Computational Fluid Dynamics coupled with Discrete Element Method (CFD-DEM) has been extensively utilized for studying hydrodynamics and heat transfer in fluidization processes. This study specifically focuses on improving hydrodynamics and heat transfer in a biomass fluidized bed combustor with immersed tubes. The investigation involves the use of mixed biomass, exploring the effects of biomass types, biomass loading, and blending ratios to propose criteria for selecting suitable biomass fuel for the system. Design parameters related to the immersed tubes, such as the angle between tubes, tube diameters, and distance between tubes, were also considered. A data-driven model was developed based on the CFD-DEM results to predict system parameters. This model offers potential applications in real-time practical engineering without the need for CFD-DEM simulation.
The CFD-DEM model was developed for cylindrical biomass and spherical silica sand particles, incorporating an appropriate drag force model. The results showed that the use of mixed biomass with a blending ratio of 1:3 of wood chips to coarse bagasse, with a 5% biomass loading using a ratio of the average equivalent diameter of the biomasses to the equivalent diameter of silica sand of 4.44, was identified as suitable conditions in a general case to obtain the most efficient hydrodynamic results in this study. The criteria for selecting biomass are cylindrical biomass with a greater equivalent diameter than silica sand. Additionally, the proportion of wood chips should be less than that of bagasse. Furthermore, through a 2k factorial design analysis, it was determined that the angle between tubes and tube diameter had the most significant influence on the system behaviors. Moreover, the development of Artificial Neural Network (ANN) models was successful using the simulation dataset, enabling accurate predictions of the mixing index, solid volume fraction, and solid temperature within the system. This study enhances the understanding of hydrodynamics and heat transfer within systems and provides valuable insights for optimizing the design and operation of applications involving mixed biomass and silica sand.