SEARCH · Search NASA

Results for “Continual Learning”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 721 records · Page 40

All sources NASA NTRS DOE OSTI

DOE OSTI2026

Accelerating kinetic plasma simulations with machine-learning-generated initial conditions

Computational models of plasma technologies often solve for the system operating conditions by time-stepping an initial value problem to a quasi-steady solution. However, the strongly nonlinear and multi-timescale nature of plasma dynamics often necessitate millions, or even hundreds of millions, of steps to reach convergence, reducing the effectiveness of these simulations for computer-aided engineering. We consider acceleration of kinetic plasma simulations via data-driven machine-learning-generated initial conditions, which initialize the simulations close to their final quasi-steady-state, thereby reducing the number of steps to reach convergence. Three machine-learning models are developed to predict the density and ion kinetic profiles of capacitively coupled plasma discharges relevant to the microelectronics industry. The models are trained on kinetic simulations over a range of device operating frequencies and pressures. Best performance was observed when simulations were initialized with ion kinetic profiles generated by a convolutional neural network, reducing the mean number of steps to reach convergence by 17.1× when compared to initialization with a zero-dimensional global model. We also outline a workflow for continuous data-driven model improvement and simulation speedup, with the aim of generating sufficient data for full device digital twins.

Artificial neural networks↗

DOE OSTI2024

Accelerating science: The usage of commercial clouds in ATLAS Distributed Computing

The ATLAS experiment at CERN is one of the largest scientific machines built to date and will have ever growing computing needs as the Large Hadron Collider collects an increasingly larger volume of data over the next 20 years. ATLAS is conducting R&D projects on Amazon Web Services and Google Cloud as complementary resources for distributed computing, focusing on some of the key features of commercial clouds: lightweight operation, elasticity and availability of multiple chip architectures. The proof of concept phases have concluded with the cloud-native, vendoragnostic integration with the experiment’s data and workload management frameworks. Google Cloud has been used to evaluate elastic batch computing, ramping up ephemeral clusters of up to O(100k) cores to process tasks requiring quick turnaround. Amazon Web Services has been exploited for the successful physics validation of the Athena simulation software on ARM processors. We have also set up an interactive facility for physics analysis allowing endusers to spin up private, on-demand clusters for parallel computing with up to 4 000 cores, or run GPU enabled notebooks and jobs for machine learning applications. The success of the proof of concept phases has led to the extension of the Google Cloud project, where ATLAS will study the total cost of ownership of a production cloud site during 15 months with 10k cores on average, fully integrated with distributed grid computing resources and continue the R&D projects.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2026

Monte Carlo Event Generation with Continuous Normalizing Flows

We apply continuous normalizing flows trained with the flow matching method to the problem of phase-space sampling in Monte Carlo event generation for high-energy collider physics. Focusing on lepton-pair and top-quark pair production with multiple jets, the two computationally most expensive processes at the Large Hadron Collider, we train helicity-conditioned continuous normalizing flows to remap the random numbers used in matrix element evaluation. Compared to standard methods, we achieve unweighting efficiency improvements by factors of up to 184 and 25 for the two processes at their respective highest jet number, at the cost of an increased evaluation time. When combining the advantages of continuous normalizing flows with the fast evaluation times of coupling-layer-based flows, using the RegFlow approach, we find parton-level unweighted event generation walltime gains of about a factor of 10 at the highest jet numbers. These substantial gains highlight the promise of samplers based on machine learning for next-generation collider experiments.

Bothmann, Enrico [CERN; Gottingen U.] (ORCID:00000↗

DOE OSTI2025

Synopsis of NREL's Automated Mobility District (AMD) Research Program and Associated Publications

An automated mobility district (AMD) envisions a system of integrated mobility options that serves major activity centers such as campuses, central business districts, and large medical facilities. The National Renewable Energy Laboratory (NREL) has been investigating the implementation prospects for fully automated passenger transport systems that are deployed to operate within dense urban settings. This document provides a synopsis of findings revealed over the last three phases of work, which have yielded insights into the creation and management of AMDs anticipated to use automated vehicle (AV) technology over the next decade. Phase I and Phase II tracked the deployment and lessons learned from 10 early-stage demonstrations of automated shuttle deployments, and their associated insights into the challenges for automated driving systems to achieve safe operations within district-scale deployments. Phase III began in-depth investigations of critical subsystem components, as automation, electrification, and on-demand service continue to converge within initial AMD operations. These directed studies focus on elements of electrification, curbfront/station management, the role of infrastructure sensing, and overall integration of AMD safety management in central, simultaneous coordination of multiple AMD fleets. Future research in AMDs includes systems engineering methodology (more frequently referred to as "digital twins") for planning, design, testing, and ongoing operation of AMDs; location (or co-location) of management functions; and human supervision and passenger communications for safety and security in unattended vehicles. The synopsis references the foundational research products (papers and presentations) that have been published through conference proceedings, journal articles, and NREL reports.

33 ADVANCED PROPULSION SYSTEMS↗

NASA NTRS1992

USRA/RIACS

The Research Institute for Advanced Computer Science (RIACS) was established by the Universities Space Research Association (USRA) at the NASA Ames Research Center (ARC) on 6 June 1983. RIACS is privately operated by USRA, a consortium of universities with research programs in the aerospace sciences, under a cooperative agreement with NASA. The primary mission of RIACS is to provide research and expertise in computer science and scientific computing to support the scientific missions of NASA ARC. The research carried out at RIACS must change its emphasis from year to year in response to NASA ARC's changing needs and technological opportunities. A flexible scientific staff is provided through a university faculty visitor program, a post doctoral program, and a student visitor program. Not only does this provide appropriate expertise but it also introduces scientists outside of NASA to NASA problems. A small group of core RIACS staff provides continuity and interacts with an ARC technical monitor and scientific advisory group to determine the RIACS mission. RIACS activities are reviewed and monitored by a USRA advisory council and ARC technical monitor. Research at RIACS is currently being done in the following areas: Parallel Computing; Advanced Methods for Scientific Computing; Learning Systems; High Performance Networks and Technology; Graphics, Visualization, and Virtual Environments.

Oliger, Joseph↗

NASA NTRSDate not supplied

Having a Come-Apart: Lessons Learned from Additively Manufactured Hardware Failures

NASA has been engaged with additively manufactured (AM) process and component development since the 2000’s. AM offers various technical advantages, such as enhanced hardware design complexity, part consolidation, and processing of novel alloys in addition to programmatic advantages for reduction in processing time and cost. The focus of much of the AM development at NASA has been to mature the various processes, characterize material properties, develop standards, produce demonstrator parts, and integrate AM hardware in liquid rocket engines. These aspects have been demonstrated through process and design iterations using a methodical characterization, test-fail-fix cycles, as well as application and dissemination of lessons learned. In addition to these fundamental demonstrations of the AM process and hardware development, alloys that provide performance advantages in the high temperature and high-pressure environments have been matured for use in rocket engines. These environments are challenging for any alloy and any design, and the AM process is required to fully meet the intended design requirements. The importance of proper AM process was made evident in the failure of a Laser Powder Bed Fusion (L-PBF) copper-alloy combustion chamber during a hot-fire test due to a degraded material quality resulted from an AM process issue. The hot-fire test aimed to demonstrate high duty cycle under a risk-tolerant development project, where consequences of component failure would be minimal. However, the unintentional component failure emphasized the necessity of robust material characterization and rigorous process control procedures for the safe use of AM components in critical applications. In part, such concerns motivate the AM certification approach that NASA has recently adopted in NASA-STD-6030 “Additive Manufacturing Requirements for Spaceflight Systems”. This presentation provides an overview of the previously mentioned failure, a discussion on the evaluation of the failed chamber and supplemental chambers produced at the same time, a representative material samples that included intentional build witness lines, and a summary of the key results and recommendations from the evaluations. NASA continues to approach AM processes and designs with a level of risk and acceptance of failures that is appropriate for the project objectives, with the overall goal of safe implementation of AM technology and transferring AM technology into commercial space applications. The objective of this presentation is to provide awareness to the community working critical and non-critical AM components and the lessons learned on proper implementation of AM.

Additive Manufacturing↗

DOE OSTI2024

Physical discovery in representation learning via conditioning on prior knowledge

Recent advances in electron, scanning probe, optical, and chemical imaging and spectroscopy yield bespoke data sets containing the information of structure and functionality of complex systems. In many cases, the resulting data sets are underpinned by low-dimensional simple representations encoding the factors of variability within the data. The representation learning methods seek to discover these factors of variability, ideally further connecting them with relevant physical mechanisms. However, generally, the task of identifying the latent variables corresponding to actual physical mechanisms is extremely complex. Here, we present an empirical study of an approach based on conditioning the data on the known (continuous) physical parameters and systematically compare it with the previously introduced approach based on the invariant variational autoencoders. The conditional variational autoencoder (cVAE) approach does not rely on the existence of the invariant transforms and hence allows for much greater flexibility and applicability. Interestingly, cVAE allows for limited extrapolation outside of the original domain of the conditional variable. However, this extrapolation is limited compared to the cases when true physical mechanisms are known, and the physical factor of variability can be disentangled in full. We further show that introducing the known conditioning results in the simplification of the latent distribution if the conditioning vector is correlated with the factor of variability in the data, thus allowing us to separate relevant physical factors. We initially demonstrate this approach using 1D and 2D examples on a synthetic data set and then extend it to the analysis of experimental data on ferroelectric domain dynamics visualized via piezoresponse force microscopy.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

Network Anomaly Detection in Distributed Edge Computing Infrastructure

As networks continue to grow in complexity and scale, detecting anomalies has become increasingly challenging, particularly in diverse and geographically dispersed environments. Traditional approaches often struggle with managing the computational burden associated with analyzing large-scale network traffic to identify anomalies. This paper introduces a distributed edge computing framework that integrates federated learning with Apache Spark and Kubernetes to address these challenges. We hypothesize that our approach, which enables collaborative model training across distributed nodes, significantly enhances the detection accuracy of network anomalies across different network types. We show that by leveraging distributed computing and containerization technologies, our framework not only improves scalability and fault tolerance but also achieves superior detection performance compared to state-of-the-art methods. Extensive experiments on the UNSW-NB15 and ROAD datasets validate the effectiveness of our approach, demonstrating statistically significant improvements in detection accuracy and training efficiency over baseline models, as confirmed by MannWhitney U and Kolmogorov-Smirnov tests (p<0.05).

Marfo, William [University of Texas at El Paso,Dep↗

DOE OSTI2024

Late-Stage Research Development and Demonstration Sub-activities Updates – FY24 Q3

Oak Ridge National Laboratory (ORNL), in collaboration with the Pacific Northwest National Laboratory (PNNL), the National Renewable Energy Laboratory (NREL), the Lawrence Berkeley National Laboratory (LBNL), and the Hummingbird Firm (a specialized consulting firm focused on promoting diversity, equity, and inclusion considerations), has initiated a national initiative known as the Heat Pump (HP) and Heat Pump Water Heater (HPWH) Field Validation Partnership. This effort involves active participation from numerous critical entities involved in research and market transformation within the field. The ORNL team is responsible for leading Late-Stage Research Development and Demonstration (LSRDD) among the four different topics. The overall outcomes of this project will be: (1) A structured Field Validation Partnership between DOE, the national labs, research, implementation, and market transformation organizations. This will result in unique way to coordinate field validation plans and collect relevant data from around the country into the HP and HPWH Field Validation Database. (2) The Field Validation Partnership will result in a continuous stream of information between DOE and the major industry players in the space of HPs and HPWHs. If desired, DOE could use this information to inform roadmaps related to HP and HPWH market adoption and research going forward. (3) The structure of this Partnership provides a mechanism for sharing lessons learned directly between Late-Stage RD&D, Building Integration Barriers, Regional Market and Policy and Workforce Development efforts. The result will be training content that is well-reviewed by the Partnership which will lead to a workforce that meets the industry’s quality and workforce supply demands. (4) The structure of this Partnership also provides an opportunity for regions to share lessons learned on policy and market transformation with each other through the Market and Policy core Committee. This report includes an update in Late-Stage Research Development and Demonstration.

99 GENERAL AND MISCELLANEOUS↗

NASA NTRS2012

A Systematic Approach for Real-Time Operator Functional State Assessment

A task overload condition often leads to high stress for an operator, causing performance degradation and possibly disastrous consequences. Just as dangerous, with automated flight systems, an operator may experience a task underload condition (during the en-route flight phase, for example), becoming easily bored and finding it difficult to maintain sustained attention. When an unexpected event occurs, either internal or external to the automated system, the disengaged operator may neglect, misunderstand, or respond slowly/inappropriately to the situation. In this paper, we discuss an approach for Operator Functional State (OFS) monitoring in a typical aviation environment. A systematic ground truth finding procedure has been designed based on subjective evaluations, performance measures, and strong physiological indicators. The derived OFS ground truth is continuous in time compared to a very sparse estimation of OFS based on an expert review or subjective evaluations. It can capture the variations of OFS during a mission to better guide through the training process of the OFS assessment model. Furthermore, an OFS assessment model framework based on advanced machine learning techniques was designed and the systematic approach was then verified and validated with experimental data collected in a high fidelity Boeing 737 simulator. Preliminary results show highly accurate engagement/disengagement detection making it suitable for real-time applications to assess pilot engagement.

Zhang, Guangfan↗

NASA NTRS2013

Estimating Global Precipitation for Science and Application

Over the past two decades there has been vigorous development in the satellite assets and the algorithms necessary to estimate precipitation around the globe. In particular the highly successful joint NASAJAXA Tropical Rainfall Measuring Mission (TRMM) and the upcoming Global Precipitation Measurement (GPM) mission, also joint between NASA and JAXA, have driven these issues. At the same time, the long-running Global Precipitation Climatology Project (GPCP) continues to extend a stable, climate-oriented view of global precipitation. This talk will provide an overview of these projects and the wider international community of precipitation datasets, sketch plans for next-generation products, and provide some examples of the best use for the different products. One key lesson learned is that different data sets are needed to address the variety of issues that need precipitation data, including detailed 3-D views of hurricanes, flash flood forecasting, drought analysis, and global change.

global precipitation↗

NASA NTRS2020

Ground System Milestones: 20 Years of Earth Observing Satellite Mission Operations

Aqua, Aura, and Terra are Earth observing satellites that serve in the Earth Observing System (EOS). Terra, the flagship mission of the EOS, was launched in 1998. Aqua was launched in 2002 and Aura was launched in 2004. Each spacecraft was designed for a 6 year lifespan. Currently, each spacecraft meets all mission and science requirements. As technology advances, the onboard hardware and software remain the same, however, our ground system progresses. In the past twenty years, many changes have taken place. Our goal is to describe the life of the ground system from launch, to present day, and touch on our future plans. The focus of our presentation will be how EOS has and continues to modernize our ground system for the future. We would like to share our experiences and give insight into the operational challenges of a 20 year old ground system. Our hope is for future mission operators and those maintaining potential age defying satellites to use our past experiences and lessons learned for future operational concepts.

Barrett, Grant↗

NASA NTRS2020

COSMIC: Content-based Onboard Summarization to Monitor Infrequent Change

Interplanetary exploration occurs at vast distancesthat severely limit communication bandwidth to spacecraft exploringother planets. It is possible to collect much morescientific data than can ever be downlinked given current communicationcapabilities. Therefore, we are developing a systemcalled COSMIC (Content-based Onboard Summarization toMonitor Infrequent Change) that will opportunistically analyzedata onboard a Mars orbiter to alert scientists when meaningfulchanges have occurred. COSMIC will allow future spacecraftto continuously collect data to search for rare, transient phenomenasuch as fresh impacts or seasonally changing polarlandforms under a constrained downlink budget. In this paper,we describe the overall goals and architecture of COSMIC,plans to enable specific scientific studies, label acquisition toenable supervised approaches to surface landform classification,a new machine learning evaluation framework for analyzingthe trade-offs between classifier accuracy and computationalrequirements, and lessons learned about constraints that COSMICwill face operating onboard a spacecraft. In particular, wediscuss design considerations surrounding computational andstorage constraints, change detection strategies, and localizingdetected landforms of interest within a global coordinate frame.Finally, we describe challenges and open research questions thatmust be addressed prior to deploying COSMIC.

Trockman, Asher↗

DOE OSTI2025

High-Temperature Aquifer Thermal Energy Storage (HT-ATES) Projects in Germany and the Netherlands—Review and Lessons Learned

Aquifer thermal energy storage (ATES) is a concept that can help to address heating and cooling needs through the use of the subsurface as a seasonal thermal energy storage (STES) system. Over 2800 ATES systems have been deployed with storage temperatures typically below 25 °C and only a few with higher temperatures (>40 °C), which would increase the energy density and utility of the stored thermal fluids. Until now, only a few high-temperature aquifer thermal energy storage (HT-ATES) projects have been initiated and are still in operation. These HT-ATES projects have encountered a range of technical and non-technical challenges. This study reviews ten such projects: four in Germany and six in the Netherlands. The non-technical issues include public acceptance, a lack of regulatory framework for these systems, managing overlapping uses of the subsurface, managing changes with the providers and off-takers of thermal energy, and obtaining financing to implement these projects. Common technical issues include geological factors such as incomplete characterization of the subsurface and reservoir heterogeneity; geochemical issues such as mineral scaling, corrosion, and biofouling; lower than expected thermal recovery; and issues with system design and reliability. This review highlights benefits and challenges faced by HT-ATES projects with the goal to use the lessons learned to improve the siting, design, development, and operation of such systems. Recommendations include improved initial subsurface site characterization, use of coupled process models to optimize system design and predict system performance, cascaded uses of stored thermal energy to better utilize the stored heat, monitoring networks to provide feedback on system performance, and expanded system scale to allow for continued operation even when maintenance of some system components is required. Techno-economic modeling and risk analysis could be used to optimize such HT-ATES project design and identify key factors that will affect sustained economic viability. In addition, design flexibility is important for these systems to allow for changing conditions regarding the supply and demand of thermal energy. Adopting these findings should improve the performance and reduce the risks for future HT-ATES projects worldwide.

15 - GEOTHERMAL ENERGY↗

NASA NTRS1994

Lessons Learned During the Integration Phase of the NASA IN-STEP Cryo System Experiment

The Cryo System Experiment (CSE), a NASA In-Space Technology Experiments Proram (IN-STEP) Class D Flight Experiment, was developed by Hughes Aircraft Company (Hughes) to validate in zero-g space a 65 K cryogenic system for focal planes, optics, instruments, or other equipment (gramma-ray spectrometers and infrared and submillimeter imaging instruments) that require continuous cryogenic cooling.

cryogenic launch-vibration restraints thermal stra↗

DOE OSTI2026

Lyapunov-Based Iterative Learning of Regions of Attraction for Autonomous Systems

This paper proposes a novel algorithm for estimating the region of attraction of equilibrium points for nonlinear discrete-time autonomous systems. The method iteratively expands an initial estimate of the region of attraction by constructing unions of sublevel sets of learned functions parametrized as neural networks. Unlike conventional techniques that rely on a single global Lyapunov function, the proposed approach provides a collection of local Lyapunov-like functions, enabling richer representations and potentially larger region of attraction estimates. These functions are trained using sampled state-space data, and their Lipschitz continuity ensures that desirable properties extend beyond the training samples. The devised strategy is tested via numerical simulations, demonstrating the effectiveness of the proposed approach.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Distributed Stochastic Optimization of a Neural Representation Network for Time-Space Tomography Reconstruction

4D time-space reconstruction of dynamic events or deforming objects using X-ray computed tomography (CT) is an important inverse problem in non-destructive evaluation. Conventional back-projection based reconstruction methods assume that the object remains static for the duration of several tens or hundreds of X-ray projection measurement images (reconstruction of consecutive limited-angle CT scans). However, this is an unrealistic assumption for many in-situ experiments that causes spurious artifacts and inaccurate morphological reconstructions of the object. To solve this problem, we propose to perform a 4D time-space reconstruction using a distributed implicit neural representation (DINR) network that is trained using a novel distributed stochastic training algorithm. Our DINR network learns to reconstruct the object at its output by iterative optimization of its network parameters such that the measured projection images best match the output of the CT forward measurement model. Here, we use a forward measurement model that is a function of the DINR outputs at a sparsely sampled set of continuous valued 4D object coordinates. Unlike previous neural representation architectures that forward and back propagate through dense voxel grids that sample the object's entire time-space coordinates, we only propagate through the DINR at a small subset of object coordinates in each iteration resulting in an order-of-magnitude reduction in memory and compute for training. DINR leverages distributed computation across several compute nodes and GPUs to produce high-fidelity 4D time-space reconstructions. We use both simulated parallel-beam and experimental cone-beam X-ray CT datasets to demonstrate the superior performance of our approach.

36 MATERIALS SCIENCE↗

DOE OSTI2026

Lyapunov-Based Iterative Learning of the Region of Attraction for Autonomous Systems

This presentation introduces a novel algorithm for estimating the region of attraction of equilibrium points for nonlinear discrete-time autonomous systems. The method iteratively expands an initial estimate of the region of attraction by constructing unions of sublevel sets of learned functions parametrized as neural networks. Unlike conventional techniques that rely on a single global Lyapunov function, the proposed approach provides a collection of local Lyapunov-like functions, enabling richer representations and potentially larger region of attraction estimates. These functions are trained using sampled state-space data, and their Lipschitz continuity ensures that desirable properties extend beyond the training samples. The devised strategy is tested via numerical simulations, demonstrating the effectiveness of the proposed approach.

97 MATHEMATICS AND COMPUTING↗