Enhancing the computational performance of granular flow simulations in DEM and CFD-DEM with adaptive sparse contacts (ASC)

The Discrete Element Method (DEM) is a popular numerical method to predict granular flows. However, systems containing a considerable number of particles entail significant computational costs. In this work, we present the Adaptive Sparse Contacts (ASC), designed to enhance the computational efficiency of DEM simulations, particularly for particles in a quasi-static state, through granular temperature evaluation. An extension to CFD-DEM with an advection term that allows the fluid-driven particle motion is also presented. Results obtained with a rectangular hopper case show the adaptability of the algorithm with respect to the variation in particulate flow dynamics with a speedup up to 2.4x. The sensitivity of the method to the granular temperature threshold is assessed using a dam break case. Evaluation of the ASC on fluidized bed and pneumatic conveying cases shows suitability for capturing dense particle-laden flows in CFD-DEM simulations.