Search NASA⌕ Search

SEARCH · Search NASA

Results for “Model serving”

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 703 records · Page 39

Development of Graphical User Interface for ARRBOD (Acute Radiation Risk and BRYNTRN Organ Dose Projection)

The space radiation environment, particularly solar particle events (SPEs), poses the risk of acute radiation sickness (ARS) to humans; and organ doses from SPE exposure may reach critical levels during extra vehicular activities (EVAs) or within lightly shielded spacecraft. NASA has developed an organ dose projection model using the BRYNTRN with SUMDOSE computer codes, and a probabilistic model of Acute Radiation Risk (ARR). The codes BRYNTRN and SUMDOSE, written in FORTRAN, are a Baryon transport code and an output data processing code, respectively. The ARR code is written in C. The risk projection models of organ doses and ARR take the output from BRYNTRN as an input to their calculations. BRYNTRN code operation requires extensive input preparation. With a graphical user interface (GUI) to handle input and output for BRYNTRN, the response models can be connected easily and correctly to BRYNTRN in friendly way. A GUI for the Acute Radiation Risk and BRYNTRN Organ Dose (ARRBOD) projection code provides seamless integration of input and output manipulations, which are required for operations of the ARRBOD modules: BRYNTRN, SUMDOSE, and the ARR probabilistic response model. The ARRBOD GUI is intended for mission planners, radiation shield designers, space operations in the mission operations directorate (MOD), and space biophysics researchers. The ARRBOD GUI will serve as a proof-of-concept example for future integration of other human space applications risk projection models. The current version of the ARRBOD GUI is a new self-contained product and will have follow-on versions, as options are added: 1) human geometries of MAX/FAX in addition to CAM/CAF; 2) shielding distributions for spacecraft, Mars surface and atmosphere; 3) various space environmental and biophysical models; and 4) other response models to be connected to the BRYNTRN. The major components of the overall system, the subsystem interconnections, and external interfaces are described in this report; and the ARRBOD GUI product is explained step by step in order to serve as a tutorial.

Kim, Myung-Hee↗

The effect of model uncertainty on some optimal routing problems

The effect of model uncertainties on optimal routing in a system of parallel queues is examined. The uncertainty arises in modeling the service time distribution for the customers (jobs, packets) to be served. For a Poisson arrival process and Bernoulli routing, the optimal mean system delay generally depends on the variance of this distribution. However, as the input traffic load approaches the system capacity the optimal routing assignment and corresponding mean system delay are shown to converge to a variance-invariant point. The implications of these results are examined in the context of gradient-based routing algorithms. An example of a model-independent algorithm using online gradient estimation is also included.

Mohanty, Bibhu↗

Cabin Environment Physics Risk Model

This paper presents a Cabin Environment Physics Risk (CEPR) model that predicts the time for an initial failure of Environmental Control and Life Support System (ECLSS) functionality to propagate into a hazardous environment and trigger a loss-of-crew (LOC) event. This physics-of failure model allows a probabilistic risk assessment of a crewed spacecraft to account for the cabin environment, which can serve as a buffer to protect the crew during an abort from orbit and ultimately enable a safe return. The results of the CEPR model replace the assumption that failure of the crew critical ECLSS functionality causes LOC instantly, and provide a more accurate representation of the spacecraft's risk posture. The instant-LOC assumption is shown to be excessively conservative and, moreover, can impact the relative risk drivers identified for the spacecraft. This, in turn, could lead the design team to allocate mass for equipment to reduce overly conservative risk estimates in a suboptimal configuration, which inherently increases the overall risk to the crew. For example, available mass could be poorly used to add redundant ECLSS components that have a negligible benefit but appear to make the vehicle safer due to poor assumptions about the propagation time of ECLSS failures.

Risk↗

Statistical prediction with Kanerva's sparse distributed memory

A new viewpoint of the processing performed by Kanerva's sparse distributed memory (SDM) is presented. In conditions of near- or over-capacity, where the associative-memory behavior of the model breaks down, the processing performed by the model can be interpreted as that of a statistical predictor. Mathematical results are presented which serve as the framework for a new statistical viewpoint of sparse distributed memory and for which the standard formulation of SDM is a special case. This viewpoint suggests possible enhancements to the SDM model, including a procedure for improving the predictiveness of the system based on Holland's work with genetic algorithms, and a method for improving the capacity of SDM even when used as an associative memory.

Rogers, David↗

Unsteady Aerodynamic Modeling of Atmospheric Entry Vehicles in Subsonic and Incompressible flow: A Frequency Response Approach

The determination of time-varying lift force and pitch moment generated by a purely pitching Earth-entry capsule is investigated. Experiments were conducted in the 12-foot Low-Speed Tunnel at NASA Langley Research Center, testing a range of oscillation frequencies at a zero-mean angle of attack with a pitching amplitude of 10 degrees. In light of these measurements, a closed-form set of analytically derived equations for lift and moment was used to develop a semi-empirical formulation, incorporating empirically determined values from the experimental runs. The equations are grounded in potential flow theory, Theodorsen's classical theory of unsteady aerodynamics, and the Joukowski theorem of conformal mapping. The unsteady aerodynamics generated by the oscillating body are then modeled by constructing frequency response functions, with quasi-steady forces and moments serving as inputs and unsteady forces and moments as outputs. The experimentally determined gain and phase variations characterize the unsteady nature of the flow and the system's response and flow time-lag to input flow parameters for a blunt-body entry vehicle. The final semi-empirical model is validated with a set of parameters beyond the initial test matrix.

Entry Systems Modeling↗

Unsteady Aerodynamic Modeling of Atmospheric Entry Vehicles in Subsonic and Incompressible Flow: A Frequency Response Approach

The determination of time-varying lift force and pitch moment generated by a purely pitching Earth-entry capsule is investigated. Experiments were conducted in the 12-foot Low-Speed Tunnel at NASA Langley Research Center, testing a range of oscillation frequencies at a zero-mean angle of attack with a pitching amplitude of 10 degrees. In light of these measurements, a closed-form set of analytically derived equations for lift and moment was used to develop a semi-empirical formulation, incorporating empirically determined values from the experimental runs. The equations are grounded in potential flow theory, Theodorsen's classical theory of unsteady aerodynamics, and the Joukowski theorem of conformal mapping. The unsteady aerodynamics generated by the oscillating body are then modeled by constructing frequency response functions, with quasi-steady forces and moments serving as inputs and unsteady forces and moments as outputs. The experimentally determined gain and phase variations characterize the unsteady nature of the flow and the system's response and flow time-lag to input flow parameters for a blunt-body entry vehicle. The final semi-empirical model is validated with a set of parameters beyond the initial test matrix.

Entry Systems Modeling↗

Forward Bay Cover Separation Modeling and Testing for the Orion Multi-Purpose Crew Vehicle

Spacecraft multi‐body separation events during atmospheric descent require complex testing and analysis to validate the flight separation dynamics model and to verify no recontact. NASA Orion Multi‐Purpose Crew Vehicle (MPCV) teams examined key model parameters and risk areas to develop a robust but affordable test campaign in order to validate and verify the Forward Bay Cover (FBC) separation event for Exploration Flight Test‐1 (EFT‐1). The FBC jettison simulation model is highly complex, consisting of dozens of parameters varied simultaneously, with numerous multi‐parameter interactions (coupling and feedback) among the various model elements, and encompassing distinct near‐field, mid‐field, and far‐field regimes. The test campaign was composed of component‐level testing (for example gas‐piston thrusters and parachute mortars), ground FBC jettison tests, and FBC jettison air‐drop tests that were accomplished by a highly multi‐disciplinary team. Three ground jettison tests isolated the testing of mechanisms and structures to anchor the simulation models excluding aerodynamic effects. Subsequently, two air‐drop tests added aerodynamic and parachute parameters, and served as integrated system demonstrations, which had been preliminarily explored during the Orion Pad Abort‐1 (PA‐1) flight test in May 2010. Both ground and drop tests provided extensive data to validate analytical models and to verify the FBC jettison event for EFT‐1, but more testing is required to support human certification, for which NASA and Lockheed Martin are applying knowledge from Apollo and EFT‐1 testing and modeling to develop a robust but affordable human spacecraft capability.

Ali, Yasmin↗

Dynamical evolution of the Oort cloud

New studies of the dynamical evolution of cometary orbits in the Oort cloud are made using a revised version of Weissman's (1982) Monte Carlo simulation model, which more accurately mimics the perturbation of comets by the giant planets. It is shown that perturbations by Saturn and Jupiter provide a substantial barrier to the diffusion of cometary perihelia into the inner solar system. Perturbations by Uranus and Neptune are rarely great enough to remove comets from the Oort cloud, but do serve to scatter the comets in the cloud in initial energy. The new model gives a population of 1.8 to 2.1 x 10 to the 12th comets for the present-day Oort cloud, and a mass of 7 to 8 earth masses. Perturbation of the Oort cloud by giant molecular clouds in the galaxy is discussed, as is evidence for a massive 'inner Oort cloud' internal to the observed one. The possibility of an unseen solar companion orbiting in the Oort cloud and causing periodic comet showers is shown to be dynamically plausible but unlikely, based on the observed cratering rate on the earth and moon.

Weissman, P. R.↗

An airport community noise-impact assessment model

A computer model was developed to assess the noise impact of an airport on the community which it serves. Assessments are made using the Fractional Impact Method by which a single number describes the community aircraft noise environment in terms of exposed population and multiple event noise level. The model is comprised of three elements: a conventional noise footprint model, a site specific population distribution model, and a dose response transfer function. The footprint model provides the noise distribution for a given aircraft operating scenario. This is combined with the site specific population distribution obtained from a national census data base to yield the number of residents exposed to a given level of noise. The dose response relationship relates noise exposure levels to the percentage of individuals highly annoyed by those levels.

Deloach, R.↗

Aero-Engines AI - A Machine-Learning App for Aircraft Engine Concepts Assessment

Effective deployment of trained machine-learning models could drive a high level of efficiency in aircraft engine conceptual design. Aero-Engines AI is a Windows app that has been created to deploy trained machine-learning models to assess aircraft engine concepts. It was created using tkinter, a GUI (graphical user interface) module that is built into the standard Python library. Employing tkinter greatly facilitates the sharing of machine-learning application as an executable file which can be run on Windows machines (without the need to have Python or any library installed). Current version of the app focuses on the performance prediction of conventional turbofans. The app gets user input for a turbofan design, preprocesses the input data, and deploys trained machine-learning models to predict turbofan thrust specific fuel consumption (TSFC), engine weight, core size, and turbomachinery stage-counts. The machine-learning predictive models were built by employing supervised deep-learning algorithm to study patterns in an existing open-source database of production and research turbofan engines. They were trained, cross-validated, and tested in Keras, an open-source neural networks API (application programming interface) written in Python, with TensorFlow (Google open-source artificial intelligence library) serving as the backend engine. The smooth deployment of these machine-learning models using the app shows that Aero-Engines AI is an easy-to-use and a time-saving tool for aircraft engine design-space exploration during the conceptual design stage.

machine learning↗

Landsat Derived Bathymetry of Lakes on the Arctic Coastal Plan of Northern Alaska

The Pleistocene sand sea on the Arctic Coastal Plain (ACP) of northern Alaska is underlain by anancient sand dune field, a geological feature that affects regional lake characteristics. Many of these lakes, whichcover approximately 20 % of the Pleistocene sand sea, are relatively deep (up to 25 m). In addition to the nat-ural importance of ACP sand sea lakes for water storage, energy balance, and ecological habitat, the need forwinter water for industrial development and exploration activities makes lakes in this region a valuable resource.However, ACP sand sea lakes have received little prior study. Here, we collect in situ bathymetric data to test12 model variants for predicting sand sea lake depth based on analysis of Landsat-8 Operational Land Imager(OLI) images. Lake depth gradients were measured at 17 lakes in midsummer 2017 using a Humminbird 798ciHD SI Combo automatic sonar system. The field-measured data points were compared to red–green–blue (RGB)bands of a Landsat-8 OLI image acquired on 8 August 2016 to select and calibrate the most accurate spectral-depth model for each study lake and map bathymetry. Exponential functions using a simple band ratio (withbands selected based on lake turbidity and bed substrate) yielded the most successful model variants. For eachlake, the most accurate model explained 81.8 % of the variation in depth, on average. Modeled lake bathymetrieswere integrated with remotely sensed lake surface area to quantify lake water storage volumes, which rangedfrom 1.056×10−3to 57.416×10−3km3. Due to variations in depth maxima, substrate, and turbidity betweenlakes, a regional model is currently infeasible, rendering necessary the acquisition of additional in situ datawith which to develop a regional model solution. Estimating lake water volumes using remote sensing will fa-cilitate better management of expanding development activities and serve as a baseline by which to evaluatefuture responses to ongoing and rapid climate change in the Arctic. All sonar depth data and modeled lakebathymetry rasters can be freely accessed at https://doi.org/10.18739/A2SN01440 (Simpson and Arp, 2018) andhttps://doi.org/10.18739/A2HT2GC6G (Simpson, 2019), respectively.

Claire E Simpson↗

Development of a Bayesian Belief Network Runway Incursion Model

In a previous paper, a statistical analysis of runway incursion (RI) events was conducted to ascertain their relevance to the top ten Technical Challenges (TC) of the National Aeronautics and Space Administration (NASA) Aviation Safety Program (AvSP). The study revealed connections to perhaps several of the AvSP top ten TC. That data also identified several primary causes and contributing factors for RI events that served as the basis for developing a system-level Bayesian Belief Network (BBN) model for RI events. The system-level BBN model will allow NASA to generically model the causes of RI events and to assess the effectiveness of technology products being developed under NASA funding. These products are intended to reduce the frequency of RI events in particular, and to improve runway safety in general. The development, structure and assessment of that BBN for RI events by a Subject Matter Expert panel are documented in this paper.

Green, Lawrence L.↗

Constitutive Modeling of Nanotube/Polymer Composites with Various Nanotube Orientations

In this study, a technique has been proposed for developing constitutive models for polymer composite systems reinforced with single-walled carbon nanotubes (SWNT) with various orientations with respect to the bulk material coordinates. A nanotube, the local polymer adjacent to the nanotube, and the nanotube/polymer interface have been modeled as an equivalent-continuum fiber by using an equivalent-continuum modeling method. The equivalent-continuum fiber accounts for the local molecular structure and bonding information and serves as a means for incorporating micromechanical analyses for the prediction of bulk mechanical properties of SWNT/polymer composite. As an example, the proposed approach is used for the constitutive modeling of a SWNT/LaRC-SI (with a PmPV interface) composite system, with aligned nanotubes, three-dimensionally randomly oriented nanotubes, and nanotubes oriented with varying degrees of axisymmetry. It is shown that the Young s modulus is highly dependent on the SWNT orientation distribution.

Odegard, Gregory M.↗

Towards an Aviation Large Language Model by Fine-tuning and Evaluating Transformers

In the aviation domain, there are many applications for machine learning and artificial intelligence tools that utilize natural language. For example, there is a desire to know the commonalities in written safety reports such as voluntary post incidents reports or aerial wildfire operations reports to better understand the risks present. Another use-case is the possibility of extracting airspace procedures and constraints currently written in documents such as Letters of Agreement. These applications can benefit from the use of state-of-the-art natural language processing techniques when adapted to the language/phraseology specific to the aviation domain. This paper evaluates the viability of adaptation of NLP tools to the aviation domain by fine-tuning transformer based models using aviation data sets. In 2018, a novel language model based on neural units (also called transformers) was created and became known as “Bidirectional Encoder Representations from Transformers” or BERT. This architecture combined with large amounts of English training data and innovative semi-supervised training tasks set the standard for what would later emerge as Large Language Models. The performance of these models was further improved by hyperparameter tuning and refinement of the semi-supervised training task and resulted in “Robustly Optimized BERT Pre-training Approach through hyperparameter tuning” or RoBERTa models. These pre-trained Large Language Models proved to be useful for a wide variety of natural language processing tasks such as text classification and question answering through a process called fine-tuning. The transformer architecture with pre-trained weights served as the basis with the last few layers replaced with layers fine-tuned to perform a new task e.g., a layer that provides a label for the entire input text. This process of fine-tuning can also be used to adapt the models to new domains; e.g., BioBERT started with the pre-trained BERT model and was completed by additional fine-tuning and training on biomedical documents. Transformer-based architectures can also be used to create rich representations of text called embeddings which can serve as the input to other machine learning models. This allows simpler algorithms such as logistic regression to use context-rich representations of the text while still remaining quick to train and evaluate. In the world of aviation, there is a growing demand for natural language processing and understanding but the domain presents unique challenges. Due to the technical content (and specialized language) of most aviation documents, fine-tuning pre-trained Large Language Models to specific tasks has not met the benchmark on natural language processing tasks set by simpler models trained from scratch on the data. To address this deficiency, this paper evaluates the improvements from fine-tuning a Large Language Model on a large set of aviation documents using the original semi-supervised training tasks before performing specific natural language tasks. In fine-tuning, a domain-specific dataset is used on the original training task but with the pre-trained Large Language Model instead of starting from a random initialization. This approach allows the model to be adapted to the specific domain language without discarding the information gained from training on general English data. This paper utilized two major dataset types to train and assess the RoBERTa fine-tuning performance. The first are 7,057 Letters of Agreement which are Federal Aviation Administration (FAA) documents that formalize airspace operations across the national airspace system. They contain many examples of ‘aviation English’ using domain specific terminology and phrasing which serves as a representative basis to perform the semi-supervised fine-tuning. The second type is the 494 document classification labels to be used for evaluation. This down-stream evaluation aims to show the performance of the fine-tuned model, better understand how much data is needed for an effective fine-tuning, and how fine-tuning can be adapted for different applications in-the domain. After semi-supervised training, evaluation begins by encoding the documents for classification using the fine-tuned RoBERTa model. Then a logistic regression classifier is trained to label the document type and compared against our ground truth labels. This currently leads to a 82.8% accuracy on 10-fold cross validation showing improvement over baseline RoBERTa which achieved 81.0%. We plan to measure the improvements on additional tasks and it is expected that these improvements will lead to more robust models that can tackle the natural language processing challenges present in aviation datasets.

ATM↗

Various Numerical Applications on Tropical Convective Systems Using a Cloud Resolving Model

In recent years, increasing attention has been given to cloud resolving models (CRMs or cloud ensemble models-CEMs) for their ability to simulate the radiative-convective system, which plays a significant role in determining the regional heat and moisture budgets in the Tropics. The growing popularity of CRM usage can be credited to its inclusion of crucial and physically relatively realistic features such as explicit cloud-scale dynamics, sophisticated microphysical processes, and explicit cloud-radiation interaction. On the other hand, impacts of the environmental conditions (for example, the large-scale wind fields, heat and moisture advections as well as sea surface temperature) on the convective system can also be plausibly investigated using the CRMs with imposed explicit forcing. In this paper, by basically using a Goddard Cumulus Ensemble (GCE) model, three different studies on tropical convective systems are briefly presented. Each of these studies serves a different goal as well as uses a different approach. In the first study, which uses more of an idealized approach, the respective impacts of the large-scale horizontal wind shear and surface fluxes on the modeled tropical quasi-equilibrium states of temperature and water vapor are examined. In this 2-D study, the imposed large-scale horizontal wind shear is ideally either nudged (wind shear maintained strong) or mixed (wind shear weakened), while the minimum surface wind speed used for computing surface fluxes varies among various numerical experiments. For the second study, a handful of real tropical episodes (TRMM Kwajalein Experiment - KWAJEX, 1999; TRMM South China Sea Monsoon Experiment - SCSMEX, 1998) have been simulated such that several major atmospheric characteristics such as the rainfall amount and its associated stratiform contribution, the Qlheat and Q2/moisture budgets are investigated. In this study, the observed large-scale heat and moisture advections are continuously applied to the 2-D model. The modeled cloud generated from such an approach is termed continuously forced convection or continuous large-scale forced convection. A third study, which focuses on the respective impact of atmospheric components on upper Ocean heat and salt budgets, will be presented in the end. Unlike the two previous 2-D studies, this study employs the 3-D GCE-simulated diabatic source terms (using TOGA COARE observations) - radiation (longwave and shortwave), surface fluxes (sensible and latent heat, and wind stress), and precipitation as input for the Ocean mixed-layer (OML) model.

Shie, C.-L.↗

Formalization and visualization of domain-specific software architectures

This paper describes a domain-specific software design system based on the concepts of software architectures engineering and domain-specific models and languages. In this system, software architectures are used as high level abstractions to formulate a domain-specific software design. The software architecture serves as a framework for composing architectural fragments (e.g., domain objects, system components, and hardware interfaces) that make up the knowledge (or model) base for solving a problem in a particular application area. A corresponding software design is generated by analyzing and describing a system in the context of the software architecture. While the software architecture serves as the framework for the design, this concept is insufficient by itself for supplying the additional details required for a specific design. Additional domain knowledge is still needed to instantiate components of the architecture and develop optimized algorithms for the problem domain. One possible way to obtain the additional details is through the use of domain-specific languages. Thus, the general concept of a software architecture and the specific design details provided by domain-specific languages are combined to create what can be termed a domain-specific software architecture (DSSA).

Bailor, Paul D.↗

Radio scintillations observed during atmospheric occultations of Voyager: Internal gravity waves at Titan and magnetic field orientations at Jupiter and Saturn

The refractive index of planetary atmospheres at microwave frequencies is discussed. Physical models proposed for the refractive irregularities in the ionosphere and neutral atmosphere serve to characterize the atmospheric scattering structures, and are used subsequently to compute theoretical scintillation spectra for comparison with the Voyager occultation measurements. A technique for systematically analyzing and interpreting the signal fluctuations observed during planetary occultations is presented and applied to process the dual-wavelength data from the Voyager radio occultations by Jupiter, Saturn, and Titan. Results concerning the plasma irregularities in the upper ionospheres of Jupiter and Saturn are reported. The measured orientation of the irregularities is used to infer the magnetic field direction at several locations in the ionospheres of these two planets; the occultation measurements conflict with the predictions of Jovian magnetic field models, but generally confirm current models of Saturn's field. Wave parameters, including the vertical fluxes of energy and momentum, are estimated, and the source of the internal gravity waves discovered in Titan's upper atmosphere is considered.

Hinson, D. P.↗

Development of a Bayesian Belief Network Runway Incursion and Excursion Model

In a previous work, a statistical analysis of runway incursion (RI) event data was conducted to ascertain the relevance of this data to the top ten Technical Challenges (TC) of the National Aeronautics and Space Administration (NASA) Aviation Safety Program (AvSP). The study revealed connections to several of the AvSP top ten TC and identified numerous primary causes and contributing factors of RI events. The statistical analysis served as the basis for developing a system-level Bayesian Belief Network (BBN) model for RI events, also previously reported. Through literature searches and data analysis, this RI event network has now been extended to also model runway excursion (RE) events. These RI and RE event networks have been further modified and vetted by a Subject Matter Expert (SME) panel. The combined system-level BBN model will allow NASA to generically model the causes of RI and RE events and to assess the effectiveness of technology products being developed under NASA funding. These products are intended to reduce the frequency of runway safety incidents/accidents, and to improve runway safety in general. The development and structure of the BBN for both RI and RE events are documented in this paper.

Green, Lawrence L.↗