Increasing competitive pressure is confronting the automotive industry with major challenges. As a result, conventional reactive maintenance is being transformed into predictive maintenance. In this context, wearing and aging effects no longer lead to plant failure since they are predicted at an earlier stage based on comprehensive data analysis.

    Furthermore, the evolution towards Smart Factory has given rise to virtual commissioning in the planning phase of production plants. In this process, a Hardware-in-the-Loop (HiL) system combines the real controls (e.g., PLC) and a virtual model of the plant. These HiL systems are used to simulate commissioning activities in advance, thus saving time and money during actual commissioning. The resulting complex virtual models are not further used in the series production.

    This paper builds upon virtual commissioning models to develop a Digital Twin, which provides inputs for predictive maintenance. The resulting approach is a methodology for building a hybrid predictive maintenance system. A hybrid prediction model combines the advantages of data-driven and physical models. Data-driven models analyse and predict wearing patterns based on real machine data. Physical models are used to reproduce the behaviour of a system. From the simulation of the hybrid model, additional insights for the predictions can be derived.

    The conceptual methodology for a hybrid predictive maintenance system is validated by the successful implementation in a bottleneck process of the electric engine production for an automotive manufacturer. Ultimately, an outlook on further possible applications of the hybrid model is presented.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Unlocking the Potential of Digital Twins


    Untertitel :

    Development of the Virtual Commissioning Model to a Hybrid Predictive Maintenance System


    Weitere Titelangaben:

    ARENA2036


    Beteiligte:
    Kiefl, Niklas (Herausgeber:in) / Wulle, Frederik (Herausgeber:in) / Ackermann, Clemens (Herausgeber:in) / Holder, Daniel (Herausgeber:in) / Pohlkötter, Fabian J. (Autor:in) / Straubinger, Dominik (Autor:in) / Kuhn, Alexander M. (Autor:in) / Imgrund, Christian (Autor:in) / Tekouo, William (Autor:in)

    Kongress:

    Stuttgart Conference on Automotive Production ; 2022 ; Stuttgart, Germany November 16, 2022 - November 18, 2022



    Erscheinungsdatum :

    2023-06-05


    Format / Umfang :

    10 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    Digital twins

    Puig Costa, Janina / Duran, Jaume | BASE | 2010

    Freier Zugriff

    Digital Twins

    Brucherseifer, Eva / Fay, Alexander | Deutsches Zentrum für Luft- und Raumfahrt (DLR) | 2021

    Freier Zugriff

    Digital twins

    Vrabič, Rok / Erkoyuncu, John / Butala, Peter et al. | BASE | 2019

    Freier Zugriff

    DEMYSTIFYING DIGITAL TWINS

    Rigby, J. / Mansfield, R. | TIBKAT | 2020


    Digital Twins for Prognostic Profiling

    Thukaram, Painuri / Mohan, Sreeram | SAE Technical Papers | 2019