Engine control systems include several sensors, which are used to acquire information about different engine parameters. A fault in any sensor may develop into system failure causing the engine to stop. This has led to the need to introduce a fault-tolerant sensor system (FTSS) that can handle faults in sensors. Fault-tolerance can be achieved by hardware redundancy or analytical redundancy. Hardware redundancy is the addition of extra components to replace the faulty component in the event of failure. However, a greater number of sensors will be required which increases costs, reduces reliability, and increases complexity. As most engine parameters are related to each other, information about one parameter can be obtained from others. This characteristic can be used to create analytical redundancy, which can be employed to accommodate sensor failures. Due to their learning and adaptation capabilities, artificial neural networks are very appealing for the purpose of providing fault tolerance capabilities in automotive engines. The paper considers the application of neural networks to accommodate failures in sensors. A faulty sensor is reconstructed based on the data from related sensors. The effectiveness of this method is tested using real data collected from a fully automated diesel engine test rig. Three sensors are considered: Engine speed, Manifold absolute pressure, and Throttle position.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Fault accommodation for diesel engine sensor system using neural networks


    Beteiligte:
    Badri, A. (Autor:in) / Berry, E. (Autor:in) / Ball, A.D. (Autor:in)


    Erscheinungsdatum :

    2001


    Format / Umfang :

    6 Seiten, 7 Bilder, 4 Quellen



    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Print


    Sprache :

    Englisch