Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed training”

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 1,153 records · Page 64

Prostate brachytherapy postimplant dosimetry: Seed orientation and the impact of dosimetric anisotropy in stranded implants

Purpose: In postimplant dosimetry for prostate brachytherapy, dose is commonly calculated using the TG-43 1D formalism, because seed orientations are difficult to determine from CT images, the current standard for the procedure. However, the orientation of stranded seeds soon after implantation is predictable, as these seeds tend to maintain their relative spacing, and orient themselves along the implant trajectory. The aim of this study was to develop a method for determining seed orientations from reconstructed strand trajectories, and to use this information to investigate the dosimetric impact of applying the TG-43 2D formalism to clinical postimplant analysis. Methods: Using in-house software, the preplan to postimplant seed correspondence was determined for a cohort of 30 patients during routine day-0 CT-based postimplant dosimetry. All patients were implanted with stranded-seed trains. Spline curves were fit to each set of seeds composing a strand, with the requirement that the distance along the spline between seeds be equal to the seed spacing within the strand. The orientations of the seeds were estimated by the tangents to the spline at each seed centroid. Dose distributions were then determined using the 1D and 2D TG-43 formalisms. These were compared using the TG-137 recommended dose metrics for the prostate, prostatic urethra, and rectum. Results: Seven hundred and sixty one strands were analyzed in total. Defining the z-axis to be cranial-positive and the x-axis to be left-lateral positive in the CT coordinate system, the average seed had an inclination of 21 deg. {+-} 10 deg. and an azimuth of -81 deg. {+-} 57 deg. These values correspond to the average strand rising anteriorly from apex to base, approximately parallel to the midsagittal plane. Clinically minor but statistically significant differences in dose metrics were noted. Compared to the 2D calculation, the 1D calculation underestimated prostate V100 by 1.1% and D90 by 2.3 Gy, while overestimating V150 and V200 by 1.6% and 1.3%, respectively. Urethral and rectal dose quantifiers tended to be underestimated by the 1D calculation. The most pronounced differences were in the urethral D30 and rectal D2cc, which rose by 3.8 and 1.9 Gy, respectively, using the 2D calculation. The total volume of the 100% isodose region as a percentage of the prostate volume was found to increase by 0.4%. Conclusions: Stranded seeds in the supine patient are not oriented in a uniformly random manner, nor are they aligned along the axis of the CT scanner. Instead, this study identified a consistent anterior pitch that is likely attributable to differences in patient pose between implant and CT imaging. The angle of the ultrasound probe with respect to the patient during implant may have also been a contributing factor. The dose metrics derived using the 1D formalism were found to be within 2%, on average, of those derived using the 2D formalism. For greater accuracy, 2D dosimetry can be pursued using the strand-fitting method described in this work. If a 1D representation is used, integrating over the empirically determined seed orientation density reported here may be more appropriate than assuming that seed inclinations are distributed uniformly.

61 RADIATION PROTECTION AND DOSIMETRY↗

PowerNet: Multi-agent Deep Reinforcement Learning for Scalable Powergrid Control

This paper develops an efficient multi-agent deep reinforcement learning algorithm for cooperative controls in powergrids. Specifically, we consider the decentralized inverter-based secondary voltage control problem in distributed generators (DGs), which is first formulated as a cooperative multi-agent reinforcement learning (MARL) problem. We then propose a novel on-policy MARL algorithm, PowerNet, in which each agent (DG) learns a control policy based on (sub-)global reward but local states and encoded communication messages from its neighbors. Motivated by the fact that a local control from one agent has limited impact on agents distant from it, we exploit a novel spatial discount factor to reduce the effect from remote agents, to expedite the training process and improve scalability. Furthermore, a differentiable, learning-based communication protocol is employed to foster the collaborations among neighboring agents. In addition, to mitigate the effects of system uncertainty and random noise introduced during on-policy learning, we utilize an action smoothing factor to stabilize the policy execution. To facilitate training and evaluation, we develop PGSim, an efficient, high-fidelity powergrid simulation platform. Here, experimental results in two microgrid setups show that the developed PowerNet outperforms the conventional model-based control method, as well as several state-of-the-art MARL algorithms. The decentralized learning scheme and high sample efficiency also make it viable to large-scale power grids.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Thermodynamics of Restricted Boltzmann Machines and Related Learning Dynamics

We investigate the thermodynamic properties of a restricted Boltzmann machine (RBM), a simple energy-based generative model used in the context of unsupervised learning. Assuming the information content of this model to be mainly reflected by the spectral properties of its weight matrix W, we try to make a realistic analysis by averaging over an appropriate statistical ensemble of RBMs. First, a phase diagram is derived. Otherwise similar to that of the Sherrington–Kirkpatrick (SK) model with ferromagnetic couplings, the RBM’s phase diagram presents a ferromagnetic phase which may or may not be of compositional type depending on the kurtosis of the distribution of the components of the singular vectors of W. Subsequently, the learning dynamics of the RBM is studied in the thermodynamic limit. A “typical” learning trajectory is shown to solve an effective dynamical equation, based on the aforementioned ensemble average and explicitly involving order parameters obtained from the thermodynamic analysis. In particular, this let us show how the evolution of the dominant singular values of W, and thus of the unstable modes, is driven by the input data. At the beginning of the training, in which the RBM is found to operate in the linear regime, the unstable modes reflect the dominant covariance modes of the data. In the non-linear regime, instead, the selected modes interact and eventually impose a matching of the order parameters to their empirical counterparts estimated from the data. Finally, we illustrate our considerations by performing experiments on both artificial and real data, showing in particular how the RBM operates in the ferromagnetic compositional phase.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Portable chemical detection platform for on-site monitoring of odorant levels in natural gas

The adequate odorization of natural gas is critical to identify gas leaks and to reduce accidents. To ensure odorization, natural gas utility companies collect samples to be processed at core facilities or a trained human technician smells a diluted natural gas sample. In this work, we report a detection platform that addresses the lack of mobile solutions capable of providing quantitative analysis of mercaptans, a class of compounds used to odorize natural gas. Detailed description of the platform hardware and software components is provided. Designed to be portable, the platform hardware facilitates extraction of mercaptans from natural gas, separation of individual mercaptan species, and quantification of odorant concentration, with results reported at point-of-sampling. The software was developed to accommodate skilled users as well as minimally trained operators. Detection and quantification of six commonly used mercaptan compounds (ethyl mercaptan, dimethyl sulfide, n-propylmercaptan, isopropyl mercaptan, tert–butyl mercaptan, and tetrahydrothiophene) at typical odorizing concentrations of 0.1–5 ppm was performed using the device. We demonstrate the potential of this technology to ensure natural gas odorizing concentrations throughout distribution systems.

03 NATURAL GAS↗

Convolution Neural Network for Fault Identification in Distribution Feeder with High Penetration Solar PV

Identification and zonal classification of the faults is a decisive factor in the relay’s decision to trip or not. Different types of fault like three-phase, line-to-line-to-ground and single-line-to-ground can occur at various locations in the feeder. These faults are seen as the variation in the instantaneous values of three-phase voltages and currents, i.e., waveforms, that are measured at the relay location. The objective of this work is to develop a machine learning model that can identify a fault and classify it to various protection zones based on measured waveforms. In this work, a data-driven relay based on Convolutional Neural Network (CNN) is proposed for fault identification in distribution feeders with high penetration solar PV. The proposed CNN model takes local current and voltage waveforms as input and classify it into fault, no-fault or a capacitor switching. Further, the CNN also attempts to identify fault zones based on the images of waveforms. The overall testing accuracy of the trained model exceeds 95%.

Ramesh, Meghana↗

A Guide for Public Utility Commissions: Building Internal Technical Capacity and Recruiting Talent for Grid Resilience

This guide offers insight into how PUCs can strategically expand their technical workforce to meet evolving gird resilience demands. It outlines critical skill sets needed to support informed regulatory decision-making around resilience, such as modeling and weather forecasting, electric power systems analysis, and advanced data interpretation. In addition, it provides strategies for developing technical talent both internally and through additional recruitment efforts. Two appendices provide a list of grid resilience training resources and a compendium of sample job descriptions that reflect grid resilience technical expertise for PUC consideration.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Improved Bayesian Infrasonic Source Localization for regional infrasound

The Bayesian Infrasonic Source Localization (BISL) methodology is examined and simplified providing a generalized method of estimating the source location and time for an infrasonic event and the mathematical framework is used therein. The likelihood function describing an infrasonic detection used in BISL has been redefined to include the von Mises distribution developed in directional statistics and propagation-based, physically derived celerity-range and azimuth deviation models. Frameworks for constructing propagation-based celerity-range and azimuth deviation statistics are presented to demonstrate how stochastic propagation modelling methods can be used to improve the precision and accuracy of the posterior probability density function describing the source localization. Infrasonic signals recorded at a number of arrays in the western United States produced by rocket motor detonations at the Utah Test and Training Range are used to demonstrate the application of the new mathematical framework and to quantify the improvement obtained by using the stochastic propagation modelling methods. Moreover, using propagation-based priors, the spatial and temporal confidence bounds of the source decreased by more than 40 per cent in all cases and by as much as 80 per cent in one case. Further, the accuracy of the estimates remained high, keeping the ground truth within the 99 per cent confidence bounds for all cases.

58 GEOSCIENCES↗

Computer Vision Pipeline for Image Analysis for Freeze‐Fracture Electron Microscopy: Rosette Cellulose Synthase Complexes Case

In materials science, plant biology, agriculture, and environmental research, the automated analysis of high-magnification, complex microscopy images, such as those generated by freeze-fracture electron microscopy (FF-TEM), remains a critical challenge that limits the scalability of data interpretation. We present a deep learning computer vision pipeline for high-throughput detection and morphological characterization analysis of cellulose synthase complexes (CSCs, or rosettes) in FF-TEM images. The pipeline integrates preprocessing, detection, human-in-the-loop verification, and semantic segmentation to quantify features such as rosette diameter and inter-lobe spacing. The approach was trained and tested on a curated dataset of high-resolution FF-TEM micrographs of Physcomitrium patens, expanded via strategic tiling and augmentation to over 650 images. We compare YOLOv8 and YOLOv9 architectures and demonstrate that YOLOv9 achieves superior performance in both localization accuracy (mAP50-95 = 0.854) and inference speed. The resulting distributions revealed biological variability consistent with prior manual studies, validating the approach for high-throughput applications. Our results show that the pipeline achieves human-expert level accuracy while dramatically reducing analysis time, enabling scalable, reproducible structural characterization of intramembrane protein complexes. The pipeline is broadly applicable to other domains requiring precise interpretation of complex microscopy data and establishes a foundation for future artificial intelligence (AI)-assisted workflows in biological imaging.

59 BASIC BIOLOGICAL SCIENCES↗

Characterization of mechanical discontinuities based on data-driven classification of compressional-wave travel times

Wave propagation and diffusive transport phenomena are influenced by the mechanical discontinuities in material. Here this study shows that certain bulk properties of the network of low-velocity mechanical discontinuities (e.g. air-filled cracks) in a material can be characterized by processing compressional-wave travel times using traditional data-driven classification techniques. To that end, we perform three tasks in chronological order: (1) use the discrete fracture network (DFN) method to create two-dimensional (2D) numerical models of crack-bearing material embedded with various types of low-velocity mechanical discontinuities, (2) use the fast marching method (FMM) to simulate the propagation of the wave/diffusion front from a single source through the 2D crack-bearing material to multiple receivers placed on the boundary of the material, and (3) train 9 data-driven classifiers to characterize the crack-bearing materials (i.e. bulk properties of the network of mechanical discontinuities in the crack-bearing material) by learning from the simulations of travel times detected by multiple receivers placed around the crack-bearing material. The classifiers identified the orientation, spatial distribution, and dispersion of the low-velocity mechanical discontinuities. Voting classifier performs the best among the 9 classifiers. For the characterization of bulk dispersion and distribution of discontinuities, the sensors located on the adjacent boundaries are more important; whereas for the characterization of bulk orientation of discontinuities, the sensors located on the opposite side are more important.

42 ENGINEERING↗

Prediction of the Cu oxidation state from EELS and XAS spectra using supervised machine learning

Abstract Electron energy loss spectroscopy (EELS) and X-ray absorption spectroscopy (XAS) provide detailed information about bonding, distributions and locations of atoms, and their coordination numbers and oxidation states. However, analysis of XAS/EELS data often relies on matching an unknown experimental sample to a series of simulated or experimental standard samples. This limits analysis throughput and the ability to extract quantitative information from a sample. In this work, we have trained a random forest model capable of predicting the oxidation state of copper based on its L-edge spectrum. Our model attains an R 2 score of 0.85 and a root mean square error of 0.24 on simulated data. It has also successfully predicted experimental L-edge EELS spectra taken in this work and XAS spectra extracted from the literature. We further demonstrate the utility of this model by predicting simulated and experimental spectra of mixed valence samples generated by this work. This model can be integrated into a real-time EELS/XAS analysis pipeline on mixtures of copper-containing materials of unknown composition and oxidation state. By expanding the training data, this methodology can be extended to data-driven spectral analysis of a broad range of materials.

36 MATERIALS SCIENCE↗

Designing a quantum-accurate machine-learning potential to enable large-scale simulations of deuterium under shock

Large-scale molecular dynamics of deuterium under shock can elucidate kinetic processes vital to the target design in inertial confinement fusion and high-energy-density experiments. However, modeling the complex evolution of this material from an insulating molecular state at ambient pressure to an ionized, atomic fluid under strong shock is beyond the capability of simple pair and even bond order potentials. We thus train a quantum-accurate and broadly transferable machine-learning interatomic potential for deuterium using the Chebyshev Interaction Model for Efficient Simulations framework. We show that due to an improved description of the molecular-to-atomic transition, our model is able to better reproduce the ab initio equation of state, radial distribution functions, and principal Hugoniot than bond order potentials. This represents an important step toward large-scale quantum-accurate and nonequilibrium simulations of complicated systems under dynamic changes including phase transitions.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Accelerated Probabilistic Marching Cubes by Deep Learning for Time-Varying Scalar Ensembles

Visualizing the uncertainty of ensemble simulations is challenging due to the large size and multivariate and temporal features of en-semble data sets. One popular approach to studying the uncertainty of ensembles is analyzing the positional uncertainty of the level sets. Probabilistic marching cubes is a technique that performs Monte Carlo sampling of multivariate Gaussian noise distributions for positional uncertainty visualization of level sets. However, the technique suffers from high computational time, making interactive visualization and analysis impossible to achieve. This paper introduces a deep-learning-based approach to learning the level-set uncertainty for two-dimensional ensemble data with a multivariate Gaussian noise assumption. We train the model using the first few time steps from time-varying ensemble data in our workflow. We demonstrate that our trained model accurately infers uncertainty in level sets for new time steps and is up to 170X faster than that of the original probabilistic model with serial computation and 10X faster than that of the original parallel computation.

Han, Mengjiao↗

Improve Learning from Crowds via Generative Augmentation

Crowdsourcing provides an efficient label collection schema for supervised machine learning. However, to control annotation cost, each instance in the crowdsourced data is typically annotated by a small number of annotators. This creates a sparsity issue and limits the quality of machine learning models trained on such data. In this paper, we study how to handle sparsity in crowdsourced data using data augmentation. Specifically, we propose to directly learn a classifier by augmenting the raw sparse annotations. We implement two principles of high-quality augmentation using Generative Adversarial Networks: 1) the generated annotations should follow the distribution of authentic ones, which is measured by a discriminator; 2) the generated annotations should have high mutual information with the ground-truth labels, which is measured by an auxiliary network. Extensive experiments and comparisons against an array of state-of-the-art learning from crowds methods on three real-world datasets proved the effectiveness of our data augmentation framework. It shows the potential of our algorithm for low-budget crowdsourcing in general.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Machine Learning the Sixth Dimension: Stellar Radial Velocities from 5D Phase-space Correlations

The Gaia satellite will observe the positions and velocities of over a billion Milky Way stars. In the early data releases, the majority of observed stars do not have complete 6D phase-space information. In this Letter, we demonstrate the ability to infer the missing line-of-sight velocities until more spectroscopic observations become available. Here, we utilize a novel neural network architecture that, after being trained on a subset of data with complete phase-space information, takes in a star's 5D astrometry (angular coordinates, proper motions, and parallax) and outputs a predicted line-of-sight velocity with an associated uncertainty. Working with a mock Gaia catalog, we show that the network can successfully recover the distributions and correlations of each velocity component for stars that fall within ~5 kpc of the Sun. We also demonstrate that the network can accurately reconstruct the velocity distribution of a kinematic substructure in the stellar halo that is spatially uniform, even when it comprises a small fraction of the total star count.

79 ASTRONOMY AND ASTROPHYSICS↗

Lessons Learned in Employing Data Analytics to Predict Oxidation Kinetics and Spallation Behavior of High-Temperature NiCr-Based Alloys

Machine learning (ML) can offer many advantages in predicting material properties over traditional materials development methods based solely on limited experimental investigations or physical-based simulations with the capability to reduce development cost, risk, and time. However, so far, limited efforts have been made to predict alloy oxidation kinetics and spallation behavior via ML due to the lack of consistently measured and sufficient experimental data and the inherent complexity in oxidation behavior of multicomponent high-temperature alloys. A previous study reported the ability of ML to predict oxidation kinetics of NiCr-based alloys as a function of alloy composition and operating conditions. Here, the performance of a ML model in predicting rate constants and spallation probability was evaluated in light of the roles of the data distribution of the experimental dataset (data analytics), the alloy composition, the exposure environment and the chosen oxidation approach to extracting kinetic values from the measured mass changes (but using either a simple parabolic law or a statistical cyclic oxidation model). Potential strategies to improve the predictions and enhance the extrapolative capability of the previously trained model will be discussed.

36 MATERIALS SCIENCE↗

How good are learning-based control v.s. model-based control for load shifting? Investigations on a single zone building energy system

Both model predictive control (MPC) and deep reinforcement learning control (DRL) have been presented as a way to approximate the true optimality of a dynamic programming problem, and these two have shown significant operational cost saving potentials for building energy systems. Furthermore, there is still a lack of in-depth quantitative studies on their approximation levels to the true optimality, especially in the building energy domain. To fill in the gap, this paper provides a numerical framework that enables the evaluation of the optimality levels of different controllers for building energy systems. This framework is then used to comprehensively compare the optimal control performance of both MPC and DRL controllers with given computation budgets for a single zone fan coil unit system. Note the optimality is estimated based on a user-specific selection of trade-off weights among energy costs, thermal comfort and control slew rates. Compared with the best optimality we can find through expensive optimization simulations, the best DRL agent can maximally approximate the optimality by 96.54%, which outperforms the best MPC whose optimality level is 90.11%. However, due to the stochasticity, the DRL agent is only expected to approximate the optimality by 90.42%, which is almost equivalent to the best MPC. Except for Proximal Policy Optimization (PPO), all DRL agents can have a better approximation to the optimality than the best MPC, and are expected to have better approximation than the MPC with a prediction horizon of 32 steps (15 min per step). In terms of reducing energy cost and thermal discomfort, MPC can outperform the rule-based control (RBC) by 18.47%–25.44%. DRL can be expected to outperform RBC by 18.95%–25.65% ,and the best DRL control policy can outperform RBC by 20.29%–29.72%. Although the comparison of the optimality level is performed in a perfect setting, e.g., MPC assumes perfect models, and DRL assumes a perfect offline training process and online deployment process, this can shed insight on their capabilities of approximating to the original dynamic programming problem.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Enhanced dynamic equivalent identification method of large-scale power systems using multiple events

The increasing complexity of the interconnected power system makes high-fidelity dynamic simulation models computationally more intensive. To improve computation efficiency, model reduction techniques have been investigated to only preserve the dynamics in a limited area of interest (study area), while deriving equivalent representation for the external area. For this purpose, a measurement-based reduction approach using system identification techniques has been previously proposed. Accordingly, external areas are represented by dynamic equivalent loads using transfer function estimation. In this paper, in order to enhance the accuracy of the reduced model in preserving dynamics of the study area, multiple “grid” events of different types and at different locations are considered for identifying the parameters of the equivalent loads. Additionally, case studies are carried out in the NPCC 140-bus system. Comparisons are made between multiple events training and single event training, highlighting the advantages of the proposed method in providing a better representation of the grid dynamics under different operating conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Variable magnetic field electron spectrometer to measure hot electrons in the range of 50–460 keV

Resonance absorption (RA) occurs when a p-polarized electromagnetic wave, obliquely incident on an inhomogeneous plasma, tunnels past its turning point and resonantly excites an electron plasma wave (EPW) at the critical density. This phenomenon is important, for instance, in the direct drive approach to inertial fusion energy and is a particular example of a wider phenomenon in plasma physics known as mode conversion, which is crucial for heating magnetic fusion devices, such as tokamaks, via RF heating. Direct measurement of these RA-generated EPW accelerated hot electrons, with energy in the range of a few tens to a few hundreds of keV, is a challenging task due to the relatively low deflecting magnetic fields needed. The solution described here is a magnetic electron spectrometer (MES) with a continually changing magnetic field, lower at the entrance of the MES and gradually increasing toward the end, that enables the measurement of a wide spectral range of electrons with energies between 50 and 460 keV. Electron spectra taken in a LaserNetUS RA experiment were acquired from plasmas generated by irradiating polymer targets with the combination of an ~300 ps pulse followed by a series of ten high intensity 50–200 fs duration laser pulses from the ALEPH laser at Colorado State University. In conclusion, the high intensity beam is designed as spike trains of uneven duration and delay pulses in order to modify the RA phenomenon.

47 OTHER INSTRUMENTATION↗