Particle Filtration DPM CFD Simulation With Brownian Force and UDF, ANSYS Fluent Training

Particle filtration is an essential step for modern engineering applications, including environmental protection, industrial air quality management, and clean room technologies. This work examines the intricate dynamics of particle filtration via advanced CFD simulation utilizing ANSYS Fluent, specifically emphasizing the inclusion of Brownian force effects through User-Defined Functions (UDFs). The Discrete Phase Model (DPM) facilitates careful tracking of particle trajectories and their interactions with the fluid medium, which is particularly vital for submicron particles where Brownian motion markedly affects filtration effectiveness. This study seeks to utilize advanced numerical techniques to clarify the microscopic behavior of particles in filtration processes, hence enhancing the design and performance predicts of filtration systems. This study established based on the reference paper entitled “Modeling particle filtration in disordered 2-D domains: A comparison with cell models [1]” published in Separation and Purification Technology Journal.

• Reference [1]: Hosseini, S. A., and H. Vahedi Tafreshi. “Modeling particle filtration in disordered 2-D domains: A comparison with cell models.” Separation and Purification Technology2 (2010): 160-169.

Figure 1: A sample of Particle Filtration configuration [1]

Simulation Process

To generate 2-D random fibrous geometries, a Fibrous Filters computer program is developed to produce fibrous structures of different porosities. It then transferred to Gambit & ANSYS Spaceclaim software. Then, it was meshed via ANSYS Meshing. Due to sensitivity to grid, a mesh independence study is conducted. The results are shown in the following Table:

As can be seen, grid four with 532600 elements is selected for further investigation. The air carries the particles which were modeled by Discrete Phase Model (DPM). Plenty of surface injections are provided with various particle diameters ranging from micron to smaller ones. More importantly, a user-defined function (UDF) is written to adopt Brownian force effects.

Figure 2: grid generation over particle filter using ANSYS Meshing

Post-processing

The Discrete Phase Model (DPM) simulation shows detailed particle-fluid interactions within the filtration system, as indicated by the velocity and particle tracking contours. The velocity field exhibits considerable flow alteration around the filter elements, with localized acceleration in the gaps between particles and the formation of twisted flow pathways typical of effective particle filtration systems. The streamline visualization illustrates the shift of the flow pattern from uniform upstream circumstances to difficult, linking routes within the filter medium, which directly influences particle capture processes. The complex flow dynamics, along with the effects of Brownian motion incorporated by UDFs, yield a pressure drop of 184.115 Pa across the filter, indicating high flow resistance characteristic of high-efficiency particle filtration systems.

Figure 3: Pressure and velocity field inside Particle Filtration DPM CFD Simulation With Brownian Force and UDF, ANSYS Fluent Training

The examination of particle trajectories and residence time distribution gives essential insights into filtration ability. The particle tracking visualization illustrates the interaction of particles with the filter medium, where certain particles are captured via direct interception, while others navigate deeper trajectories impacted by Brownian motion. The pressure contour indicates a gradual decline in pressure from the inlet to the outlet, with the most pronounced pressure gradient observed across the filter media, where particle capture is most rapid. The time-dependent behavior of particles, as depicted by the particle-particle time contour, reveals differing residence times within the filter media, where extended residence times (represented by warmer colors) correlate with regions of elevated particle capture probability.

Figure 4: Particles residence time in particle filtration CFD simulation