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 1,171 records · Page 65

Artificial intelligence-based predictive modeling for imaging neutral particle analyzers on the DIII-D tokamak

The Imaging Neutral Particle Analyzer (INPA) at DIII-D is a diagnostic system used to accurately resolve the energy and spatial distributions of fast ions in fusion plasmas. A novel artificial intelligence (AI) technique named INPA-net is based on Reservoir Computing Networks and developed here to predict active and passive signals produced by charge-exchange reactions from injected and edge-cold neutrals, respectively, in magnetically confined fusion plasmas. This model is trained using a set of 21 time domain signals between 0 s to 3.35 s that includes injected beam and thermal plasma information, and 6444 real 2D experimental images of the INPA in 12 plasma discharges at DIII-D. The trained neural network is able to forecast experimental images in real-time. The model achieves an R-squared value of 0.91, which is higher than the 0.83 value achieved by a simple linear regression model. This improvement highlights the model's enhanced predictive accuracy for measured images from the validation set. This AI approach is valuable due to its rapid response times and potential for integration into real-time plasma control systems. A version of this model capable of generating syntehic images would be useful for the real-time monitoring of fast-ion transport. A comprehensive sensitivity study reveals that INPA-net maintains high performance even with variations in the input parameters, indicating the model's robustness and reliability. While developed for the INPA, the underlying architecture is adaptable and may be applied to various 2D imaging diagnostics in fusion research.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Stochastic Spectral Descent for Discrete Graphical Models

Interest in deep probabilistic graphical models has in-creased in recent years, due to their state-of-the-art performance on many machine learning applications. Such models are typically trained with the stochastic gradient method, which can take a significant number of iterations to converge. Since the computational cost of gradient estimation is prohibitive even for modestly sized models, training becomes slow and practically usable models are kept small. In this paper we propose a new, largely tuning-free algorithm to address this problem. Our approach derives novel majorization bounds based on the Schatten- norm. Intriguingly, the minimizers of these bounds can be interpreted as gradient methods in a non-Euclidean space. We thus propose using a stochastic gradient method in non-Euclidean space. We both provide simple conditions under which our algorithm is guaranteed to converge, and demonstrate empirically that our algorithm leads to dramatically faster training and improved predictive ability compared to stochastic gradient descent for both directed and undirected graphical models.

97 MATHEMATICS AND COMPUTING↗

Value Gradient Learning Approach in Power and Frequency Regulation of Grid-Connected Synchronverters

In this paper, a neural network adaptive critic design (ACD) method based on value gradient learning (VGL) is used to optimally control a grid-connected synchronverter. The main drawback of the traditional synchronverters is their infeasibility to face non-inductive grids. To be able to implement a synchronverter technique in any impedance angle, a neural network-based adaptive controller is used. The advantage of adaptive dynamic programing is its ability to adjust itself when it faces changes and uncertainties in the power system. The proposed VGL consists of two subnetworks: the critic network and the action network. The action network is trained during the operation, and the critic network can be pretrained offline or can be simultaneously trained with the action network. To compare the effectiveness of the traditional synchronverter to the VGL-based synchronverter, the simulation results are provided.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Sensor Error Estimation for Reactor Coolant System with Generalization Error Equations

We consider a simplified version of the sensor error estimation problem in pressure sensors in the primary coolant system of a nuclear power plant under the condition that a majority of sensors are error-free. We present a sensor fusion method to estimate sensor errors in the primary coolant system of a power reactor by using measurements collected by multiple sensors. The underlying relationship between the measured and estimated parameters is inferred using measurements collected under controlled no-error conditions of the plant. We propose a machine learning approach to train fusers that provide an estimate of the sensor measurement by using measurements from other sensors, and the difference between the actual sensor measurement and its estimate provides the sensor estimate. We present generalization error equations for these methods that characterize the performance of their sensor error estimates on future measurements using the distribution-free machine learning formulation.

Rao, Nageswara S.↗

Advanced control techniques for modern inertia based inverters

In this research three artificial intelligent (AI)-based techniques are proposed to regulate the voltage and frequency of a grid-connected inverter. The increase in the penetration of renewable energy sources (RESs) into the power grid has led to the increase in the penetration of fast-responding inertia-less power converters. The increase in the penetration of these power electronics converters changes the nature of the conventional grid, in which the existing kinetic inertia in the rotating parts of the enormous generators plays a vital role. The concept of virtual inertia control scheme is proposed to make the behavior of grid connected inverters more similar to the synchronous generators, by mimicking the mechanical behavior of a synchronous generator. Conventional control techniques lack to perform optimally in nonlinear, uncertain, inaccurate power grids. Besides, the decoupled control assumption in conventional VSGs makes them nonoptimal in resistive grids. The neural network predictive controller, the heuristic dynamic programming, and the dual heuristic dynamic programming techniques are presented in this research to overcome the draw backs of conventional VSGs. The nonlinear characteristics of neural networks, and the online training enable the proposed methods to perform as robust and optimal controllers. The simulation and the experimental laboratory prototype results are provided to demonstrate the effectiveness of the proposed techniques.

24 POWER TRANSMISSION AND DISTRIBUTION↗

An efficient Bayesian inversion of a geothermal prospect using a multivariate adaptive regression spline method

In this study, an efficient Bayesian framework equipped with a multivariate adaptive regression spline (MARS) technique is developed to alleviate computational burdens encountered in a conventional Bayesian inversion of a geothermal prospect. Fast MARS models are developed from training dataset generated by CPU-intensive hydrothermal models and used as surrogate of high-fidelity physical models in Markov Chain Monte Carlo (MCMC) sampling. This Bayesian inference with MARS-enabled MCMC method is used to reduce prior estimates of uncertainty in structural or characteristic hydrothermal flow parameters of the model to posterior distributions. A geothermal prospect near Superstition Mountain in Imperial County of California in USA is used to illustrate the proposed framework and demonstrate the computational efficiency of MARS-based Bayesian inversion. The developed MARS models are also used to efficiently drive calculation of Sobol’ total sensitivity indices. Furthermore, only top sensitive parameters are included in Bayesian inference to further improve the computational efficiency of inversion. Sensitivity analysis also confirms that water circulation through high permeable structures, rather than heat conduction through impermeable granite, is the primary heat transfer method. The presented framework is demonstrated an efficient tool to update knowledge of geothermal prospects by inversing field data. Although only thermal data is used in this study, other type of data, such as flow and transport observations, can be jointly used in this method for underground hydrocarbon reservoirs.

15 GEOTHERMAL ENERGY↗

High-performance laser processing using manipulated ultrafast laser pulses

We employ manipulated ultrafast laser pulses to realize microprocessing with high-performance. Efficient microwelding of glass substrates by irradiation by a double-pulse train of ultrafast laser pulses is demonstrated. The bonding strength of two photostructurable glass substrates welded by double-pulse irradiation was evaluated to be 22.9 MPa, which is approximately 22% greater than that of a sample prepared by conventional irradiation by a single pulse train. Additionally, the fabrication of hollow microfluidic channels with a circular cross-sectional shape embedded in fused silica is realized by spatiotemporally focusing the ultrafast laser beam. We show both theoretically and experimentally that the spatiotemporal focusing of ultrafast laser beam allows for the creation of a three-dimensionally symmetric spherical peak intensity distribution at the focal spot.

36 MATERIALS SCIENCE↗

Extraction of the muon signals recorded with the surface detector of the Pierre Auger Observatory using recurrent neural networks

The Pierre Auger Observatory, at present the largest cosmic-ray observatory ever built, is instrumented with a ground array of 1600 water-Cherenkov detectors, known as the Surface Detector (SD). The SD samples the secondary particle content (mostly photons, electrons, positrons and muons) of extensive air showers initiated by cosmic rays with energies ranging from 1017eV up to more than 1020eV. Measuring the independent contribution of the muon component to the total registered signal is crucial to enhance the capability of the Observatory to estimate the mass of the cosmic rays on an event-by-event basis. However, with the current design of the SD, it is difficult to straightforwardly separate the contributions of muons to the SD time traces from those of photons, electrons and positrons. In this paper, we present a method aimed at extracting the muon component of the time traces registered with each individual detector of the SD using Recurrent Neural Networks. We derive the performances of the method by training the neural network on simulations, in which the muon and the electromagnetic components of the traces are known. We conclude this work showing the performance of this method on experimental data of the Pierre Auger Observatory. We find that our predictions agree with the parameterizations obtained by the AGASA collaboration to describe the lateral distributions of the electromagnetic and muonic components of extensive air showers.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Robust neural network-enhanced estimation of local primordial non-Gaussianity

When applied to the nonlinear matter distribution of the universe, neural networks have been shown to be very statistically sensitive probes of cosmological parameters, such as the linear perturbation amplitude σ 8 . However, when used as a “black box,” neural networks are not robust to baryonic uncertainty. We propose a robust architecture for constraining primordial non-Gaussianity f NL , by training a neural network to locally estimate σ 8 , and correlating these local estimates with the large-scale density field. We apply our method to N-body simulations, and show that σ⁡(f NL ) is 3.5 times better than the constraint obtained from a standard halo-based approach. Finally, we show that our method has the same robustness property as large-scale halo bias: baryonic physics can change the normalization of the estimated f NL , but cannot change whether f NL is detected.

79 ASTRONOMY AND ASTROPHYSICS↗

Adaptive Computing (AC) [SWR-24-106]

The Adaptive Computing (AC) software stack supports goal-based computing, for which a simulation workload is created on the fly adapting to the results of calculations. Application-specific code defines an objective, which may be to solve an optimization problem or to train a surrogate model with minimal uncertainty. Then, the AC driver decides where in the design parameter space to run simulations to best achieve that objective. This process is iterative and online; as new data is returned from simulations, the AC driver chooses new simulations to run. The AC driver can strategically run simulations on distributed hardware resources (including high performance computing machines, cloud resources, and edge devices) to maximize throughput and obey resource constraints.

Griffin, Kevin [National Renewable Energy Laborato↗

Adaptive Computing (AC) (Open Source) [SWR-24-106]

The Adaptive Computing (AC) software stack supports goal-based computing, for which a simulation workload is created on the fly, adapting to the results of calculations. Application-specific code defines an objective, which may be to solve an optimization problem or to train a surrogate model with minimal uncertainty. Then, the AC driver decides where in the design parameter space to run simulations to best achieve that objective. This process is iterative and online; as new data is returned from simulations, the AC driver chooses new simulations to run. The AC driver can strategically run simulations on distributed hardware resources (including high performance computing machines, cloud resources, and edge devices) to maximize throughput and obey resource constraints.

Griffin, Kevin [National Laboratory of the Rockies↗

ANALYSIS OF MULTIBUNCH SPECTRUM FOR AN UNEVEN BUNCH DISTRIBUTION IN A STORAGE RING

Modern storage-ring designs often require an uneven bunch distribution pattern. An uneven bunch fill pattern can result in complex structures for the beam current spectra. Particularly at high average beam currents, these complex current spectra need to be taken into account in concern of beam-dynamical effects. In this study, we analyze a beam current spectrum for various filling patterns with bunch trains and gaps. The characteristics of the resulting beam current spectra are illustrated and discussed.

Li, Rui↗

Tribal Colleges and Universities/American Indian Research and Education Initiatives Advanced Manufacturing Technical Assistance Project

The overall goal of this project is to establish a network of TCUs with essential advanced manufacturing (AM) facilities, associated training and education programs, and private sector and federal agency partnerships to both prepare an American Indian AM workforce and create economic and employment opportunities within Tribal communities through design, manufacturing, and marketing of high quality products. Some examples of high quality products involve next generation grid components such as mechanical energy storage, cabling for distribution of energy, and electrochemical energy storage enclosures. Sandia National Laboratories (Sandia) is tasked to provide technical advising, planning, and academic program development support for the TCU/American Indian Higher Education Consortium (AIHEC) Advanced Manufacturing Project. The TCUs include Bay Mills Community College (BMCC), Cankdeska Cikana Community College (CCCC), Navajo Technical University (NTU), Southwestern Indian Polytechnic Institute (SIPI), and Salish Kooteani College. AIHEC and Sandia, with collaboration from SIPI, will be establishing an 8-week summer institute on the SIPI campus during the summer of 2017. Up to 20 students from TCUs are anticipated to take part in the summer program. The goal of the program is to bring AM science, technology, engineering, and mathematics (STEM) awareness and opportunities for the American Indian students. Prior to the summer institute, Sandia will be providing reviews on curriculum plans at the each of the TCUs to ensure the content is consistent with current AM design and engineering practice. In addition, Sandia will provide technical assistance to each of the TCUs in regards to their current AM activities.

99 GENERAL AND MISCELLANEOUS↗

ClimateNet: an expert-labeled open dataset and deep learning architecture for enabling high-precision analyses of extreme weather

Abstract. Identifying, detecting, and localizing extreme weather events is a crucial first step in understanding how they may vary under different climate change scenarios. Pattern recognition tasks such as classification, object detection, and segmentation (i.e., pixel-level classification) have remained challenging problems in the weather and climate sciences. While there exist many empirical heuristics for detecting extreme events, the disparities between the output of these different methods even for a single event are large and often difficult to reconcile. Given the success of deep learning (DL) in tackling similar problems in computer vision, we advocate a DL-based approach. DL, however, works best in the context of supervised learning – when labeled datasets are readily available. Reliable labeled training data for extreme weather and climate events is scarce. We create “ClimateNet” – an open, community-sourced human-expert-labeled curated dataset that captures tropical cyclones (TCs) and atmospheric rivers (ARs) in high-resolution climate model output from a simulation of a recent historical period. We use the curated ClimateNet dataset to train a state-of-the-art DL model for pixel-level identification – i.e., segmentation – of TCs and ARs. We then apply the trained DL model to historical and climate change scenarios simulated by the Community Atmospheric Model (CAM5.1) and show that the DL model accurately segments the data into TCs, ARs, or “the background” at a pixel level. Further, we show how the segmentation results can be used to conduct spatially and temporally precise analytics by quantifying distributions of extreme precipitation conditioned on event types (TC or AR) at regional scales. The key contribution of this work is that it paves the way for DL-based automated, high-fidelity, and highly precise analytics of climate data using a curated expert-labeled dataset – ClimateNet. ClimateNet and the DL-based segmentation method provide several unique capabilities: (i) they can be used to calculate a variety of TC and AR statistics at a fine-grained level; (ii) they can be applied to different climate scenarios and different datasets without tuning as they do not rely on threshold conditions; and (iii) the proposed DL method is suitable for rapidly analyzing large amounts of climate model output. While our study has been conducted for two important extreme weather patterns (TCs and ARs) in simulation datasets, we believe that this methodology can be applied to a much broader class of patterns and applied to observational and reanalysis data products via transfer learning.

54 ENVIRONMENTAL SCIENCES↗

Simulating Study of Premixed Charge Compression Ignition on Light-Duty Diesel Fuel Economy and Emissions Control

We utilize the Powertrain Systems Analysis Toolkit (PSAT) combined with transient engine and aftertreatment component models to simulate the impact of premixed charge compression ignition (PCCI) on the fuel economy and emissions of light-duty (LD) diesel-powered conventional and hybrid electric vehicles (HEVs). Our simulated aftertreatment train consists of a diesel oxidation catalyst (DOC), lean NOx trap (LNT), and catalyzed diesel particulate filter (DPF). The results indicate that utilizing PCCI combustion significantly reduces fuel consumption and tailpipe emissions for the conventional diesel-powered vehicle with NOx and particulate emissions controls. These benefits result from a favorable engine speed-load distribution over the cycle combined with a corresponding reduction in the need to regenerate the LNT and DPF. However, the current PCCI technology appears to offer less potential benefit for diesel HEVs equipped with similar emissions controls. This is because PCCI can only be activated over a relatively small part of the drive cycle. Thus we conclude that future utilization of PCCI in diesel HEVs will require significant extension of the available speed-load range for PCCI and revision of current HEV engine management strategies before significant benefits can be realized.

33 ADVANCED PROPULSION SYSTEMS↗

Fiber-optic diagnostic system for future accelerator magnets

The next generation high energy physics accelerators will require magnetic fields at ~20 T. HTS coils will be an essential component of future accelerator magnets and several efforts are currently dedicated on designing 20 T HTS- LTS hybrid magnets. Among the existing challenges, there is the lack of a robust quench detection system for hybrid magnet technology. Another big challenge is represented by the high number of training quenches required by Nb3Sn magnets to reach performance level. In this framework it is important to find a tool that allow local real-time monitoring of magnet strain and temperature. In this paper, we propose the use of fiber optics sensors for diagnostic and quench detection in future accelerator superconducting magnets. Discrete and distributed fiber optic sensors have demonstrated to be a promising tool. The goal is to instrument hundreds of accelerator superconducting magnets and to move beyond the proof-of-concept level. Significant developments are still needed. Here, we are going to present the most recent results and discuss the most urgent technical developments in order to make those sensors a robust and reliable diagnostic tool for accelerator superconducting magnets over the next 10 year. We foresee that discrete fiber sensors will be a stable diagnostic probe for superconducting magnets over the next 3 to 5 years. More R&D work will be necessary for distributed fibers. The most urgent needs are the increase of sample rate and sensitivity. Close collaboration with vendors will be necessary to improve mechanical properties and fabrication processes in order to produce hundreds of meters of fiber and instrument a large number of accelerator superconducting magnets. Those R&D efforts will last up to 10 years with a founding level that spans between 5-10 M$.

43 PARTICLE ACCELERATORS↗

MO-DE-BRA-01: Flipped Physics Courses Within a Radiologic Technologist Program: Video Production and Long Term Outcomes

Purpose: To determine if the flipped class format has an effect on examination results for a radiologic technologist (RT) program and discuss benefits from creating video resources. Methods: From 2001–2015, students had taken both a radiological physics and quality control (QC) class as a part of their didactic training. In 2005/2006, the creation of videos of didactic lectures and QC test demonstrations allowed for a flip where content was studied at home while exercises and reviews were done in-class. Final examinations were retrospectively reviewed from this timeframe. 12 multiple choice physics questions (MCP) and 5 short answer QC questions (SAQC) were common to pre and post flip exams. The RT program’s ARRT exam scores were also obtained and compared to national averages. Results: In total, 36 lecture videos and 65 quality control videos were created for the flipped content. Data was ∼2.4GB and distributed to students via USB or CD media. For MCP questions, scores improved by 7.9% with the flipped format and significance (Student’s t-test, p<0.05) was found for 3 of the 12 questions. SAQC questions showed improvement by 14.6% and significance was found for 2 of the 5 questions. Student enrollment increased from ∼14 (2001–2004) to ∼23 students (2005–15). Content was continuously added post-flip due to the efficiency of delivery. The QC class in 2003 covered 45 test setups in-class while 65 were covered with video segments in 2014. Flipped materials are currently being repurposed. In 2015, this video content was restructured into an ARRT exam review guide and in 2016, the content was reorganized for fluoroscopy training for physicians. Conclusion: We believe that flipped classes can improve efficiency of content delivery and improve student performance even with an increase in class size. This format allows for flexibility in learning as well as re-use in multiple applications.

07 ISOTOPE AND RADIATION SOURCES↗

Towards Improved Understanding of the Applicability of Uncertainty Forecasts in the Electric Power Industry

Around the world wind energy is starting to become a major energy provider in electricity markets, as well as participating in ancillary services markets to help maintain grid stability. The reliability of system operations and smooth integration of wind energy into electricity markets has been strongly supported by years of improvement in weather and wind power forecasting systems. Deterministic forecasts are still predominant in utility practice although truly optimal decisions and risk hedging are only possible with the adoption of uncertainty forecasts. One of the main barriers for the industrial adoption of uncertainty forecasts is the lack of understanding of its information content (e.g., its physical and statistical modeling) and standardization of uncertainty forecast products, which frequently leads to mistrust towards uncertainty forecasts and their applicability in practice. Our paper aims at improving this understanding by establishing a common terminology and reviewing the methods to determine, estimate, and communicate the uncertainty in weather and wind power forecasts. This conceptual analysis of the state of the art highlights that: (i) end-users should start to look at the forecast's properties in order to map different uncertainty representations to specific wind energy-related user requirements; (ii) a multidisciplinary team is required to foster the integration of stochastic methods in the industry sector. Furthermore, a set of recommendations for standardization and improved training of operators are provided along with examples of best practices.

24 POWER TRANSMISSION AND DISTRIBUTION↗