Multi-scale computational fluid dynamics (CFD) simulation bridges the gaps between particle and reactor scales in the modeling of biomass pyrolysis. Its accuracy depends on the coupling of sub-models in different scales. Recent progress includes (a) detailed kinetics with 32 reactions and 59 species; (b) efficient intra-particle models for thermally thick particles; (c) hybrid drag models for sand biomass and biochar; (d) convection heat transfer models for nonspherical particles; (e) shape-resolved models e.g., glued spheres and Superquadrics; (e) machine learning derived models. These sub-models are coupled with MFiX multi-scale CFD solvers: (1) coupling corrected 0-D intra-particle model with MFiX-CGP (coarse grained particle); (2) coupling 1-D intra-particle model with MFiX-PIC (particle in cell); (3) coupling 1-D intra-particle model with Superquadrics MFiX-DEM (discrete element model); (4) coupling 3-D intra-particle model with glued spheres MFiX-DEM. These solvers are validated against experiments and used in the simulation of pyrolysis reactors processing varied feedstocks.

多尺度计算流体动力学 (CFD) 模拟弥合了生物质热解建模中粒子和反应器尺度之间的差距。其精度取决于不同尺度子模型的耦合。最近的进展包括 (a) 32 个反应和 59 个物种的详细动力学；(b) 热厚粒子的有效粒子内模型；(c) 沙子生物质和生物炭的混合阻力模型；(d) 非球形颗粒的对流传热模型；(e) 形状解析模型，例如胶合球体和超二次曲面；(e) 机器学习派生模型。这些子模型与 MFiX 多尺度 CFD 求解器耦合： (1) 耦合校正的 0-D 粒子内模型与 MFiX-CGP（粗粒度粒子）；(2) 将一维粒子内模型与MFiX-PIC（粒子在细胞中）耦合；(3) 将一维粒子内模型与 Superquadrics MFiX-DEM（离散元模型）耦合；(4) 将 3-D 粒子内模型与胶合球体 MFiX-DEM 耦合。这些求解器经过实验验证，并用于模拟处理各种原料的热解反应器。