Search NASA⌕ Search

SEARCH · Search NASA

Results for “Operator 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.

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

At least 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Collaboration and Negotiation

Collaboration and Negotiation is a critical high-level function of an Autonomous Intelligent Cyber-Defense Agent (AICA) that enables communication among agents, central cyber C2, and human operators. Maintaining the Confidentiality, Integrity, and Availability (CIA) triad while achieving mission goals requires stealthy AICA agents to exercise: 1) minimal communication as needed for avoiding detection, 2) verification of information received with possibly limited resources, and 3) active learning during operations to address dynamic conditions. Moreover, negotiations to jointly identify and execute a Course of Action (COA) solution will require building consensus under distributed and/or decentralized multiagent settings with information uncertainties. This chapter presents algorithmic approaches for enabling the collaboration and negotiation function. Strengths and limitations of potential techniques are identified, and a representative example is illustrated. Recommendations for future development are also discussed.

Chatterjee, Samrat↗
DOE OSTI2023

Tuning Phase Lock Loop Controller of Grid Following Inverters by Reinforcement Learning to Support Networked Microgrid Operations

The dynamic operation of networked microgrids leads to varying topological configurations and generator commitments and dispatches. These variations correspond to systems with different electrical characteristics. The fixed control gains of high-speed power electronic devices may result in undesirable system performance when the electrical characteristics change significantly. As such, it is necessary to tune the control gains of power electronics devices to adapt to the changing system characteristics. This paper uses observer-based reinforcement learning to automatically tune the proportional-integral (PI) gains of phase lock loop (PLL) controller of grid-following (GFL) inverters to adapt to the changing system strengths, that would be seen in networked microgrid operations. Simulation results using an operational electric distribution system, modeled as networked microgrids, are presented to demonstrate the need and effectiveness of the proposed adaptive controls.

networked microgrids, reinforcement learning, grid↗
DOE OSTI2022

Combining Deep Learning With Physics Based Features in Explosion-Earthquake Discrimination

This paper combines the power of deep-learning with the generalizability of physics-based features, to present an advanced method for seismic discrimination between earthquakes and explosions. The proposed method contains two branches: a deep learning branch operating directly on seismic waveforms or spectrograms, and a second branch operating on physics-based parametric features. These features are high-frequency P/S amplitude ratios and the difference between local magnitude (ML) and coda duration magnitude (MC). The combination achieves better generalization performance when applied to new regions than models that are developed solely with deep learning. Further, we also examined which parts of the waveform data dominate deep learning decisions (i.e., via Grad-CAM). Such visualization provides a window into the black-box nature of the machine-learning models and offers new insight into how the deep learning derived models use data to make decisions.

58 GEOSCIENCES↗
DOE OSTI2025

A resolution independent neural operator

The Deep operator network (DeepONet) is a powerful yet simple neural operator architecture that utilizes two deep neural networks to learn mappings between infinite-dimensional function spaces. This architecture is highly flexible, allowing the evaluation of the solution field at any location within the desired domain. However, it imposes a strict constraint on the input space, requiring all input functions to be discretized at the same locations; this limits its practical applications. Here, in this work, we introduce a general framework for operator learning from input–output data with arbitrary number and locations of sensors. This begins by introducing a resolution-independent DeepONet (RI-DeepONet), enabling it to handle input functions that are arbitrarily, but sufficiently finely, discretized. To this end, we propose two dictionary learning algorithms to adaptively learn a set of appropriate continuous basis functions, parameterized as implicit neural representations (INRs), from correlated signals defined on arbitrary point cloud data. These basis functions are then used to project arbitrary input function data as a point cloud onto an embedding space (i.e., a vector space of finite dimensions) with dimensionality equal to the dictionary size, which can be directly used by DeepONet without any architectural changes. In particular, we utilize sinusoidal representation networks (SIRENs) as trainable INR basis functions. The introduced dictionary learning algorithms are then used in a similar way to learn an appropriate dictionary of basis functions for the output function data, which defines a new neural operator architecture referred to as the R esolution I ndependent N eural O perator (RINO). In the RINO, the operator learning task simplifies to learning a mapping from the coefficients of input basis functions to the coefficients of output basis functions. We demonstrate the robustness and applicability of RINO in handling arbitrarily (but sufficiently richly) sampled input and output functions during both training and inference through several numerical examples.

Deep operator network (DeepONet)↗
DOE OSTI2021

Evaluation of Extreme Weather Impacts on Utility-scale Photovoltaic Plant Performance in the United States

The global energy system is undergoing significant changes, including a shift in energy generating technologies to more renewable energy sources. However, the dependence of renewable energy sources on local environmental conditions could also increase disruptions in service through exposures to compound, extreme weather events. By fusing three diverse datasets (operations and maintenance tickets, weather data, and production data), this analysis presents a novel methodology to identify and evaluate performance impacts arising from extreme weather events across diverse geographical regions. Text analysis of maintenance tickets identified snow, hurricanes, and storms as the leading extreme weather events affecting photovoltaic plants in the United States. Statistical techniques and machine learning were then implemented to identify the magnitude and variability of these extreme weather impacts on site performance. Impacts varied between event and non-event days, with snow events causing the greatest reductions in performance (54.5%), followed by hurricanes (12.6%) and storms (1.1%). Machine learning analysis identified key features in determining if a day is categorized as low performing, such as low irradiance, geographic location, weather features, and site size. The analysis improves our understanding of compound, extreme weather event impacts on photovoltaic systems, which can inform planning activities, especially as the industry continues to expand into new geographic and climatic regions around the world.

14 SOLAR ENERGY↗
DOE OSTI2024

Separable physics-informed DeepONet: Breaking the curse of dimensionality in physics-informed machine learning

The deep operator network (DeepONet) has shown remarkable potential in solving partial differential equations (PDEs) by mapping between infinite-dimensional function spaces using labeled datasets. However, in scenarios lacking labeled data, the physics-informed DeepONet (PI-DeepONet) approach, which utilizes the residual loss of the governing PDE to optimize the network parameters, faces significant computational challenges, particularly due to the curse of dimensionality. This limitation has hindered its application to high-dimensional problems, making even standard 3D spatial with 1D temporal problems computationally prohibitive. Additionally, the computational requirement increases exponentially with the discretization density of the domain. Here, to address these challenges and enhance scalability for high-dimensional PDEs, we introduce the Separable physics-informed DeepONet (Sep-PI-DeepONet). This framework employs a factorization technique, utilizing sub-networks for individual one-dimensional coordinates, thereby reducing the number of forward passes and the size of the Jacobian matrix required for gradient computations. By incorporating forward-mode automatic differentiation (AD), we further optimize computational efficiency, achieving linear scaling of computational cost with discretization density and dimensionality, making our approach highly suitable for high-dimensional PDEs. We demonstrate the effectiveness of Sep-PI-DeepONet through three benchmark PDE models: the viscous Burgers’ equation, Biot’s consolidation theory, and a parameterized heat equation. Our framework maintains accuracy comparable to the conventional PI-DeepONet while reducing training time by two orders of magnitude. Notably, for the heat equation solved as a 4D problem, the conventional PI-DeepONet was computationally infeasible (estimated 289.35 h), while the Sep-PI-DeepONet completed training in just 2.5 h. These results underscore the potential of Sep-PI-DeepONet in efficiently solving complex, high-dimensional PDEs, marking a significant advancement in physics-informed machine learning.

Neural operator↗
DOE OSTI2022

Dynamic Decarbonization through Autonomous Physics-Centric Deep Learning and Optimization of Building Operations (Abstract only)

This project directly addresses the primary goal of Area of Interest 2 in the CRADA call: to advance optimization-based integrated energy management systems in commercial and residential buildings. Pacific Northwest National Laboratory (PNNL) and its industry partner PassiveLogic aim to accomplish this by reaching three key objectives. First, to ensure a broad impact in the building controls industry, PNNL will extend its open-source library for predictive control synthesis by augmenting its capabilities with data-driven self-learning of building models and auto-calibration of predictive controllers. The effort will focus on building use cases selected in collaboration with PassiveLogic. The team will specifically address the development of methods for data-driven adaptation of building models, investigation of model architectures that best address specific building types, and automated synthesis of differentiable predictive controllers that optimize diverse objectives. Second, PNNL will collaborate with PassiveLogic to integrate the aforementioned methods with PasiveLogic’s advanced controls platform. The collaborative integration effort will inform the developments under the first objective by providing specific data on the attainable performance of model learning on resource-constrained edge computing platforms. This software integration effort will increase the technical maturity of the developed libraries by exploring the use of software integration tools and methods. Third, PNNL and PassiveLogic will work to improve the technology readiness of the developed predictive controllers by testing their performance in relevant test environments, such as high-fidelity simulation, hardware in the loop, and actual test buildings.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Accelerating phase field simulations through a hybrid adaptive Fourier neural operator with U-net backbone

Prolonged contact between a corrosive liquid and metal alloys can cause progressive dealloying. For one such process as liquid-metal dealloying (LMD), phase field models have been developed to understand the mechanisms leading to complex morphologies. However, the LMD governing equations in these models often involve coupled non-linear partial differential equations (PDE), which are challenging to solve numerically. In particular, numerical stiffness in the PDEs requires an extremely refined time step size (on the order of 10 -12 s or smaller). This computational bottleneck is especially problematic when running LMD simulation until a late time horizon is required. This motivates the development of surrogate models capable of leaping forward in time, by skipping several consecutive time steps at-once. In this paper, we propose a U-shaped adaptive Fourier neural operator (U-AFNO), a machine learning (ML) based model inspired by recent advances in neural operator learning. U-AFNO employs U-Nets for extracting and reconstructing local features within the physical fields, and passes the latent space through a vision transformer (ViT) implemented in the Fourier space (AFNO). We use U-AFNOs to learn the dynamics of mapping the field at a current time step into a later time step. We also identify global quantities of interest (QoI) describing the corrosion process (e.g., the deformation of the liquid-metal interface, lost metal, etc.) and show that our proposed U-AFNO model is able to accurately predict the field dynamics, in spite of the chaotic nature of LMD. Most notably, our model reproduces the key microstructure statistics and QoIs with a level of accuracy on par with the high-fidelity numerical solver, while achieving a significant 11, 200 × speed-up on a high-resolution grid when comparing the computational expense per time step. Finally, we also investigate the opportunity of using hybrid simulations, in which we alternate forward leaps in time using the U-AFNO with high-fidelity time stepping. We demonstrate that while advantageous for some surrogate model design choices, our proposed U-AFNO model in fully auto-regressive settings consistently outperforms hybrid schemes.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Sludge Removal: Success and Partnership with T Plant - 20510

U.S. DOE contractor CH2M Hill Plateau Remediation Company (CHPRC) has safely and efficiently managed the removal of highly radioactive sludge from the 105-K West Reactor fuel storage basin (105- KW Basin) near the Columbia River. In mid-September 2019, sludge retrieval was completed after 21 storage containers were safely loaded with highly radioactive sludge and transported approximately 13 miles from the 105-KW Basin to T Plant, located in the 200 West Area on the Central Plateau of the Hanford Site. Sludge removal represents the last major source term reduction necessary before the K West Reactor and fuel storage basin can transition to closure activities in preparation for final deactivation and demolition. The technical complexity of the sludge retrieval process, coupled with the challenging physical and radiological characteristics of the waste, necessitated a methodical and deliberate approach using unique design and operational solutions to safely conduct the work. This challenge was further complicated by the need to plan and conduct the work in two separate facilities, each subject to the controls and requirements of its own nuclear facility safety basis. The project overcame these challenges through creating an integrated team that actively engaged, communicated and coordinated each phase of the project to ensure successful completion. This paper will cover how these key elements of the Sludge Removal Project led to its success: - Integrating project planning and management; - Lessons learned from design and construction; - Preparing for a successful campaign (lessons learned from testing and start-up); - Lessons learned from operating in two separate facilities, transporting sludge containers on the Hanford Site and ensuring compliance with a range of regulatory and safety basis requirements. The establishment of an integrated group spanning two facilities and multiple organizations created a team capable of overcoming the challenges necessary to successfully plan and execute the sludge retrieval mission. Sharing lessons learned from this successful project can enhance the ability of teams across the DOE complex to successfully plan and execute complex projects. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2022

Case Study: NREL Campus Chilled Water Storage Potential: Benchmark Datasets Development and Applications, Task 4 - Use Case Demonstration

The Benchmark Datasets Development and Applications project is a three-year collaboration between the National Renewable Energy Laboratory (NREL), Oak Ridge National Laboratory, Pacific Northwest National Laboratory, and Lawrence Berkeley National Laboratory. The project seeks to collect and curate high-resolution, well-calibrated time series of building operational and indoor/outdoor environmental data, which are crucial to understanding and optimizing building energy efficiency performance and demand flexibility capabilities as well as benchmarking energy algorithms. Project outcomes include approximately twelve high-fidelity building datasets, enhanced data representation tools, and four case studies to illustrate example applications. The goal of these case studies is to define and execute analyses that demonstrate how one or more datasets collected through this project can address a data gap or challenge historically faced by building stakeholders. This technical paper summarizes the findings of one of these case studies, in which we studied the operational efficiencies of the central cooling system at NREL. We looked at three years of data from the three chillers in the Field Test Laboratory Building (FTLB), from 2019 to 2021, to compare equipment operation and demand throughout the time period. Our analysis indicates that all three chillers are operating at or below the optimal loading conditions for most of the operation time, and thus there was no efficiency drop due to loading of the chillers at full capacity. Our recommendation is that no chiller capacity increase is needed; instead, the central plant could benefit from adopting advanced control logics for optimal sequencing of chillers during part load operations. Analysis of adding chilled water thermal storage to the central plant indicated 34% savings in demand cost and 24.5% savings in total cost (energy consumption and demand charge cost). The payback period is estimated to be 11-22 years with an assumed TES cost of $\$$100-$200 per ton. This case study shows how a selected dataset is used to solve a practical building problem - learning the operational status of its components, analyzing the effectiveness of a proposed new technique, and aiding decision-making for the building operations and maintenance team.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS2017

EVA Systems Technology Gaps and Priorities 2017

Performance of Extra-Vehicular Activities (EVA) has been and will continue to be a critical capability for human space flight. Human exploration missions beyond LEO will require EVA capability for either contingency or nominal activities to support mission objectives and reduce mission risk. EVA systems encompass a wide array of products across pressure suits, life support systems, EVA tools and unique spacecraft interface hardware (i.e. EVA Translation Paths and EVA Worksites). In a fiscally limited environment with evolving transportation and habitation options, it is paramount that the EVA community's strategic planning and architecture integration products be reviewed and vetted for traceability between the mission needs far into the future to the known technology and knowledge gaps to the current investments across EVA systems. To ascertain EVA technology and knowledge gaps many things need to be brought together, assessed and analyzed. This includes an understanding of the destination environments, various mission concept of operations, current state of the art of EVA systems, EVA operational lessons learned, and reference advanced capabilities. A combined assessment of these inputs should result in well-defined list of gaps. This list can then be prioritized depending on the mission need dates and time scale of the technology or knowledge gap closure plan. This paper will summarize the current state of EVA related technology and knowledge gaps derived from NASA's Exploration EVA Reference Architecture and Operations Concept products. By linking these products and articulating NASA's approach to strategic development for EVA across all credible destinations an EVA could be done in, the identification of these gaps is then used to illustrate the tactical and strategic planning for the EVA technology development portfolio. Finally, this paper illustrates the various "touch points" with other human exploration risk identification areas including human health and performance.

Johnson, Brian J.↗
DOE OSTI2023

Predictive Analytics for Hydropower Fleet Intelligence

A primary challenge in hydropower industry is the ability to maintain cost-competitiveness, reliability, and security of hydropower assets through evolving power system contexts and aging of the fleet. Maintaining cost-effective and reliable operations under these conditions is expected to require new modernization and maintenance paradigms for changing contexts. Changes in existing practices for O&M will require an understanding of the current state and health of hydropower assets, and the impact of changing paradigms on asset health and reliability. The Hydropower Fleet Intelligence project is developing and evaluating standardized methodologies and analysis tools for data-driven asset reliability and management technologies for hydropower, leading to eventual predictive maintenance planning, repair/replacement decision making, and asset-reliability and cost-optimized operations. A key question is the feasibility of using existing data sets at hydropower facilities to perform assessments of asset reliability. This document uses data from hydropower facilities to assess the potential for using available analytics methods for asset reliability estimates. In addition to reliability assessments, the feasibility of using existing analytics techniques for several other potential applications is discussed. Finally, a case study that a data-driven model is trained to learn nominal operations via vibration data from an asset of a certain plant, and then utilized to identify anomalies on a similar asset from a different plant, highlighting the generic use of proposed Prognostics and Health Management (PHM) approaches.

Yucesan, Yigit↗
NASA NTRS2013

Forecasting Propagation and Evolution of CMEs in an Operational Setting: What Has Been Learned

One of the major types of solar eruption, coronal mass ejections (CMEs) not only impact space weather, but also can have significant societal consequences. CMEs cause intense geomagnetic storms and drive fast mode shocks that accelerate charged particles, potentially resulting in enhanced radiation levels both in ions and electrons. Human and technological assets in space can be endangered as a result. CMEs are also the major contributor to generating large amplitude Geomagnetically Induced Currents (GICs), which are a source of concern for power grid safety. Due to their space weather significance, forecasting the evolution and impacts of CMEs has become a much desired capability for space weather operations worldwide. Based on our operational experience at Space Weather Research Center at NASA Goddard Space Flight Center (http://swrc.gsfc.nasa.gov), we present here some of the insights gained about accurately predicting CME impacts, particularly in relation to space weather operations. These include: 1. The need to maximize information to get an accurate handle of three-dimensional (3-D) CME kinetic parameters and therefore improve CME forecast; 2. The potential use of CME simulation results for qualitative prediction of regions of space where solar energetic particles (SEPs) may be found; 3. The need to include all CMEs occurring within a ~24 h period for a better representation of the CME interactions; 4. Various other important parameters in forecasting CME evolution in interplanetary space, with special emphasis on the CME propagation direction. It is noted that a future direction for our CME forecasting is to employ the ensemble modeling approach.

forecasting↗
NASA NTRSDate not supplied

Insights and Observations From Operating A Laser Communication Terminal on the International Space Station (ISS)

The Integrated Laser Communication Relay Demonstration (LCRD) Low Earth Orbit User Modem and Amplifier Terminal (ILLUMA-T) is the first user platform for NASA’s first laser communication relay, LCRD. ILLUMA-T is hosted on the International Space Station (ISS) as an external payload of the Japanese Experiment Module – Exposed Facility (JEM-EF) situated in slot 3. The LCRD payload is hosted on the STPSat-6 spacecraft, which orbits Earth in a geostationary orbit at position 112 W. The ILLUMA-T-to-LCRD system provides a bidirectional, space-to-ground laser communication relay link transferring data up to 1.2 Gbps utilizing two optical ground stations (OGS), OGS-1 in California and OGS-2 in Hawai’i. This paper describes lessons learned from operating the ILLUMA-T system and examines topics such as the end-to-end system availability, predictive versus actual ephemeris, acquisition adjustments, modem timing and handshaking, the terminal’s physical placement and accommodations, and experiment development, execution, and analysis. The authors of this paper operate both the LCRD and ILLUMA-T laser terminals on a daily basis including pass planning, procedure execution, acquisition analysis, and real-time terminal commanding and telemetry monitoring. LCRD is a joint project involving NASA Goddard Space Flight Center (GSFC), the California Institute of Technology Jet Propulsion Laboratory (JPL), and Massachusetts Institute of Technology Lincoln Laboratory (MIT LL). The ILLUMA-T payload is managed by NASA GSFC in Greenbelt, Maryland whose partners include the ISS program office at NASA’s Johnson Space Center (JSC) in Houston, Texas; the Huntsville Operations Support Center (HOSC) at Marshall Space Flight Center (MSFC) in Huntsville, Alabama; the Glenn Research Center (GRC) in Cleveland, Ohio; and the MIT LL in Lexington, Massachusetts. ILLUMA-T is funded by the Space Communications and Navigation (SCaN) Program at NASA Headquarters in Washington, D.C.

Laser Communications↗
NASA NTRS1990

The Johnson Space Center Management Information Systems (JSCMIS): An interface for organizational databases

The Management Information and Decision Support Environment (MIDSE) is a research activity to build and test a prototype of a generic human interface on the Johnson Space Center (JSC) Information Network (CIN). The existing interfaces were developed specifically to support operations rather than the type of data which management could use. The diversity of the many interfaces and their relative difficulty discouraged occasional users from attempting to use them for their purposes. The MIDSE activity approached this problem by designing and building an interface to one JSC data base - the personnel statistics tables of the NASA Personnel and Payroll System (NPPS). The interface was designed against the following requirements: generic (use with any relational NOMAD data base); easy to learn (intuitive operations for new users); easy to use (efficient operations for experienced users); self-documenting (help facility which informs users about the data base structure as well as the operation of the interface); and low maintenance (easy configuration to new applications). A prototype interface entitled the JSC Management Information Systems (JSCMIS) was produced. It resides on CIN/PROFS and is available to JSC management who request it. The interface has passed management review and is ready for early use. Three kinds of data are now available: personnel statistics, personnel register, and plan/actual cost.

Bishop, Peter C.↗
DOE OSTI2022

Machine Learning Pattern Recognition Algorithm With Applications to Coherent Laser Combination

Herein we analyze a new kind of machine learning algorithm designed to feedback stabilize coherently combined lasers. This algorithm learns differential, rather than absolute, values of action in phase space, in order to facilitate learning on initially unstable systems. Experiments have shown that this approach can control small-scale spatial beam combination with high stability. In this paper we analyze the algorithm's performance and limitations in depth, showing that it can continuously learn during operation in order to track changes. Using simulation, we extend the application to temporal combination, and show that it scales to more complex instances by combining 81 beams.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2011

Lunar Reconnaissance Orbiter (LRO) Thermal On-Orbit Performance

The Lunar Reconnaissance Orbiter (LRO) was launched on June 18, 2009. The nominal mission ended on September 15, 2010 and LRO is now on a four-year extended mission. The LRO performances in hot and cold cases are compared to pre-launch analysis predicts, and operational lessons learned are discussed. One instrument has required tighter-than-anticipated thermal control, and two others have frequently requested unanticipated calibration maneuvers that had to be evaluated for their thermal performance. A series of off nadir thermal analyses of the entire orbiter were performed prior to launch, and these predictions are compared to actual maneuvers, with a discussion of the process by which maneuvers can be rapidly evaluated for thermal concerns. On December 21st, 2010, LRO experienced its first severe Lunar Eclipse. Operationally, this required the Spacecraft to pre-heat its main avionics panel in order to minimize control heater power during the period when the Earth blocks the sun from the moon. The operational design and in-flight performance are summarized.

Baker, Charles↗
NASA NTRS2018

Cognitive Anti-Jamming Satellite-To-Ground Communications on NASA's SCaN Testbed

Machine learning aided cognitive anti-jamming communications is designed, developed and demonstrated on a live satellite-to-ground link. A wideband autonomous cognitive radio (WACR) is designed and implemented as a hardware-in-the- loop (HITL) prototype. The cognitive engine (CE) of the WACR is implemented on a PC while the software-defined radio (SDR) platform utilized two different radios for spectrum sensing and actual communications. The cognitive engine performs spectrum knowledge acquisition over the complete spectrum range available for the SATCOM system operation and learns an anti-jamming communications protocol to avoid both intentional jammers and inadvertent interferers using reinforcement learning. When the current satellite-to-ground link is jammed, the cognitive engine of the ground receiver directs the satellite transmitter to switch to a new channel that is predicted to be jammer-free for the longest possible duration. The end-to-end, closed-loop system was tested on the NASA Space Communications and Navigation (SCaN) Testbed on the International Space Station (ISS). The experimental results demonstrated the feasibility of satellite-to-ground cognitive anti-jamming communications along with excellent anti-jamming capability of machine learning aided cognitive protocols against several different types of jammers.

Jayaweera, Sudharman K.↗