This article proposed two novel stochastic process corrosion growth models for buried pipelines. These models represent corrosion rate as pulses of a Poisson square wave process, and hence are able to represent the inherent time-variability of corrosion growth and produce continuous corrosion growth histories. A linear and a non-linear (power-law) version of the Poisson square wave corrosion rate model have been presented. The linear model suffers from the same limitation of linear random variable models, that is: the model is overly conservative and cannot be extrapolated back in time. The non-linear stochastic process corrosion growth model was shown to precisely fit actual corrosion data for two inspections, and to represent problem physics much better (time to corrosion initiation and inherent time-variability of corrosion growth properly represented). In application to an example pipeline problem, it was shown that, on the long run, linear models present conservative predictions of future failure probabilities, especially the random variable one. Conservativeness is fair to design and operate pipeline systems, but adversely affects solutions as the cost-effective optimization of inspection intervals. Results obtained herein emphasize the importance of calibrating prediction models to the appropriate amount of corrosion data, and also the importance of using appropriate uncertainty models for the corrosion growth process.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Stochastic process corrosion growth models for pipeline reliability


    Weitere Titelangaben:

    Korrosionsfortschrittsmodelle für die Pipeline-Zuverlässigkeit durch Zufallsprozesse



    Erschienen in:

    Corrosion Science ; 74 ; 50-58


    Erscheinungsdatum :

    2013


    Format / Umfang :

    9 Seiten, 9 Bilder, 4 Tabellen, 32 Quellen




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Print


    Sprache :

    Englisch