Different mathematical models are compared with regard to their usage in the model-based leak-monitoring scheme. The pipeline is represented as a two-port system. Three models are compared: the nonlinear distributed parameters model, the linear distributed parameters model and the linear lumped parameters model were used for leak detection and localization. The nonlinear distributed parameters model was simulated using a special program PIPESIM and provides the best results, however it involves the highest computational demand. With very small changes of the signals around a working point, the linear models provide useful results however, with a greater change of working point conditions which may be caused by the leak, the results of linear models become biased. Investigations into adaptation of the linear models to the current working point are under way as well as the knowledge-based gray box modeling.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Pipeline supervision: comparative evaluation of different models for leak detection and localization


    Additional title:

    Pipelineüberwachung: Vergleich verschiedener Prozessmodelle für die Leckerkennung und -suche


    Contributors:
    Matko, D. (author) / Geiger, G. (author)


    Publication date :

    2001


    Size :

    4 Seiten, 2 Bilder, 13 Quellen


    Type of media :

    Conference paper


    Type of material :

    Storage medium


    Language :

    English





    Post-earthquake pipeline leak detection technologies

    Du, Winncy Y. / Yelich, Scott W. | Tema Archive | 2008


    Performance Study of FBG Hoop Strain Sensor for Pipeline Leak Detection and Localization

    Jia, Ziguang / Ren, Liang / Li, Hongnan et al. | ASCE | 2018


    Canadian pipeline installs leak-detection system

    Yoon, M.S. / Mensik, M. / Luk, W.Y. | Tema Archive | 1988


    Study outlines pipeline leak-detection selection

    Kristiansen, Morten | Tema Archive | 2012