A Lagrangian differencing dynamics method for granular flow modeling

Abstract In this paper, a new Lagrangian differencing dynamics (LDD) method is presented for the simulation of granular flows. LDD is a truly meshless method, which employs second-order consistent spatial operators derived from the Taylor expansion and point renormalization. A decoupled pressure-velocity formulation is employed to derive the pressure Poisson equation, which ensures smooth pressure results in time. Granular media are modeled as viscoplastic materials with the Drucker-Prager yield surface. The velocity equation is formulated and solved in an implicit form, in order to handle large viscosities from the constitutive model. Furthermore, a position-based dynamics technique is used to maintain uniform distribution of the Lagrangian points. The presented LDD method for granular flows is validated by simulating two numerical examples. The results are compared with experimental and numerical data from the literature.