This paper describes a computerized expert system that detects mechanical faults in rotating machines. It was developed for the U.S. fleet of aircraft carriers and has been used successfully aboard ship. Twelve years experience in processing and analyzing vibration data for the fleet has provided a substantial database and the expertise to formulate the program. Order normalized vibration amplitude spectra are screened to obtain significant peaks, their change from previous and relation to statistical averages. Cepstral analysis identifies those peaks included in series of harmonics and sidebands. Screened data are transformed into diagnostic data matrices according to machine configuration. Over 250 diagnostic templates, tabular arrays using the same format as the data matrices, were created to detect specific mechanical faults. Templates employ logic symbols and empirical thresholds to obtain diagnostic results. Diagnostic scores are evaluated to determine the degree of fault severity. New machine configurations can be incorporated with little effort. Using only a portable vibration data collector and laptop computer, hundreds of machines can be tested and diagnosed quickly on site.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    A portable, automated machine condition diagnostics program using order normalized vibration data


    Additional title:

    Tragbares, automatisches Maschinenzustandsdiagnoseprogramm mit Hilfe geordneter Vibrationsdaten


    Contributors:
    Watts, W.A. (author) / Vandyke, D.J. (author)


    Publication date :

    1989


    Size :

    7 Seiten, 4 Bilder, 2 Quellen


    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English