基于实际熔喷纤维电镜扫描(SEM)图像，建立了不同褶间角、层数、直径、纤维个数、曲率的褶型空气过滤介质微观结构模型，实现了对褶型空气过滤介质微观结构的动态控制. 通过计算流体力学和离散单元法(CFD?DEM)耦合的方法对褶型过滤介质含尘过滤阶段的过滤特性进行模拟，将模拟结果与相关经验关联式比较. 结果表明，所建模型与实际过滤介质电镜图像基本相似，压力损失模拟值与计算值误差在10%以内，表明CFD?DEM耦合计算方法可行；在含尘过滤阶段，同一进口风速(v)下随颗粒沉积量增加，压力损失非线性增大；不同进口风速下的压力损失增加均随过滤时间增加而增大，v=0.6 m/s时压力损失增加最大，为0.214 Pa, v=0.4 m/s时压力损失增加最少，为0.133 Pa. 当3.216 ms时，颗粒出现明显沉积与团聚，纤维彼此贯穿的区域沉积更明显，表明纤维的排列方式对颗粒的沉积影响显著.

Three-dimensional random model of the pleated air filter media was built based on the actual melt blown fiber, and its microstructure was established considering a series of parameters, including pleated angle, layer, diameter, number of fibers, curvature, and so on, which had dynamic control of the pleated air filter media. The established model was basically similar to the actual electron microscope image of filter media. On this basis, the filtration characteristic of the pleated air filter media in the dust filtration stage was simulated using CFD?DEM coupling method, and the simulation results were compared with the relevant empirical correlations. The results show that the error between the simulation results and the empirical correlations is less than 10%, which indicates that the calculation based on CFD?DEM coupling method was feasible. In the dust filtration stage, the pressure drop increases nonlinearly with the increasing number of deposition particles at the same inlet gas velocity. The increased pressure drop increases with the increase of filtration time under different inlet velocities, the maximum increased pressure drop (0.214 Pa) appears at 0.6 m/s of the inlet gas velocity and the minimal increased pressure drop (0.133 Pa) appears at 0.4 m/s of the inlet gas velocity. The phenomenon of the particles’ deposition and agglomeration is obvious when the filtration time, T is 3.216 ms, and this phenomenon is more apparent in the penetrated region of fibers, which also indicates that the arrangement of fibers have significant influence on the deposition of particles.