Assessment of rolling resistance models in discrete element simulations

Jun Ai, Jian-Fei Chen, J. Michael Rotter, Jin Y. Ooi

    Research output: Contribution to journalArticlepeer-review

    426 Citations (Scopus)


    Particulate systems are of interest in many disciplines. They are often investigated using the discrete element method because of its capability to investigate particulate systems at the individual particle scale. To model the interaction between two particles and between a particle and a boundary, conventional discrete element models use springs and dampers in both the normal and tangential directions. The significance of particle rotation has been highlighted in both numerical studies and physical experiments. Several researchers have attempted to incorporate a rotational torque to account for the rolling resistance or rolling friction by developing different models. This paper presents a review of the commonly used models for rolling resistance and proposes a more general model. These models are classified into four categories according to their key characteristics. The robustness of these models in reproducing rolling resistance effects arising from different physical situations was assessed by using several benchmarking test cases. The proposed model can be seen to be more general and suitable for modelling problems involving both dynamic and pseudo-static regimes. An example simulation of the formation of a 2D sandpile is also shown. For simplicity, all formulations and examples are presented in 2D form, though the general conclusions are also applicable to 3D systems.
    Original languageEnglish
    Pages (from-to)269-282
    Number of pages14
    JournalPowder Technology
    Issue number3
    Early online date29 Sep 2010
    Publication statusPublished - 30 Jan 2011


    • Rolling friction
    • Rolling resistance
    • Discrete element method
    • Granular material
    • Granular flow
    • Sandpile
    • Angle of repose

    Fingerprint Dive into the research topics of 'Assessment of rolling resistance models in discrete element simulations'. Together they form a unique fingerprint.

    Cite this