Tension-bending analysis of flexible pipe by a repeated unit cell finite element model

Troels Vestergaard Lukassen*, Egil Gunnarsson, Steen Krenk, Kristian Glejbøl, Anders Lyckegaard, Christian Berggreen

*Corresponding author for this work

    Research output: Contribution to journalJournal articleResearchpeer-review

    Abstract

    Unbonded flexible pipes consist of multiple subcomponents which interact through frictional contact. A full 3-D finite element analysis of unbonded flexible pipes is computationally expensive, and a more efficient approach for practical engineering purposes is required. This work presents a repeated unit cell (RUC) finite element model for analyzing flexible pipes subjected to combined constant tension and curvature. Periodic boundary conditions reduce the model size by taking advantage of the structural and loading periodicities, and by assuming uniform wire behavior in the armor layer. The RUC model is suitable for resolving the local tensile armor stress distribution and the global pipe response. A flexible pipe is studied with the RUC model for various tension-bending load configurations and the results have been compared with existing analytical models for validation. The study showed strong correlation between the RUC model and the analytical models, with some difference in the wire bending stresses. This difference can to some degree be explained by the difference in the wire kinematics assumptions. It is found that the proposed RUC model is a robust and computationally efficient approach for analyzing flexible pipes.
    Original languageEnglish
    JournalMarine Structures
    Volume64
    Pages (from-to)401-420
    ISSN0951-8339
    DOIs
    Publication statusPublished - 2019

    Keywords

    • Flexible pipes
    • Repeated unit cell
    • Tensile armor bending stress
    • Helical layer
    • Frictional contact
    • Finite elements

    Fingerprint

    Dive into the research topics of 'Tension-bending analysis of flexible pipe by a repeated unit cell finite element model'. Together they form a unique fingerprint.

    Cite this