Predictive Model of Air Filter Efficiency Using a Single-Fiber Model
Acoustics is reported to enhance fibrous filter efficiency; however, the degree of improvement against non-acoustic case is difficult to quantify in the absence of a reference model. We developed a COMSOL reference model that predicts capture efficiency across a wide range of fibrous filters, including nanofiber and microfiber, woven and nonwoven, and homogeneous and heterogeneous media. We first solve the flow field around a single fiber and then use a one-way coupling approach to study particle capture. While the initial model assumes each particle as captured when hit the fiber surface, we account for particle rebound using an empirical correlation mimicking real-time capture. This single-fiber efficiency is then mapped to obtain the fibrous filter overall capture efficiency using the depth filtration theory. Our model is validated against experiments, showing less than 10% mean absolute error, while capturing the trend of experiments with greater than 90% confidence based on the Pearson correlation coefficient.
