Mesoscale-structure-based dynamic multiscale method for gas-solid flow

Abstract

Multiscale methods have recently underpinned the successful simulation of gas-solid flow. In our prior study (Chen and Wang, AIChE Journal, 2017, 63: 3681–3691), a dynamic multiscale method (DMM) has been proposed to concurrently couple different simulation models in different simulation domains where discrete particle model (DPM) is used in the near wall regions and the rest of simulation domain is solved by two-fluid model (TFM). In this study, a mesoscale-structure-based mapping method was further proposed and implemented into the framework of DMM. The recently developed EMMS-based particle velocity distribution function (Wang et al., AIChE Journal, 2016, 62: 2649–2657), which was derived by assuming that there is a two-phase structure in a computational grid, was used for mapping between different models. In addition, the stability condition of the EMMS model was undertaken to locate the newly inserted particles during the boundary mapping process from the TFM to the DPM. The simulation results of gas-solid flow in a riser containing Geldart B particles using the proposed DMM have compared with those of using the full TFM and full DPM, which verifies its fidelities.