烯烃催化裂解固定床工艺中的反应过程对压力敏感，深入研究催化剂堆积颗粒结构中的流动及压力分布对优化固定床结构及操作参数有重要意义。颗粒解析模拟方法广泛用于固定床内堆积结构的模拟，可以准确描述堆积结构中的流体力学行为，但对于复杂堆积结构网格生成困难。采用基于多孔介质模型的浸入边界法(PMM-IBM)结合网格自适应，实现了对固定床堆积结构的颗粒解析模拟，既解决了网格划分困难的问题，又节省了计算资源。采用网格自适应技术后，与均匀网格相比，堆积结构的网格总数减少大约80%。通过与贴体网格法的单颗粒表面受力分析对比，确定了此浸入边界法的关键模拟参数。随后模拟预测了三种床层与颗粒直径比值条件下堆积结构的空隙率及其内部的压力及流动分布。研究表明，堆积结构空隙中的局部轴向速度的最大值可以达到入口速度的10倍以上，轴向平均速度的径向分布与轴向平均空隙率分布一致，均成震荡衰减趋势。除此之外，预测的床层压降与Reichelt经验关联式结果较为吻合。在此基础上，耦合单颗粒内扩散和烯烃裂解的主反应，预测了反应物随孔径和孔隙率的变化，为进一步考虑外流场的变化奠定了方法基础。

Understanding the flow behaviors and pressure distribution in the interstitial void space among particles is of great importance for improving operating parameters of fixed bed chemical processes, like OCC (olefin catalytic cracking) process. Particle-resolved simulation method is extensively adopted in simulations of fixed bed reactors due to its ability of accurate describing fluid dynamics in packing structures, but its mesh generation process is rather difficult when the packing structure is complex. This study realized a particle-resolved simulation method of typical packing structures in the OCC fixed bed reactor through developing an immersed boundary method, which was based on the porous media model (PMM-IBM) and mesh adaption technology. This method simplified mesh generation process and reduces the computational cost. Compared to the uniform mesh, the total mesh number of the packing structure was reduced about 80% after the mesh adaption. The key parameters of the PMM-IBM were first determined by comparing the forces acting on the particle surface between PMM-IBM and body-fitted mesh method. Further, for three different bed-to-particle diameter ratios, the voidage, pressure and velocity distribution of the packing structures were simulated. It was found that the maximum local axial velocity in the packing structures was 10 times higher than the inlet velocity. The radial distribution of the average axial velocity was consistent with that of the average axial voidage, which showed an oscillation attenuation trend along the center of the bed. Moreover, the pressure drop was almost consistent with the results from Reichelt empirical correlation. Finally, the mass transfer and the main reaction of OCC were considered. The mass fraction variation of the reactant according to pore size and porosity was predicted, which provided the possibility for including the external flow structures and variations in future work.