Virtual coupling is effective to improve the flexibility and efficiency of railway services. It forms multiple train units as a virtually coupled train set (VCTS). To separate units safely by a minimal distance, relative-braking distance (RBD) is employed. However, it is numerically calculated without explicit models in practice, which makes the design of VCTS control approaches challenging. To solve this problem, this paper proposes a data-driven model predictive control (DDMPC) approach that deploys behavioral systems theory and does not need an explicit model of RBD. Specifically, Hankel matrices are constructed to formulate the unknown dynamics of VCTS, based on the prior-measured trajectories of VCTS operation. Then, using the past data and information received from the preceding unit, the controller of each unit is yielded by solving a computationally-efficient optimal control problem. Finally, experiments are conducted, and results show that our approach can stably operate VCTS, and saves by over 99% computation time than the nonlinear model predictive control approach.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Data-Driven Model Predictive Control for Virtually Coupled Train Set Using Behavioral Systems Theory


    Beteiligte:
    Luo, Xiaolin (Autor:in) / Wang, Dongming (Autor:in) / Tang, Tao (Autor:in) / Zhang, Yong (Autor:in) / Liu, Hongjie (Autor:in)


    Erscheinungsdatum :

    24.09.2024


    Format / Umfang :

    727276 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch