SEARCH · Search NASA
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.
This paper describes the MDS approach to state knowledge representation, contrasting it with past efforts, and offering motivation for the approach as a general recipe for similar efforts.
Collective buffering and data pattern solutions are provided for storage, retrieval, and/or analysis of data in a collective parallel processing environment. For example, a method can be provided for data storage in a collective parallel processing environment. The method comprises receiving data to be written for a plurality of collective processes within a collective parallel processing environment, extracting a data pattern for the data to be written for the plurality of collective processes, generating a representation describing the data pattern, and saving the data and the representation.
A computer program, designed to assist in the development and debugging of other software, generates a two-dimensional (2D) representation of a possibly complex n-dimensional (where n is an integer >2) data structure or abstract rank-n object in that other software. The nature of the 2D representation is such that it can be displayed on a non-graphical output device and distributed by non-graphical means.
Critical to the understanding of data is the ability to provide pictorial or visual representation of those data, particularly in support of correlative data analysis. Despite the advancement of visualization techniques for scientific data over the last several years, there are still significant problems in bringing today's hardware and software technology into the hands of the typical scientist. For example, there are other computer science domains outside of computer graphics that are required to make visualization effective such as data management. Well-defined, flexible mechanisms for data access and management must be combined with rendering algorithms, data transformation, etc. to form a generic visualization pipeline. A generalized approach to data visualization is critical for the correlative analysis of distinct, complex, multidimensional data sets in the space and Earth sciences. Different classes of data representation techniques must be used within such a framework, which can range from simple, static two- and three-dimensional line plots to animation, surface rendering, and volumetric imaging. Static examples of actual data analyses will illustrate the importance of an effective pipeline in data visualization system.
An efficient means of storing data in a first-order predicate calculus theorem-proving system is described. The data structure is oriented for large scale question-answering (QA) systems. An algorithm is outlined which uses the data structure to unify a given literal in parallel against all literals in all clauses in the data base. The data structure permits a compact representation of data within a QA system. Some suggestions are made for heuristics which can be used to speed-up the unification algorithm in systems.
A BDD (Boolean Decision Diagram) is a data structure for the compact representation of a Boolean function. It is equipped with efficient algorithms for minimization and for applying Boolean operators. The use of BDDs for representing Boolean functions, combined with symbolic algorithms, facilitated a leap in the capability of model checking for the verification of systems with a huge number of states. Recently BDDs were considered as an efficient representation of data for Runtime Verification (RV). We review here the basic theory of BDDs and summarize their use in model checking and specifically in runtime verification.
Implicit neural representation (INR) has surfaced as a promising direction for solving different scientific visualization tasks due to its continuous representation and flexible input and output settings. We present STSR-INR, an INR solution for generating simultaneous spatiotemporal super-resolution for multivariate time-varying volumetric data. Inheriting the benefits of the INR-based approach, STSR-INR supports unsupervised learning and permits data upscaling with arbitrary spatial and temporal scale factors. Unlike existing GAN- or INR-based super-resolution methods, STSR-INR focuses on tackling variables or ensembles and enabling joint training across datasets of various spatiotemporal resolutions. Here we achieve this capability via a variable embedding scheme that learns latent vectors for different variables. In conjunction with a modulated structure in the network design, we employ a variational auto-decoder to optimize the learnable latent vectors to enable latent-space interpolation. To combat the slow training of INR, we leverage a multi-head strategy to improve training and inference speed with significant speedup. We demonstrate the effectiveness of STSR-INR with multiple scalar field datasets and compare it with conventional tricubic+linear interpolation and state-of-the-art deep-learning-based solutions (STNet and CoordNet).
The analysis and design of practical control systems requires that stochastic models be employed. Analysis and design tools have been developed, for example, for Markovian jump linear continuous and discrete-time systems, piecewise-deterministic processes (PDP's), and general stochastic hybrid systems (GSHS's). These model classes have been used in many applications, including fault tolerant control and networked control systems. This paper presents initial results on the analysis of a sampled-data PDP representation of a nonlinear sampled-data system with a jump linear controller. In particular, it is shown that the state of the sampled-data PDP satisfies the strong Markov property. In addition, a relation between the invariant measures of a sampled-data system driven by a stochastic process and its associated discrete-time representation are presented. As an application, when the plant is linear with no external input, a sufficient testable condition for the convergence in distribution to the invariant delta Dirac measure is given.
Linear vector space theory is used to develop a general representation of a set of data vectors or random vectors by linear combinations of orthonormal vectors such that the mean squared error of the representation is minimized. The orthonormal vectors are shown to be the eigenvectors of an operator. The general representation is applied to several specific problems involving the use of the Karhunen-Loeve expansion, principal component analysis, and empirical orthogonal functions; and the common properties of these representations are developed.
The proposed method of studying biologic rhythms permits the biologist to view physiological data dynamically without assuming that the data are stationary in time. Vector representation of data points is employed, and the summation of the vectors (train of vectors) produces a summation dial that is able to detect dynamic changes in the time of the peak as well as random walks (arrhythmia).
Techniques for the review, display, and manipulation of multidimensional data are developed and described. Multidimensional data is meant in this context to describe scalar data associated with a three dimensional geometry or otherwise too complex to be well represented by traditional graphs. Raster graphics techniques are used to display a shaded image of a three dimensional geometry. The use of color to represent scalar data associated with the geometries in shaded images is explored. Distinct hues are associated with discrete data ranges, thus emulating the traditional representation of data with isarithms, or lines of constant numerical value. Data ranges are alternatively associated with a continuous spectrum of hues to show subtler data trends. The application of raster graphics techniques to the display of bivariate functions is explored.
Summary Scientific datasets are growing rapidly and becoming critical to next‐generation scientific discoveries. The validity of scientific results relies on the quality of data used and data are often subject to change, for example, due to observation additions, quality assessments, or processing software updates. The effects of data change are not well understood and difficult to predict. Datasets are often repeatedly updated and recomputing derived data products quickly becomes time consuming and resource intensive and may in some cases not even be necessary, thus delaying scientific advance. Despite its importance, there is a lack of systematic approaches for best comparing data versions to quantify the changes, and ad‐hoc or manual processes are commonly used. In this article, we propose a novel hierarchical approach for analyzing data changes, including real‐time (online) and offline analyses. We employ a variety of fast‐to‐compute numerical analyses, graphical data change representations, and more resource‐intensive recomputations of a subset of the data product. We illustrate the application of our approach using three scientific diverse use cases, namely, satellite, cosmological, and x‐ray data. The results show that a variety of data change metrics should be employed to enable a comprehensive representation and qualitative evaluation of data changes.
The management of large geophysical and celestial data bases is now, more than ever, the most critical path to timely data analysis. With today's large volume data sets from multiple satellite missions, analysts face the task of defining useful data bases from which data and metadata (information about data) can be extracted readily in a meaningful way. Visualization, following an object-oriented design, is a fundamental method of organizing and handling data. Humans, by nature, easily accept pictorial representations of data. Therefore graphically oriented user interfaces are appealing, as long as they remain simple to produce and use. The Visual Interface for Space and Terrestrial Analysis (VISTA) system, currently under development at the Naval Research Laboratory's Backgrounds Data Center (BDC), has been designed with these goals in mind. Its graphical user interface (GUI) allows the user to perform queries, visualization, and analysis of atmospheric and celestial backgrounds data.
Abstract. Low-level jets (LLJs) occur under a variety of atmospheric conditions and influence the available wind resource for wind energy projects. In 2020, lidar-mounted buoys owned by the US Department of Energy (DOE) were deployed off the California coast in two wind energy lease areas administered by the Bureau of Ocean Energy Management: Humboldt and Morro Bay. The wind profile observations from the lidars and collocated near-surface meteorological stations (4–240 m) provide valuable year-long analyses of offshore LLJ characteristics at heights relevant to wind turbines. At Humboldt, LLJs were associated with flow reversals and north-northeasterly winds, directions that are more aligned with terrain influences than the predominant northerly flow. At Morro Bay, coastal LLJs were observed primarily during northerly flow as opposed to the predominant north-northwesterly flow. LLJs were observed more frequently in colder seasons within the lowest 250 m a.s.l. (above sea level), in contrast with the summertime occurrence of the higher-altitude California coastal jet influenced by the North Pacific High, which typically occurs at heights of 300–400 m. The lidar buoy observations also validate LLJ representation in atmospheric models that estimate potential energy yield of offshore wind farms. The European Centre for Medium-Range Weather Forecasts Reanalysis version 5 (ERA5) was unsuccessful at identifying all observed LLJs at both buoy locations within the lowest 200 m. An extension of the National Renewable Energy Laboratory (NREL) 20-year wind resource dataset for the Outer Continental Shelf off the coast of California (CA20-Ext) yielded marginally greater captures of observed LLJs using the Mellor–Yamada–Nakanishi–Niino (MYNN) planetary boundary layer (PBL) scheme than the 2023 National Offshore Wind dataset (NOW-23), which uses the Yonsei University (YSU) scheme. However, CA20-Ext also produced the most LLJ false alarms, which are instances when a model identified an LLJ but no LLJ was observed. CA20-Ext and NOW-23 exhibited a tendency to overestimate the duration of LLJ events and underestimate LLJ core heights.
Abstract not provided.
A generalized three dimensional perspective software capability was developed within the framework of a low cost computer oriented geographically based information system using the Earth Resources Laboratory Applications Software (ELAS) operating subsystem. This perspective software capability, developed primarily to support data display requirements at the NASA/NSTL Earth Resources Laboratory, provides a means of displaying three dimensional feature space object data in two dimensional picture plane coordinates and makes it possible to overlay different types of information on perspective drawings to better understand the relationship of physical features. An example topographic data base is constructed and is used as the basic input to the plotting module. Examples are shown which illustrate oblique viewing angles that convey spatial concepts and relationships represented by the topographic data planes.
AI technologies which are relevant to astronomical data bases are reviewed, including intelligent interfaces, internal representations, and data analysis. The natural language query system developed for the Hubble Space Telescope and the technique of goal directed queries are considered. Two technologies which might lead to the development of pictorial interfaces are presented: one based on Bayesian probabilities, the other on associative memories. The development of a data analysis system which can discover classes of data within a data base without any information other than the data itself is examined. A prototype data analysis assistant to automatically develop and implement plans for data reduction is described.
A computer program defines and implements an innovative kind of data structure than can be used for representing information derived from serendipitous discoveries made via collection of scientific data on long exploratory spacecraft missions. Data structures capable of collecting any kind of data can easily be implemented in advance, but the task of designing a fixed and efficient data structure suitable for processing raw data into useful information and taking advantage of serendipitous scientific discovery is becoming increasingly difficult as missions go deeper into space. The present software eases the task by enabling definition of arbitrarily complex data structures that can adapt at run time as raw data are transformed into other types of information. This software runs on a variety of computers, and can be distributed in either source code or binary code form. It must be run in conjunction with any one of a number of Lisp compilers that are available commercially or as shareware. It has no specific memory requirements and depends upon the other software with which it is used. This program is implemented as a library that is called by, and becomes folded into, the other software with which it is used.