SuperDEM-CFD Modeling of Fluidization of Spherical and Cylindrical Particle Binary Mixtures and Experiment Validation

Understanding the fluidization behavior of binary mixtures of spherical and nonspherical particles is essential for better designing and optimizing pyrolysis-fluidized bed reactors. This study experimentally investigated the fluidization behaviors of spherical and cylindrical plastic particles and their binary mixtures, and the features studied include pressure drops, particle mixing, and segregation in a rectangular fluidized bed. A machine learning-aided image processing method was developed to segment different particles from the background for the analysis of particle dynamics. The SuperDEM-CFD method in the open-source code MFiX was improved and extended to simulate the fluidization behaviors of spherical and cylindrical particles' binary mixtures and validated by comparing simulation results with experimental data. It was found that the proposed Di Felice & HS hybrid drag model could better predict the pressure drops than previous drag models. The predicted fluidization and mixing behaviors, particle height, and orientation distribution by the improved SuperDEM-CFD method agreed well with the experiments.