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 487 records · Page 27
All sourcesNASA NTRSDOE OSTI
NASA NTRS2004

Aerospace Meteorology Lessons Learned Relative to Aerospace Vehicle Design and Operations

Aerospace Meteorology came into being in the 1950s as the development of rockets for military and civilian usage grew in the United States. The term was coined to identify those involved in the development of natural environment models, design/operational requirements, and environment measurement systems to support the needs of aerospace vehicles, both launch vehicles and spacecraft. It encompassed the atmospheric environment of the Earth, including Earth orbit environments. Several groups within the United States were active in this area, including the Department of Defense, National Aeronautics and Space Administration, and a few of the aerospace industry groups. Some aerospace meteorology efforts were similar to those being undertaken relative to aviation interests. As part of the aerospace meteorology activities a number of lessons learned resulted that produced follow on efforts which benefited from these experiences, thus leading to the rather efficient and technologically current descriptions of terrestrial environment design requirements, prelaunch monitoring systems, and forecast capabilities available to support the development and operations of aerospace vehicles.

Vaughan, William W.↗
DOE OSTI2021

GOOML Big Kahuna Forecast Modeling and Genetic Optimization Files

This submission includes example files associated with the Geothermal Operational Optimization using Machine Learning (GOOML) Big Kahuna fictional power plant, which uses synthetic data to model a fictional power plant. A forecast was produced using the GOOML data model framework and fictional input data, and a genetic optimization is included which determines optimal flash plant parameters. The inputs and outputs associated with the forecast and genetic optimization are included. The input and output files consist of data, configuration files, and plots. A link to the Physics-Guided Neural Networks (phygnn) GitHub repository is also included, which augments a traditional neural network loss function with a generic loss term that can be used to guide the neural network to learn physical or theoretical constraints. phygnn is used by the GOOML framework to help integrate its machine learning models into the relevant physics and engineering applications. Note that the data included in this submission are intended to provide a demonstration of GOOML's capabilities. Additional files that have not been released to the public are needed for users to run these models and reproduce these results. Units can be found in the readme data resource.

15 GEOTHERMAL ENERGY↗
DOE OSTI2024

Updates and Lessons Learned from NuMI Beamline at Fermilab

The Neutrinos at the Main Injector (NuMI) beamline at Fermilab generates an intense muon neutrino beam for the NOvA (NuMI Off-axis 𝜈𝑒 Appearance) long baseline neutrino experiment. Over the years, the NuMI beamline has been pivotal in advancing neutrino physics, providing invaluable data and insights. This presentation offers updates and a comprehensive review of the lessons learned from the operation, maintenance, and monitoring of the NuMI beamline. Key topics include the optimization of beam performance, challenges in maintaining beamline stability, and proposed Machine Learning implementations to enhance monitoring. The talk aims to share best practices and provide a roadmap for future beamline projects, including the Long-Baseline Neutrino Facility (LBNF).

Wickremasinghe, Athula↗
NASA NTRS2016

Health Monitoring Survey of Bell 412EP Transmissions

Health and usage monitoring systems (HUMS) use vibration-based Condition Indicators (CI) to assess the health of helicopter powertrain components. A fault is detected when a CI exceeds its threshold value. The effectiveness of fault detection can be judged on the basis of assessing the condition of actual components from fleet aircraft. The Bell 412 HUMS-equipped helicopter is chosen for such an evaluation. A sample of 20 aircraft included 12 aircraft with confirmed transmission and gearbox faults (detected by CIs) and eight aircraft with no known faults. The associated CI data is classified into "healthy" and "faulted" populations based on actual condition and these populations are compared against their CI thresholds to quantify the probability of false alarm and the probability of missed detection. Receiver Operator Characteristic analysis is used to optimize thresholds. Based on the results of the analysis, shortcomings in the classification method are identified for slow-moving CI trends. Recommendations for improving classification using time-dependent receiver-operator characteristic methods are put forth. Finally, lessons learned regarding OEM-operator communication are presented.

mechanical components↗
NASA NTRSDate not supplied

Artemis Deep Space Habitation: Enabling a Sustained Human Presence on the Moon and Beyond

As NASA and its partners’ capabilities for human exploration of deep space continue to mature, so too does its roadmap toward a sustained crewed presence on the surface of the Moon and eventual human missions to Mars. The first launch of the Space Launch System and Orion crew vehicle, the contract award for the first demonstrations of a Human Landing System, and the beginning of construction on the initial elements of the lunar Gateway have marked major milestones toward NASA’s near-term exploration goals: a long-duration outpost in orbit around the Moon and the next footsteps on the lunar surface. At the same time, NASA is in the early phases of planning the capabilities that will be needed for long-term exploration. Among the common elements that will be required by long-duration stays on the lunar surface, transit to Mars, and Martian surface expeditions will be new habitats unlike any flown to date. NASA is currently working on development of both architectures for those habitats and on the technological advancements that will enable them, with an eye toward systems that will not only extend mission operations but also provide for living quarters that will keep the crew happy and healthy throughout their expeditions. Beyond the Gateway habitation needs, these capabilities will need to be defined and advanced to support the initial lunar surface missions and to prepare for human missions to the Mars system. The Surface Habitat is the current concept in consideration to serve as this initial surface habitat that will extend the crew mission durations. It will provide 30-to-60-day habitability for a crew of up to four allowing for the astronauts to explore farther and longer on each visit to the lunar surface. NASA is also currently reviewing opportunities to use current or near-term in-space habitation systems as proving grounds or precursors for keeping astronauts safe and healthy during future transits to Mars. Already, the International Space Station (ISS) is being used for implementation of next-generation life-support systems that will inform those used in exploration habitats, and the operations approach for ISS is providing lessons-learned for future science operations around or on the Moon.1U.S. Government work not protected by U.S. copyright While a suite of habitation concepts is currently under study within NASA, the agency is also working closely with U.S. industry through the Next Space Technologies for Exploration Partnerships (NextSTEP) activity to understand their concepts for commercially provided habitation capabilities as well as close coordination with international partners to understand their desires for in-space and surface habitation. This paper will provide a status of these concepts and partnership activities as well as potential future technology and architecture development paths.

Habitat↗
DOE OSTI2024

NREL ARIES MW Electrolyzer Integration

Integration and commissioning of ARIES MW Electrolyzer system is complete, and several integration and operation lessons have been learned. An outline of the ARIES MW Electrolyzer system is presented. Lessons learned in both integration of the units and operations of the electrolyzer are also covered.

ARIES↗
NASA NTRS2013

Downscaling Satellite Precipitation with Emphasis on Extremes: A Variational 1-Norm Regularization in the Derivative Domain

The increasing availability of precipitation observations from space, e.g., from the Tropical Rainfall Measuring Mission (TRMM) and the forthcoming Global Precipitation Measuring (GPM) Mission, has fueled renewed interest in developing frameworks for downscaling and multi-sensor data fusion that can handle large data sets in computationally efficient ways while optimally reproducing desired properties of the underlying rainfall fields. Of special interest is the reproduction of extreme precipitation intensities and gradients, as these are directly relevant to hazard prediction. In this paper, we present a new formalism for downscaling satellite precipitation observations, which explicitly allows for the preservation of some key geometrical and statistical properties of spatial precipitation. These include sharp intensity gradients (due to high-intensity regions embedded within lower-intensity areas), coherent spatial structures (due to regions of slowly varying rainfall),and thicker-than-Gaussian tails of precipitation gradients and intensities. Specifically, we pose the downscaling problem as a discrete inverse problem and solve it via a regularized variational approach (variational downscaling) where the regularization term is selected to impose the desired smoothness in the solution while allowing for some steep gradients(called 1-norm or total variation regularization). We demonstrate the duality between this geometrically inspired solution and its Bayesian statistical interpretation, which is equivalent to assuming a Laplace prior distribution for the precipitation intensities in the derivative (wavelet) space. When the observation operator is not known, we discuss the effect of its misspecification and explore a previously proposed dictionary-based sparse inverse downscaling methodology to indirectly learn the observation operator from a database of coincidental high- and low-resolution observations. The proposed method and ideas are illustrated in case studies featuring the downscaling of a hurricane precipitation field.

Hurricanes↗
DOE OSTI2025

Physics-informed latent neural operator for real-time predictions of time-dependent parametric PDEs

Deep operator network (DeepONet) has shown significant promise as surrogate models for systems governed by partial differential equations (PDEs), enabling accurate mappings between infinite-dimensional function spaces. However, when applied to systems with high-dimensional input-output mappings arising from large numbers of spatial and temporal collocation points, these models often require heavily overparameterized networks, leading to long training times. Latent DeepONet addresses some of these challenges by introducing a two-step approach: first learning a reduced latent space using a separate model, followed by operator learning within this latent space. While efficient, this method is inherently data-driven and lacks mechanisms for incorporating physical laws, limiting its robustness and generalizability in data-scarce settings. Here, in this work, we propose PI-Latent-NO, a physics-informed latent neural operator framework that integrates governing physics directly into the learning process. Our architecture features two coupled DeepONets trained end-to-end: a Latent-DeepONet that learns a low-dimensional representation of the solution, and a Reconstruction-DeepONet that maps this latent representation back to the physical space. By embedding PDE constraints into the training via automatic differentiation, our method eliminates the need for labeled training data and ensures physics-consistent predictions. The proposed framework is both memory and compute-efficient, exhibiting near-constant scaling with problem size and demonstrating significant speedups over traditional physics-informed operator models. We validate our approach on a range of parametric PDEs, showcasing its accuracy, scalability, and suitability for real-time prediction in complex physical systems.

Latent representations↗
DOE OSTI2019

GMLS-Nets: A Framework for Learning from Unstructured Data

Data fields sampled on irregularly spaced points arise in many applications in the sciences and engineering. For regular grids, Convolutional Neural Networks (CNNs) have been successfully used to gaining benefits from weight sharing and invariances. We generalize CNNs by introducing methods for data on unstructured point clouds based on Generalized Moving Least Squares (GMLS). GMLS is a non-parametric technique for estimating linear bounded functionals from scattered data, and has recently been used in the literature for solving partial differential equations. By parameterizing the GMLS estimator, we obtain learning methods for operators with unstructured stencils. In GMLS-Nets the necessary calculations are local, readily parallelizable, and the estimator is supported by a rigorous approximation theory. We show how the framework may be used for unstructured physical data sets to perform functional regression to identify associated differential operators and to regress quantities of interest. The results suggest the architectures to be an attractive foundation for data-driven model development in scientific machine learning applications

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1975

Design and evaluation of a sensor fail-operational control system for a digitally controlled turbofan engine

A self-learning, sensor fail-operational, control system for the TF30-P-3 afterburning turbofan engine was designed and evaluated. The sensor fail-operational control system includes a digital computer program designed to operate in conjunction with the standard TF30-P-3 bill-of-materials control. Four engine measurements and two compressor face measurements are tested. If any engine measurements are found to have failed, they are replaced by values synthesized from computer-stored information. The control system was evaluated by using a realtime, nonlinear, hybrid computer engine simulation at sea level static condition, at a typical cruise condition, and at several extreme flight conditions. Results indicate that the addition of such a system can improve the reliability of an engine digital control system.

Hrach, F. J.↗
NASA NTRS1978

Economic evaluation of DSS 13 unattended operations demonstration

The goals and data collection requirements to be used for the economic and performance evaluation indexes and life cycle cost parameters for the upcoming operations demonstration of an automated Deep Space Station (DSS) run unattended and controlled remotely from JPL are presented. These evaluation indexes compare the remote operation of telemetry at DSS 13 with the cost and performance of a comparable manned operation at DSS 11. A description is presented of the data that needs to be collected, how the data will be analyzed, and what can and cannot be learned from this operations demonstration.

Remer, D. S.↗
DOE OSTI2021

Application of a machine learning algorithm (XGBoost) to offline RHIC luminosity optimization

The operation parameter optimization in 2020 RHIC low energy run is difficult. First, the RHIC luminosity is affected by many RHIC operation parameters, as well as affected by many Low Energy RHIC electron Cooling (LEReC) operation parameters. Second, the luminosity signal in this run is noisy and not sensitive to these parameter changes, especially when these parameters are very close to their optimized values. It is not easy to distinguish the effects of one parameter from all other operation parameters separately. Therefore, it is difficult to optimize the luminosity by varying these parameters one by one. To find a way for luminosity optimization, we analyze some operation parameters via a machine learning algorithm - XGBoost. After constructing a black-box surrogate model from XGBoost and plotting their partial dependency plots (PDF) and SHAP value plots for different operation parameters, we can find the effects of an individual parameter on the RHIC luminosity and optimize it accordingly.

43 PARTICLE ACCELERATORS↗
DOE OSTI2021

Solving the Dynamics-Aware Economic Dispatch Problem with the Koopman Operator

The dynamics-aware economic dispatch (DED) problem embeds low-level generator dynamics and operational constraints to enable near real-time scheduling of generation units in a power network. DED produces a more dynamic supervisory control policy than traditional economic dispatch (T-ED) that reduces overall generation costs. However, in contrast to T-ED, DED is a nonlinear, non-convex optimization problem that is computationally prohibitive to solve. We introduce a machine learning-based operator-theoretic approach for solving the DED problem efficiently. Specifically, we develop a novel discrete-time Koopman Operator (KO) formulation that embeds domain information into the structure of the KO to learn high-fidelity approximations of the generator dynamics. Using the KO approximation, the DED problem can be reformulated as a computationally tractable linear program (abbreviated DED-KO). We demonstrate the high solution quality and computational-time savings of the DED-KO model over the original DED formulation on a 9-bus test system.

King, Ethan↗
NASA NTRS1996

Cost efficient operations: Challenge from NASA administrator and lessons learned from hunting sacred cows

The conclusions and recommendations that resulted from NASA's Hunting Sacred Cows Workshop are summarized, where a sacred cow is a belief or assumption that is so well established that it appears to be unreasonably immune to criticism. A link was identified between increased complexity and increased costs, especially in relation to automation and autonomy. An identical link was identified for outsourcing and commercialization. The work of NASA's Cost Less team is reviewed. The following conclusions were stated by the Cost Less team and considered at the workshop: the way Nasa conducts business must change; NASA makes its best contributions to the public areas not addressed by other government organizations; the management tool used for the last 30 years is no longer suitable; the most important work on any program or project is carried out before the development or operations stages; automation should only be used to achieve autonomy if the reasons for automation are well understood, and NASA's most critical resources are its personnel.

Hornstein, Rhoda Shaller↗
NASA NTRSDate not supplied

Astrobee: Completed, Current, and Future Research using Free Flying Robots on the International Space Station

After four years on the International Space Station (ISS), the Astrobee Research Facility, has completed over 130 Test Sessions logging over 1000 hours of operations. Managed by the NASA ISS Program OZ office and supported by NASA Ames Research Center (ARC) in California, the Astrobee Team maintains three identical free-flying Astrobee robots for research on the ISS. As a technology demonstration platform, the Astrobee Robots are available for Guest Scientists to use for a spectrum of research capabilities. Astrobee, propelled by battery-operated fans, is designed to autonomously operate throughout most of the USOS (US Orbital Segment), with the objective of minimizing astronaut support. Astrobee carries a suite of six cameras, a two degree-of-freedom (DOF) arm with a gripper that can grasp ISS handrails and other objects, and three payload bays that provide power and data for guest science hardware. Astrobee can autonomously execute hours-long flight plans or be teleoperated from the ground or by astronauts. While the Astrobee Team continues to improve mapping and autonomous flight capabilities, one of the main goals of Astrobee Robots is to provide research opportunities for Guest Scientists. The Astrobee Robot Software (ARS) makes extensive use of the open-source Robot Operating System (ROS). The ARS can be used interchangeably with an Astrobee Simulator or as Astrobee’s onboard software. ARS features include autonomous docking and perching, real-time teleoperations from the ground, plan based autonomous tasks, multi Astrobee communication, among other capabilities. Through simulation software and ground testing laboratories, the Astrobee Team is available to support Guest Scientists during development and testing and lead real-time ISS operations. Guest Scientists can participate in this research opportunity following the Guest Science Lifecycle (GSL) shown in Figure 1: Guest Science Lifecycle below. The Astrobee Team and Guest Scientists have complete research including Gecko materials studies, RFID and sound sensing capabilities, student Robotics Programming Challenges, and Free Flyer formation flight investigations. Current science with the Astrobee Robots includes Free Flyer self-toss studies, new docking capabilities, advanced mapping resolution capabilities, and high resolution panoramic imagery. Future Guest scientists and Astrobee Team research will focus on robotics applications for future NASA missions such as Gateway and Artemis and potential experiments involving human-robot interactions. This presentation will focus on four main subjects, 1) completed, current, and future planned research using the Astrobee robots, 2) how Guest Scientist get from conception to the ISS, 3) Astrobee Facility resources available for Guest Science ground testing and real-time ISS operations support, and 4) lessons learned from four years of ISS operations.

Astrobee↗
NASA NTRS2013

Flight Planning Branch NASA Co-op Tour

This semester I worked with the Flight Planning Branch at the NASA Johnson Space Center. I learned about the different aspects of flight planning for the International Space Station as well as the software that is used internally and ISSLive! which is used to help educate the public on the space program. I had the opportunity to do on the job training in the Mission Control Center with the planning team. I transferred old timeline records from the planning team's old software to the new software in order to preserve the data for the future when the software is retired. I learned about the operations of the International Space Station, the importance of good communication between the different parts of the planning team, and enrolled in professional development classes as well as technical classes to learn about the space station.

Marr, Aja M.↗
NASA NTRS2008

Cassini Attitude Control Flight Software: from Development to In-Flight Operation

The Cassini Attitude and Articulation Control Subsystem (AACS) Flight Software (FSW) has achieved its intended design goals by successfully guiding and controlling the Cassini-Huygens planetary mission to Saturn and its moons. This paper describes an overview of AACS FSW details from early design, development, implementation, and test to its fruition of operating and maintaining spacecraft control over an eleven year prime mission. Starting from phases of FSW development, topics expand to FSW development methodology, achievements utilizing in-flight autonomy, and summarize lessons learned during flight operations which can be useful to FSW in current and future spacecraft missions.

Architecture↗