Since its invention in 2004, the particle finite element method (PFEM) has being attracted increasing attention. So far, it has been demonstrated to be a robust and powerful numerical tool for handling various challenging engineering problems such as free-surface flow, solid-structure interaction, multiphase problems, melting problems with phase changes, etc. Nevertheless, several issues arise when adopting it for large deformation geotechnical problems. This is, to a large extent, due to the complex geomaterial behaviour. History dependency makes variable mapping between meshes inevitable if the classical PFEM is adopted. Linear elements used in the conventional PFEM do not work well for capturing soil behaviour. Although the smoothed particle finite element method, a variant version of the PFEM, allows the use of linear elements and alleviates the variable mapping requirement, stress oscillation occurs in its dynamic analysis. In this paper, challenges associated with the conventional PFEM for modelling geotechnical problems are explored followed by a new version of Nodal integration based PFEM (N-PFEM) proposed to overcome the issues. Numerical benchmarks demonstrate the correctness and robustness of the N-PFEM for dynamic analysis of geotechnical problems.
10th European Conference on Numerical Methods in Geotechnical Engineering (NUMGE2023)
2. Finite element, finite difference, discrete element, material point and other methods