Search NASA⌕ Search

SEARCH · Search NASA

Results for “Scalable Model”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
NASA NTRS2020

TopiQAL: Topic-aware Question Answering using Scalable Domain-specific Supercomputers

We all have questions. About today's temperature, scores of our favorite baseball team, the Universe, and about vaccine for COVID-19. Life, physical, and natural scientists have been trying to find answers to various topics using scientific methods and experiments, while computer scientists have built language models as a tiny step towards automatically answering all of these questions across domains given a little bit of context. In this paper, we propose an architecture using state-of-the-art Natural Language Processing language models namely Topic Models and Bidirectional Encoder Representations from Transformers (BERT) that can transparently and automatically retrieve articles of relevance to questions across domains, and fetch answers to topical questions related to COVID-19 current and historical medical research literature. We demonstrate the benefits of using domain-specific supercomputers like Tensor Processing Units (TPUs), residing on cloud-based infrastructure, using which we could achieve significant gains in training and inference times, also with very minimal cost.

Penberthy, Scott↗
NASA NTRS2013

The Solar Umbrella: A Low-cost Demonstration of Scalable Space Based Solar Power

Within the past decade, the Space Solar Power (SSP) community has seen an influx of stakeholders willing to entertain the SSP prospect of potentially boundless, base-load solar energy. Interested parties affiliated with the Department of Defense (DoD), the private sector, and various international entities have all agreed that while the benefits of SSP are tremendous and potentially profitable, the risk associated with developing an efficient end to end SSP harvesting system is still very high. In an effort to reduce the implementation risk for future SSP architectures, this study proposes a system level design that is both low-cost and seeks to demonstrate the furthest transmission of wireless power to date. The overall concept is presented and each subsystem is explained in detail with best estimates of current implementable technologies. Basic cost models were constructed based on input from JPL subject matter experts and assume that the technology demonstration would be carried out by a federally funded entity. The main thrust of the architecture is to demonstrate that a usable amount of solar power can be safely and reliably transmitted from space to the Earth's surface; however, maximum power scalability limits and their cost implications are discussed.

climate change↗
NASA NTRS2017

Structural Design Exploration of an Electric Powered Multi-Propulsor Wing Configuration

Advancements in aircraft electric propulsion may enable an expanded operational envelope for electrically powered vehicles compared to their internal combustion engine counterparts. High aspect ratio wings provide additional lift and drag reduction for a proposed multi-propulsor design, however, the challenge is to reduce the weight of wing structures while maintaining adequate structural and aeroelastic margins. Design exploration using a conventional design-and-build philosophy coupled with a finite element method (FEM)-based design of experiments (DOE) strategy are presented to examine high aspect ratio wing structures that have spanwise distributed electric motors. Multiple leading-edge-mounted engine masses presented a challenge to design a wing within acceptable limits for dynamic and aeroelastic stability. Because the first four primary bending eigenmodes of the proposed wing structure are very sensitive to outboard motor placement, safety-of-flight requirements drove the need for multiple spars, rib attachments, and outboard structural reinforcements in the design. Global aeroelasticity became an increasingly important design constraint during the on-going design process, with outboard motor pod flutter ultimately becoming a primary design constraint. Designers successively generated models to examine stress, dynamics, and aeroelasticity concurrently. This research specifically addressed satisfying multi-disciplinary design criteria to generate fluid-structure interaction solution sets, and produced high aspect ratio primary structure designs for the NASA Scalable Convergent Electric Propulsion Technology and Operations Research (SCEPTOR) project in the Aeronautic Research Mission Directorate at NASA. In this paper, a dynamics-driven, quasi-inverse design methodology is presented to address aerodynamic performance goals and structural challenges encountered for the SCEPTOR demonstrator vehicle. These results are compared with a traditional computer aided design based approach.

Moore, James B.↗
NASA NTRS2019

Liquid Nitrogen Testing of ISRU Liquefaction Methods in Unsteady Applications

To enable NASA’s planned long duration missions, the agency is putting emphasis on reusable cryogenic systems. Such systems will require replenishing of cryogens on-orbit via a cryogenic tanker or refueling depot, and potentially on the lunar or Martian surfaces with the utilization of in-situ resources. Surface replenishing requires the in-situ production of gaseous oxygen (and hydrogen if on the lunar surface), followed by liquefaction and storage. Funded by NASA’s Advanced Exploration Systems, and managed under the Advanced Cis-Lunar Space Capability Project, the Cryogenic Fluid In-Situ Liquefaction for Landers (CryoFILL) multi center team was formed to develop a liquefaction and storage system that is efficient, reliable and scalable. This presentation will demonstrate the liquefaction and storage of “In-Situ like” propellant via a Tube-On-Tank Heat Exchanger integrated with Active Cooling (cryocooler) (verify proof of concept and obtain relevant data for model validation) and gather lessons learned from “brassboard” testing which will be applied to future liquefaction system prototype testing, then eventually to an end-to-end demonstration.

Stephens, J. R.↗
NASA NTRSDate not supplied

Interplanetary Terminal Procedure Design Applied to Lunar Landing

The impending NASA Artemis mission to the Moon will enable deep space exploration. The planned cadence of crewed Moon missions under Artemis necessitates scalable, safe lunar landings. This paper describes the development of Interplanetary Terminal Procedure Design (iTERP) which leverages current terminal procedure design criteria augmented with lunar physics to support the exploration of the Moon, Mars and beyond. The iTERPS model combines lessons learned from decades of commercial airspace operations with novelties of human space exploration on other worlds. This paper will highlight some of the key functional areas in procedure design safety, training, and operations reimagined for human space flight.

terminal procedure design↗
NASA NTRSDate not supplied

Interplanetary Terminal Procedure Design Applied to Lunar Landing

The impending NASA Artemis mission to the Moon will enable deep space exploration. The planned cadence of crewed Moon missions under Artemis necessitates scalable, safe lunar landings. This paper describes the development of Interplanetary Terminal Procedure Design (iTERP) which leverages current terminal procedure design criteria augmented with lunar physics to support the exploration of the Moon, Mars and beyond. The iTERPS model combines lessons learned from decades of commercial airspace operations with novelties of human space exploration on other worlds. This paper will highlight some of the key functional areas in procedure design safety, training, and operations reimagined for human space flight.

lunar approach procedures, Artemis Moon base, inte↗
NASA NTRSDate not supplied

Structured Covariance Gaussian Networks for Orion Crew Module Aerodynamic Uncertainty Quantification

In this paper we propose a new approach for nonlinear regression and uncertainty quantification. The method is based on a pair of neural networks which parameterize mean and dense covariance functions of a multivariate Gaussian process, trained together to maximize the log-likelihood of observing the given data. The covariance matrix is made positive definite at every input by construction. We also propose a sampling approach that produces viable surrogate function realizations from the Gaussian process. We call the proposed model a Structured Covariance Gaussian Network (SCGN). We illustrate the use of SCGNs for learning an aerodynamic response surface with built-in uncertainty for the Orion crew module. We find that SCGN provides an efficient and systematic way to learn nonlinear functional relationships and dense covariances. We compare results to a baseline Gaussian process regressor and observe that the SCGN provides comparable uncertainty descriptions with improved scalability to dataset size. The sample functions generated by SCGN are fast to evaluate online and are therefore convenient for use in trajectory simulations. These results suggest that SCGN may be a viable computational method for aerodynamic uncertainty quantification.

machine learning↗
NASA NTRSDate not supplied

Structured Covariance Gaussian Networks for Orion Crew Module Aerodynamic Uncertainty Quantification

In this paper we propose a new approach for nonlinear regression and uncertainty quantification. The method is based on a pair of neural networks which parameterize mean and dense covariance functions of a multivariate Gaussian process, trained together to maximize the log-likelihood of observing the given data. The covariance matrix is made positive definite at every input by construction. We also propose a sampling approach that produces viable surrogate function realizations from the Gaussian process. We call the proposed model a Structured Covariance Gaussian Network (SCGN). We illustrate the use of SCGNs for learning an aerodynamic response surface with built-in uncertainty for the Orion crew module. We find that SCGN provides an efficient and systematic way to learn nonlinear functional relationships and dense covariances. We compare results to a baseline Gaussian process regressor and observe that the SCGN provides comparable uncertainty descriptions with improved scalability to dataset size. The sample functions generated by SCGN are fast to evaluate online and are therefore convenient for use in trajectory simulations. These results suggest that SCGN may be a viable computational method for aerodynamic uncertainty quantification.

machine learning↗
NASA NTRS1989

Advanced H2/O2 space engine parametrics

Engine cycle analyses conducted on a 3000-lbf component testing model of an H2/O2-fueled advanced orbit-transfer vehicle engine employing a dual-expander cycle have yielded pressure and temperature trend predictions. On the basis of the results obtained, the dual-expander cycle is projected to be scalable to thrust levels of as much as 50,000 lbf, with chamber pressures of 2000 psi. The high chamber pressure, in conjunction with the use of a gas-gas injector element, facilitates 10:1-range continuously variable throttling. The preferred thrust level for supporting mission studies would be of the order of 20,000 lbf.

Schneider, J. A.↗
NASA NTRS1994

A comparative study of serial and parallel aeroelastic computations of wings

A procedure for computing the aeroelasticity of wings on parallel multiple-instruction, multiple-data (MIMD) computers is presented. In this procedure, fluids are modeled using Euler equations, and structures are modeled using modal or finite element equations. The procedure is designed in such a way that each discipline can be developed and maintained independently by using a domain decomposition approach. In the present parallel procedure, each computational domain is scalable. A parallel integration scheme is used to compute aeroelastic responses by solving fluid and structural equations concurrently. The computational efficiency issues of parallel integration of both fluid and structural equations are investigated in detail. This approach, which reduces the total computational time by a factor of almost 2, is demonstrated for a typical aeroelastic wing by using various numbers of processors on the Intel iPSC/860.

Byun, Chansup↗
NASA NTRS2019

Characterization of a Photon Counting Test Bed for Space to Ground Optical Pulse Position Modulation Communications Links

The National Aeronautics and Space Administration (NASA) Glenn Research Center (GRC) has developed a laboratory transmitter and receiver prototype of a space-to-ground optical communications link. The system is meant to emulate future deep space optical communication links, such as the first crewed flight of Orion, in which the transmitted laser is modulated using pulse position modulation and the receiver is capable of detecting single photons. The transmitter prototype consists of a software defined radio, a high extinction ratio electro-optic modulator system, and a 1550 nm laser. The receiver is a scalable concept and utilizes a single-pixel array of fiber coupled superconducting nanowire single photon detectors. The transmit and receive waveforms follow the Consultative Committee for Space Data Systems (CCSDS) Optical Communications Coding and Synchronization Standard. A software model of the optical transmitter and receiver has also been implemented to predict performance of the optical test bed. This paper describes the transmitter and receiver prototypes as well as the system test configuration. System level tests results are presented and shown to align with predictions from software simulations. The validated software model can be used to in the future to reduce the design cycle of optical communications systems.

software defined radio↗
NASA NTRSDate not supplied

NASA ETM Modeling and Simulation Upper E Traffic Management Meeting December 2020

This is a slide set to be used as part of a meeting series with members of a working group aimed at the development of a concept that addresses needs and gaps in the management of high altitude airspace operations. This concept leverages elements developed through the UAS Traffic Management project with respect to a cooperative, service-based approach that provides services and capabilities in areas (e.g., Upper Class E airspace) that currently receive no or limited service from Air Traffic Control. This concept is meant to provide a safe, fair, and scalable approach to management of Upper Class E operations that reduces the burden on ATC while providing the flexibility and access desired by current and new users of the airspace. This set of slides includes an overview of the past accomplishments, current project structure, and a simulation roadmap for planned work ahead.

NASA ETM modeling and simulation↗
NASA NTRS2000

Using Model-Based Reasoning for Autonomous Instrument Operation

Multiprobe missions are an important part of NASA's future: Cluster, Magnetospheric Multi Scale, Global Electrodynamics and Magnetospheric Constellation are representatives from the Sun-Earth Connections Theme. To make such missions robust, reliable, and affordable, ideally the many spacecraft of a constellation must be at least as easy to operate as one spacecraft is today. To support this need for scalability, science instrumentation must become increasingly easy to operate, even as this same instrumentation becomes more capable and advanced. Communication and control resources will be at a premium for future instruments. Many missions will be out of contact with ground operators for extended periods either to reduce operations cost or because of orbits that limit communication to weekly perigee transits. Autonomous capability is necessary if such missions are to effectively achieve their operational objectives. An autonomous system is one that acts given its situation in a mission appropriate manner without external direction to achieve mission goals. To achieve this capability autonomy must be built into the system through judicious design or through a built-in intelligence that recognizes system state and manages system response. To recognize desired or undesired system states, the system must have an implicit or explicit understanding of its expected states given its history and self observations. The systems we are concerned with, science instruments, can have stringent requirements for system state knowledge in addition to requirements driven by health and safety concerns. Without accurate knowledge of the system state, the usefulness of the science instrument may be severely limited. At the same time, health and safety concerns often lead to overly conservative instrument operations further reducing the effectiveness of the instrument. These requirements, coupled with overall mission requirements including lack of communication opportunities and tolerance of environmental hazards, frame the problem of constructing autonomous science instruments. we are developing a model of the Low Energy Neutral Atom instrument (LENA) that is currently flying on board the Imager for Magnetosphere-to-Aurora Global Exploration (IMAGE) spacecraft. LENA is a particle detector that uses high voltage electrostatic optics and time-of-flight mass spectrometry to image neutral atom emissions from the denser regions of the Earth's magnetosphere. As with most spacecraft borne science instruments, phenomena in addition to neutral atoms are detected by LENA. Solar radiation and energetic particles from Earth's radiation belts are of particular concern because they may help generate currents that may compromise LENA's long term performance. An explicit model of the instrument response has been constructed and is currently in use on board IMAGE to dynamically adapt LENA to the presence or absence of energetic background radiations. The components of LENA are common in space science instrumentation, and lessons learned by modelling this system may be applied to other instruments. This work demonstrates that a model-based approach can be used to enhance science instrument effectiveness. Our future work involves the extension of these methods to cover more aspects of LENA operation and the generalization to other space science instrumentation.

Johnson, Mike↗
NASA NTRS1996

Unstructured Adaptive Grid Computations on an Array of SMPs

Dynamic load balancing is necessary for parallel adaptive methods to solve unsteady CFD problems on unstructured grids. We have presented such a dynamic load balancing framework called JOVE, in this paper. Results on a four-POWERnode POWER CHALLENGEarray demonstrated that load balancing gives significant performance improvements over no load balancing for such adaptive computations. The parallel speedup of JOVE, implemented using MPI on the POWER CHALLENCEarray, was significant, being as high as 31 for 32 processors. An implementation of JOVE that exploits 'an array of SMPS' architecture was also studied; this hybrid JOVE outperformed flat JOVE by up to 28% on the meshes and adaption models tested. With large, realistic meshes and actual flow-solver and adaption phases incorporated into JOVE, hybrid JOVE can be expected to yield significant advantage over flat JOVE, especially as the number of processors is increased, thus demonstrating the scalability of an array of SMPs architecture.

Biswas, Rupak↗
NASA NTRS2017

Comparison of Oxygen Liquefaction Methods for Use on the Martian Surface

In order to use oxygen that is produced on the surface of Mars from In-Situ production processes in a chemical propulsion system, the oxygen must first be converted from vapor phase to liquid phase and then stored within the propellant tanks of the propulsions system. There are multiple ways that this can be accomplished, from simply attaching a liquefaction system onto the propellant tanks to carrying separate tanks for liquefaction and storage of the propellant and loading just prior to launch (the way that traditional rocket launches occur on earth). A study was done into these various methods by which the oxygen (and methane) could be liquefied and stored on the Martian surface. Five different architectures or cycles were considered: Tube-on-Tank (also known as Broad Area Cooling or Distributed Refrigeration), Tube-in-Tank (also known as Integrated Refrigeration and Storage), a modified Linde open liquefaction/refrigeration cycle, the direct mounting of a pulse tube cryocooler onto the tank, and an in-line liquefier at ambient pressure. Models of each architecture were developed to give insight into the performance and losses of each of the options. The results were then compared across eight categories: Mass, Power (both input and heat rejection), Operability, Cost, Manufacturability, Reliability, Volumility, and Scalability. The result was that, given the current state of technology maturity, Tube-on-Tank architectures were the most attractive solution, closely followed by Tube-in-Tank. As a result of this technical analysis and other factors, NASA has determined to focus its Martian surface liquefaction activities and technology development on Tube-on-Tank liquefaction cycles.

Cryogenic propellants↗
NASA NTRS2002

National Combustion Code Parallel Performance Enhancements

The National Combustion Code (NCC) is being developed by an industry-government team for the design and analysis of combustion systems. The unstructured grid, reacting flow code uses a distributed memory, message passing model for its parallel implementation. The focus of the present effort has been to improve the performance of the NCC code to meet combustor designer requirements for model accuracy and analysis turnaround time. Improving the performance of this code contributes significantly to the overall reduction in time and cost of the combustor design cycle. This report describes recent parallel processing modifications to NCC that have improved the parallel scalability of the code, enabling a two hour turnaround for a 1.3 million element fully reacting combustion simulation on an SGI Origin 2000.

Quealy, Angela↗
NASA NTRS2019

Machine Learning-Based Atmospheric Phenomena Detection Platform

As the number of Earth pointing satellites has increased over the last several decades, the data volume retrieved from instruments onboard these satellites has also increased. It is expected that this trend will continue as more data intensive missions and small satellite constellations are launched. Currently, feature detection - namely atmospheric phenomena - in these datasets is performed manually and is thus not scalable with the growing data archives. Recent advancements in computational efficiency allow for the Earth science community to leverage machine learning to identify interesting atmospheric phenomena. Given the wide range of distinctive features in various atmospheric phenomena, a specialized machine learning model is required for accurate detection of these phenomena independently. The Phenomena Portal, developed at NASA IMPACT, is designed to provide visualization for the output from these machine learning models. In addition, detected events for each atmospheric phenomena are stored in a database that can be used to more easily use/subset larger spatiotemporal datasets. The user interface also incorporates additional features to enhance the user experience including spatiotemporal analysis, multiple base layer images, and a slider to filter events with lower probabilities of positive detection. Each detection supports user feedback on whether the detection is true or false that can then be stored and used to improve the machine learning model performance.

Gurung, Iksha↗
NASA NTRS2010

A New Method for Incremental Testing of Finite State Machines

The automatic generation of test cases is an important issue for conformance testing of several critical systems. We present a new method for the derivation of test suites when the specification is modeled as a combined Finite State Machine (FSM). A combined FSM is obtained conjoining previously tested submachines with newly added states. This new concept is used to describe a fault model suitable for incremental testing of new systems, or for retesting modified implementations. For this fault model, only the newly added or modified states need to be tested, thereby considerably reducing the size of the test suites. The new method is a generalization of the well-known W-method and the G-method, but is scalable, and so it can be used to test FSMs with an arbitrarily large number of states.

Pedrosa, Lehilton Lelis Chaves↗