Browsing by Subject "data analysis"

Now showing 1 - 4 of 4
  • No Thumbnail Available
    Item
    An Analysis Tool for Flight Dynamics Monte Carlo Simulations
    (2011-05-20) Restrepo, Carolina 1982-
    Spacecraft design is inherently difficult due to the nonlinearity of the systems involved as well as the expense of testing hardware in a realistic environment. The number and cost of flight tests can be reduced by performing extensive simulation and analysis work to understand vehicle operating limits and identify circumstances that lead to mission failure. A Monte Carlo simulation approach that varies a wide range of physical parameters is typically used to generate thousands of test cases. Currently, the data analysis process for a fully integrated spacecraft is mostly performed manually on a case-by-case basis, often requiring several analysts to write additional scripts in order to sort through the large data sets. There is no single method that can be used to identify these complex variable interactions in a reliable and timely manner as well as be applied to a wide range of flight dynamics problems. This dissertation investigates the feasibility of a unified, general approach to the process of analyzing flight dynamics Monte Carlo data. The main contribution of this work is the development of a systematic approach to finding and ranking the most influential variables and combinations of variables for a given system failure. Specifically, a practical and interactive analysis tool that uses tractable pattern recognition methods to automate the analysis process has been developed. The analysis tool has two main parts: the analysis of individual influential variables and the analysis of influential combinations of variables. This dissertation describes in detail the two main algorithms used: kernel density estimation and nearest neighbors. Both are non-parametric density estimation methods that are used to analyze hundreds of variables and combinations thereof to provide an analyst with insightful information about the potential cause for a specific system failure. Examples of dynamical systems analysis tasks using the tool are provided.
  • No Thumbnail Available
    Item
    Diagnosing spatial variation patterns in manufacturing processes
    (Texas A&M University, 2004-09-30) Lee, Ho Young
    This dissertation discusses a method that will aid in diagnosing the root causes of product and process variability in complex manufacturing processes when large quantities of multivariate in-process measurement data are available. As in any data mining application, this dissertation has as its objective the extraction of useful information from the data. A linear structured model, similar to the standard factor analysis model, is used to generically represent the variation patterns that result from the root causes. Blind source separation methods are investigated to identify spatial variation patterns in manufacturing data. Further, the existing blind source separation methods are extended, enhanced and improved to be a more effective, accurate and widely applicable method for manufacturing variation diagnosis. An overall strategy is offered to guide the use of the presented methods in conjunction with alternative methods.
  • No Thumbnail Available
    Item
    PerCon: A Personal Digital Library for Heterogeneous Data Management and Analysis
    (2015-03-31) Park, Su Inn
    Systems are needed to support access to and analysis of larger and more heterogeneous scientific datasets. Users need support in the location, organization, analysis, and interpretation of data to support their current activities with appropriate services and tools. We developed PerCon, a data management and analysis environment, to support such use. PerCon processes and integrates data gathered via queries to existing data providers to create a personal or a small group digital library of data. Users may then search, browse, visualize, annotate, and organize the data as they proceed with analysis and interpretation. Analysis and interpretation in PerCon takes place in a visual workspace in which multiple data visualizations and annotations are placed into spatial arrangements based on the current task. The system watches for patterns in the user?s data selection, exploration, and organization, then through mixed-initiative interaction assists users by suggesting potentially relevant data from unexplored data sources. In order to identify relevant data, PerCon builds up various precomputed feature tables of data objects including their metadata (e.g. similarities, distances) and a user interest model to infer the user interest or specific information need. In particular, probabilistic networks in PerCon model user interactions (i.e. event features) and predict the data type of greatest interest through network training. In turn, the most relevant data objects of interest in the inferred data type are identified through a weighted feature computation then recommended to the user. PerCon?s data location and analysis capabilities were evaluated in a controlled study with 24 users. The study participants were asked to locate and analyze heterogeneous weather and river data with and without the visual workspace and mixed-initiative interaction, respectively. Results indicate that the visual workspace facilitated information representation and aided in the identification of relationships between datasets. The system?s suggestions encouraged data exploration, leading participants to identify more evidences of correlation among data streams and more potential interactions among weather and river data.
  • Thumbnail Image
    Item
    TwoRavens Statistical Analysis and Data Repositories
    (2016-11-16) D'Orazio, Vito; Honaker, James; Bhattacharjee, Rohit; University of Texas at Dallas; Harvard University