skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Pattern classification approach to rocket engine diagnostics

Abstract

This paper presents a systems level approach to integrate state-of-the-art rocket engine technology with advanced computational techniques to develop an integrated diagnostic system (IDS) for future rocket propulsion systems. The key feature of this IDS is the use of advanced diagnostic algorithms for failure detection as opposed to the current practice of redline-based failure detection methods. The paper presents a top-down analysis of rocket engine diagnostic requirements, rocket engine operation, applicable diagnostic algorithms, and algorithm design techniques, which serve as a basis for the IDS. The concepts of hierarchical, model-based information processing are described, together with the use uf signal processing, pattern recognition, and artificial intelligence techniques which are an integral part of this diagnostic system. 27 refs.

Authors:
Publication Date:
OSTI Identifier:
5649169
Report Number(s):
AIAA-Paper--89-2850; CONF-8907118--
Resource Type:
Conference
Resource Relation:
Conference: 25. American Society of Mechanical Engineers joint propulsion conference, Monterey, CA (USA), 10-13 Jul 1989
Country of Publication:
United States
Language:
English
Subject:
42 ENGINEERING; ROCKET ENGINES; DIAGNOSTIC TECHNIQUES; ALGORITHMS; ARTIFICIAL INTELLIGENCE; MAINTENANCE; MONITORING; PATTERN RECOGNITION; PROPULSION SYSTEMS; RELIABILITY; SPACE SHUTTLES; TECHNOLOGY ASSESSMENT; AIRCRAFT; ENGINES; MATHEMATICAL LOGIC; SPACE VEHICLES; VEHICLES 420200* -- Engineering-- Facilities, Equipment, & Techniques

Citation Formats

Tulpule, S. Pattern classification approach to rocket engine diagnostics. United States: N. p., 1989. Web.
Tulpule, S. Pattern classification approach to rocket engine diagnostics. United States.
Tulpule, S. 1989. "Pattern classification approach to rocket engine diagnostics". United States. doi:.
@article{osti_5649169,
title = {Pattern classification approach to rocket engine diagnostics},
author = {Tulpule, S.},
abstractNote = {This paper presents a systems level approach to integrate state-of-the-art rocket engine technology with advanced computational techniques to develop an integrated diagnostic system (IDS) for future rocket propulsion systems. The key feature of this IDS is the use of advanced diagnostic algorithms for failure detection as opposed to the current practice of redline-based failure detection methods. The paper presents a top-down analysis of rocket engine diagnostic requirements, rocket engine operation, applicable diagnostic algorithms, and algorithm design techniques, which serve as a basis for the IDS. The concepts of hierarchical, model-based information processing are described, together with the use uf signal processing, pattern recognition, and artificial intelligence techniques which are an integral part of this diagnostic system. 27 refs.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = 1989,
month = 1
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • This guide deals with the classification of nuclear rocket engines and their development. Only very general guidance can be given on the classification of vehicles because their characteristics are not established, and because vehicle classification is essentially a military concern. Restricted Data as defined in the Atomic Energy Act of 1954 may be expected to occur in the nuclear rocket program only in the reactor itself (excepting the warhead, for which classification guidance is given in the AEC-DOD Classification Guide). Non-nuclear phases of the engine, propellant storage and handling systems, and the vehicle itself may require protection as Defense Information,more » but will not, in general, reveal Restricted Data.« less
  • Studies are reported on the development of a new map pattern correlation technique, to be applied to the analysis of Northern Hemisphere winter sea-level pressure anomaly patterns, and, where possible, the results are compared with those of principal component analysis of the same data. The map pattern recognition presented is in many ways comparable to the method of principal component eigenvectors. The pattern correlation method has several advantages over principal component eigenvectors.
  • In this paper, we demonstrate the use of self-organizing feature maps as pattern classifiers. When a set of training patterns is presented to a self-organizing network repeatedly for many iterations, the weight vectors gradually organize themselves to be the cluster centers of these patterns. In the one dimensional case, they can arrange themselves to satisfy a linear ordering relation. That is, the distance between two weight vectors increases as the physical distance between the two corresponding output nodes increases. However, the latter is true only when an adequate size of neighborhood is used in the network. We notice a cyclicmore » phenomenon among the distances between weight vectors, when the size of the neighborhood is small. 5 refs., 7 figs., 5 tabs.« less
  • Validation of Computational Fluid Dynamics (CFD) codes developed for prediction and evaluation of rocket performance is hampered by a lack of experimental data. Non-intrusive laser based diagnostics are needed to provide spatially and temporally resolved gas dynamic and fluid dynamic measurements. This paper reports the first non-intrusive temperature and species measurements in the plume of a 110 N gaseous hydrogen/oxygen thruster at and below ambient pressures, obtained with spontaneous Raman spectroscopy. Measurements at 10 mm downstream of the exit plane are compared with predictions from a numerical solution of the axisymmetric Navier-Stokes and species transport equations with chemical kinetics, whichmore » fully model the combustor-nozzle-plume flowfield. The experimentally determined oxygen number density at the centerline at 10 mm downstream of the exit plane is four times that predicted by the model. The experimental number density data fall between those numerically predicted for the exit and 10 mm downstream planes in both magnitude and radial gradient. The predicted temperature levels are within 10 to 15 percent of measured values. Some of the discrepancies between experimental data and predictions result from not modeling the three dimensional core flow injection mixing process, facility back pressure effects, and possible diffuser-thruster interactions.« less