Static artificial neural networks (ANN) are applied to determine a propeller's four-quadrant performance.The presented solution method includes two problem approaches: the signal approach based on the representation of the propeller's geometry by its first quadrant performance and the direct approach where the prropeller geometry is directly the input of the neural network. The predicted performance for the signal approach shows good agreement with experimental data, while the direct approach prediction is only qualitative. The verification is limited due to the few propellers tested in four-quadrant condition. For the signal approach, empirical formulas based on trained ANNs are presented which can be applied for practical calculations of crash-ahead, crash-back and backing propeller performance, e.g. for the simulation of ship manoeuvres in the design stage.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Application of artificial neural networks for determination of propeller's crash-ahead, crash-back and backing performance


    Weitere Titelangaben:

    Die Anwendung von künstlichen neuralen Netzwerken für die Bestimmung der Voll-voraus-, Voll-zurück und der Rückwärtsleistung des Propellers


    Beteiligte:
    Abramowski, T. (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    2001


    Format / Umfang :

    7 Seiten, 6 Bilder, 15 Quellen



    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Print


    Sprache :

    Englisch