarios such as wildfires, evacuation traffic often significantly deviates from historical patterns, rendering conventional data-driven forecasting methods less effective. To address this challenge, we propose an improved algebraic derivative estimation (ADE) incorporating particle swarm optimization (PSO) for real-time traffic flow prediction. Our approach dynamically adjusts the ADE prediction time window at each step by minimizing a cost function based on the mean and variance of accumulated forecasting errors within the window, thereby balancing bias and variability. We evaluate the method using traffic data from the January 2025 California wildfires, focusing on key road segments critical for large-scale evacuations. The results demonstrate that our approach surpasses established machine learning and deep learning models—XGBoost, LSTM, and GRU—in predictive accuracy and maintains high computational efficiency. Notably, the proposed method eliminates the need for offline model training. Moreover, rapid PSO-based tuning enables real-time deployment, which provides a crucial advantage in scenarios where evacuation timings and road closures change dynamically. In conclusion, these findings highlight the benefits of the PSO-enhanced ADE framework for emergency traffic management, where rapid, data-sparse forecasts are essential for effective evacuation planning.

Algebraic derivative estimation↗
DOE OSTI2025

GRUMDN: A Multi-Task Model for Predicting Human Patterns-of-Life from Stay Transition Data

Understanding human patterns-of-life (PoL) is essential towards ensuring safe and secure indoor facility environment as well as outdoor urban environment. Prediction of human movement in between places of interest is vital in understanding human PoL. Movement between spaces maybe represented and detected in one of the two forms: 1) trajectories: locations measured at regular time intervals by mobile sensors, bluetooth or GPS sensors; or 2) stay transitions: semantic PoI (points of interest) and stay duration data measurable by eventbased sensors that collect data when a check-in or check-out event is detected. Stay transition data provides a more compressed data format compared to trajectories data, especially in situations with longer stay durations, while preserving the information necessary for PoL analysis. Now as introduced briefly in the paper, our deployed end application (Digital Twin of a facility with non-player characters, besides the interactive user in virtual reality) needed a well-performing and validated AI/ML model for simulating high quality stay transitions behavior. In this study we thus primarily present our findings with developing and validating that model, which is a multi-task neural network for stay transition prediction. The neural network consists of two heads, for corresponding two tasks of stay category prediction and stay duration prediction. We evaluated gated recurrent units and multi-layer perceptrons of varying network sizes for stay category prediction; while mixture density networks, noisy generator-only networks, and generative adversarial networks of varying network sizes for stay duration prediction. We have then evaluated four multi-task models, constructed by combining these specialized models, on their ability to predict stay transition data. We tested our models on datasets from two different cases: 1) a simulation-generated dataset of indoor movement within the HFIR (high flux isotope reactor) nuclear reactor facility at Oak Ridge National Laboratory (ORNL); and 2) the GeoLife human mobility dataset of outdoor urban movement available in literature. Our results indicate that GRUMDN, which combines gated recurrent units (GRU) for stay category prediction task, and mixture density networks (MDN) for stay duration prediction task, did overall outperform other multitask models and the current state-of-the-art.

Gunaratne, Chathika [ORNL] (ORCID:0000000225088745↗
DOE OSTI2024

Forward variable selection enables fast and accurate dynamic system identification with Karhunen-Loève decomposed Gaussian processes

A promising approach for scalable Gaussian processes (GPs) is the Karhunen-Loève (KL) decomposition, in which the GP kernel is represented by a set of basis functions which are the eigenfunctions of the kernel operator. Such decomposed kernels have the potential to be very fast, and do not depend on the selection of a reduced set of inducing points. However KL decompositions lead to high dimensionality, and variable selection thus becomes paramount. This paper reports a new method of forward variable selection, enabled by the ordered nature of the basis functions in the KL expansion of the Bayesian Smoothing Spline ANOVA kernel (BSS-ANOVA), coupled with fast Gibbs sampling in a fully Bayesian approach. It quickly and effectively limits the number of terms, yielding a method with competitive accuracies, training and inference times for tabular datasets of low feature set dimensionality. Theoretical computational complexities are O ( N P 2 ) in training and O ( P ) per point in inference, where N is the number of instances and P the number of expansion terms. The inference speed and accuracy makes the method especially useful for dynamic systems identification, by modeling the dynamics in the tangent space as a static problem, then integrating the learned dynamics using a high-order scheme. The methods are demonstrated on two dynamic datasets: a ‘Susceptible, Infected, Recovered’ (SIR) toy problem, along with the experimental ‘Cascaded Tanks’ benchmark dataset. Comparisons on the static prediction of time derivatives are made with a random forest (RF), a residual neural network (ResNet), and the Orthogonal Additive Kernel (OAK) inducing points scalable GP, while for the timeseries prediction comparisons are made with LSTM and GRU recurrent neural networks (RNNs) along with the SINDy package.

Hayes, Kyle↗
DOE OSTI2025

Dataset for "Machine Learning Ensembles Can Enhance Hydrologic Predictions and Uncertainty Quantification" Willard et al. (2025).

This data release provides all data and code used in the paper " "Machine Learning Ensembles Can Enhance Hydrologic Predictions and Uncertainty Quantifications" Willard et al. (2025)" to model stream temperature, evaluate, and assess results. The associated manuscript explores the effect of different ensemble construction techniques across different common machine learning (ML) architectures for predictions in unmonitored basins. Modeling was done using long short-term memory (LSTM), gated recurrent unit (GRU), temporal convolution network (TCN), and extreme gradient boosting (XGBoost) models, and stream site coverage spans 1362 locations across the conterminous United States. The ensemble construction techniques investigated include ensemble by random weight initialization, differing hyperparameters, different random subsets of training data, different subselections of input features, different architectures, and Monte Carlo Dropout. The data is organized into these items items:Code repository and data for the paper " "Machine Learning Ensembles Can Enhance Hydrologic Predictions and Uncertainty Quantifications" Willard et al. (2025).Code: stream_temp_ml_regionalization.zip contains the code repositoryData to run the code:- data_dir.zip -- contains all files that should be moved to the "DATA_DIR" variable defined in the "set_env_vars.sh" script in the code repository- metadata_dir.zip -- contains all files that should be moved to the "METADATA_DIR" variable defined in the "set_env_vars.sh" script in the code repositoryData produced by the code and used in the paper:- outputs_dir.zip - contains model output and results (outputs_dir/results), model weights (outputs_dir/models), and all other outputs used for the paper including feature importances.To cite this code, please use the following BibTeX or MLA entries:bibtex:@misc{willard2025streamensembles,author = {Jared Willard and Charuleka Varadharajan},title = {Dataset for "Machine Learning Ensembles Can Enhance Hydrologic Predictions and Uncertainty Quantification"},year = {2024},doi = {10.15485/2527393},publisher = {ESS-DIVE Repository},url = {https://data.ess-dive.lbl.gov/datasets/doi:10.15485/2527393}}MLA: Willard, Jared, et al. Dataset for "Machine Learning Ensembles Can Enhance Hydrologic Predictions and Uncertainty Quantification". 2025. ESS-DIVE Repository, doi:10.15485/2448016.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Machine Learning for Slow Spill Regulation in the Fermilab Delivery Ring for Mu2e

A third-integer resonant slow extraction system is being developed for the Fermilab’s Delivery Ring to deliver protons to the Mu2e experiment. During a slow extraction process, the beam on target is liable to experience small intensity variations due to many factors. Owing to the experiment’s strict requirements in the quality of the spill, a Spill Regulation System (SRS) is currently under design. The SRS primarily consists of three components - slow regulation, fast regulation, and harmonic content tracker. In this presentation, we shall present the investigations of using Machine Learning (ML) in the fast regulation system, including further optimizations of PID controller gains for the fast regulation, prospects of an ML agent completely replacing the PID controller using supervised learning schemes such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) ML models, the simulated impact and limitation of machine response characteristics on the effectiveness of both PID and ML regulation of the spill. We also present here nascent results of Reinforcement Learning efforts, including continuous-action soft actor-critic methods, to regulate the spill rate.

43 PARTICLE ACCELERATORS↗
DOE OSTI2024

Transformer Neural Networks with Spatiotemporal Attention for Predictive Control and Optimization of Industrial Processes

In the context of real-time optimization and model predictive control of industrial systems, machine learning, and neural networks represent cutting-edge tools that hold promise for enhancing dynamic modeling. This work presents a novel transformer neural network architecture for real-time optimization and model predictive control. This network design includes a modified attention mechanism inspired by positional embedding attention from vision transformers and task-specific modifications to the input-output structure of the transformer’s decoder stack. Experiments were conducted using data from a 450 MW coal-fired power plant to evaluate this approach's effectiveness. The transformer neural network was compared with conventional recurrent models, including GRU and LSTM. The transformer exhibited a 6% increase in the R-squared (R2) value of predictions and an 83% reduction in mean squared error (MSE). Computation time was also reduced by 84% compared to conventional recurrent models.

Gallup, Ethan R.↗
DOE OSTI2022

Proper Motions, Orbits, and Tidal Influences of Milky Way Dwarf Spheroidal Galaxies

We combine Gaia early data release 3 astrometry with accurate photometry and utilize a probabilistic mixture model to measure the systemic proper motion of 52 dwarf spheroidal (dSph) satellite galaxies of the Milky Way (MW). For the 46 dSphs with literature line-of-sight velocities we compute orbits in both a MW and a combined MW + Large Magellanic Cloud (LMC) potential and identify Car II, Car III, Hor I, Hyi I, Phx II, and Ret II as likely LMC satellites. 40% of our dSph sample has a >25% change in pericenter and/or apocenter with the MW + LMC potential. For these orbits, we use a Monte Carlo sample for the observational uncertainties for each dSph and the uncertainties in the MW and LMC potentials. We predict that Ant II, Boo III, Cra II, Gru II, and Tuc III should be tidally disrupting by comparing each dSph's average density relative to the MW density at its pericenter. dSphs with large ellipticity (CVn I, Her, Tuc V, UMa I, UMa II, UMi, Wil 1) show a preference for their orbital direction to align with their major axis even for dSphs with large pericenters. We compare the dSph radial orbital phase to subhalos in MW-like N-body simulations and infer that there is not an excess of satellites near their pericenter. With projections of future Gaia data releases, we find that dSph's orbital precision will be limited by uncertainties in the distance and/or MW potential rather than in proper motion precision. Finally, we provide our membership catalogs to enable community follow-up.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

KGML-ag: a modeling framework of knowledge-guided machine learning to simulate agroecosystems: a case study of estimating N<sub>2</sub>O emission using data from mesocosm experiments

Abstract. Agricultural nitrous oxide (N2O) emission accounts for a non-trivial fraction of global greenhouse gas (GHG) budget. To date, estimating N2O fluxes from cropland remains a challenging task because the related microbial processes (e.g., nitrification and denitrification) are controlled by complex interactions among climate, soil, plant and human activities. Existing approaches such as process-based (PB) models have well-known limitations due to insufficient representations of the processes or uncertainties of model parameters, and due to leverage recent advances in machine learning (ML) a new method is needed to unlock the “black box” to overcome its limitations such as low interpretability, out-of-sample failure and massive data demand. In this study, we developed a first-of-its-kind knowledge-guided machine learning model for agroecosystems (KGML-ag) by incorporating biogeophysical and chemical domain knowledge from an advanced PB model, ecosys, and tested it by comparing simulating daily N2O fluxes with real observed data from mesocosm experiments. The gated recurrent unit (GRU) was used as the basis to build the model structure. To optimize the model performance, we have investigated a range of ideas, including (1) using initial values of intermediate variables (IMVs) instead of time series as model input to reduce data demand; (2) building hierarchical structures to explicitly estimate IMVs for further N2O prediction; (3) using multi-task learning to balance the simultaneous training on multiple variables; and (4) pre-training with millions of synthetic data generated from ecosys and fine-tuning with mesocosm observations. Six other pure ML models were developed using the same mesocosm data to serve as the benchmark for the KGML-ag model. Results show that KGML-ag did an excellent job in reproducing the mesocosm N2O fluxes (overall r2=0.81, and RMSE=3.6 mgNm-2d-1 from cross validation). Importantly, KGML-ag always outperforms the PB model and ML models in predicting N2O fluxes, especially for complex temporal dynamics and emission peaks. Besides, KGML-ag goes beyond the pure ML models by providing more interpretable predictions as well as pinpointing desired new knowledge and data to further empower the current KGML-ag. We believe the KGML-ag development in this study will stimulate a new body of research on interpretable ML for biogeochemistry and other related geoscience processes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Third-integer Resonant Extraction Regulation System for Mu2e

A third-integer resonant slow extraction system is being developed for Fermilab's Delivery Ring to deliver protons to the upcoming Mu2e experiment. The timescale of the extraction (or spill) duration is 43 milliseconds, which is extremely short and unprecedented. Additionally, the experiment's strict and challenging requirements on the quality of the spill at this time scale has led to the development of a new Spill Regulation System (SRS) design. The SRS primarily consists of three components - slow regulation, fast regulation, and harmonic content suppressor. Contributions to the first two components of the SRS, i.e., Slow Regulation and Fast Regulation subsystems, will be presented in which new adaptive learning algorithm schemes for the slow regulation of the spill -- validated using particle tracking simulations -- shall be described. In addition to these novel methods for the enhancement of the spill regulation system, results of employing Machine Learning in enhancing the performance of the resonant extraction are also presented. At the forefront of applying ML techniques to solve non-linear accelerator control problems, this work includes optimizing the PID gains as well as the replacement of the traditional PID controller using Recurrent Neural Networks and Gated Recurrent Unit (GRU) ML models to achieve efficiencies greater than a PID controller. Cutting-edge on-going Reinforcement Learning efforts, including an actor-critic family of learning algorithms, to regulate the spill rate will be reviewed, as well as present analytical calculations pertaining the transit time of particles in a third-integer resonant extraction. Detailed numerical investigations and validations of such calculations, the model of which could be exported and reliably used in future analytical modeling of any resonant extraction, are discussed.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

Ultra-low latency recurrent neural network inference on FPGAs for physics applications with hls4ml

Abstract Recurrent neural networks have been shown to be effective architectures for many tasks in high energy physics, and thus have been widely adopted. Their use in low-latency environments has, however, been limited as a result of the difficulties of implementing recurrent architectures on field-programmable gate arrays (FPGAs). In this paper we present an implementation of two types of recurrent neural network layers—long short-term memory and gated recurrent unit—within the hls4ml framework. We demonstrate that our implementation is capable of producing effective designs for both small and large models, and can be customized to meet specific design requirements for inference latencies and FPGA resources. We show the performance and synthesized designs for multiple neural networks, many of which are trained specifically for jet identification tasks at the CERN Large Hadron Collider.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Empirical Comparison of Machine Learning Approaches for Black-Box Modeling of Power Conversion System Dynamics

Inverter-based resources are key components in modern power systems, but accurately modeling their complex behavior can be challenging. Standard, generic converter models often oversimplify inverter dynamics, leading to significant errors in predicting performance. In this work, we compare several data-driven machine learning (ML) approaches for inverter modeling, performing experiments on power conversion systems, systematically varying input conditions, and recording the resulting voltages and currents. The ML models were then trained on this measured data to capture the inverter's dynamic response and to predict the inverter's output current. A performance comparison between the four ML models under study is conducted, laying the foundation for future work on hardware implementation for real-time inference.

30 DIRECT ENERGY CONVERSION↗