Mechanisms and Modeling of Acoustic Air Filtering System

Choi, Jong-Hyo

Graduate Thesis Or Dissertation

Mechanisms and Modeling of Acoustic Air Filtering System Public Deposited

Analytics

Citations

Citeable URL: https://scholar.colorado.edu/concern/graduate_thesis_or_dissertations/r207tr198

Abstract

This research explores the possibility of an alternative air filtering method through the study of acoustic agglomeration. Acoustic agglomeration is a phenomenon occurring when particles resonate in the presence of sound waves. The mechanisms of acoustic agglomeration are composed of orthokinetic interaction, primary refilling mechanisms, and secondary refilling mechanisms. Orthokinetic interaction, where direct collisions occur between particles due to differences in resonant amplitude and phase angle, is the most essential mechanism. Refilling mechanisms involve shifting particles into the agglomeration volume through repeated vibrations caused by sound waves.

This research developed an OpenFOAM particle agglomeration model, pimAAFoam, by focusing on the key concept of agglomeration volume which is decided by the orthokinetic interaction. The final model consists of acoustic collision, general collision, and wall collision models. Particle collision outcomes are determined by the concept of critical velocity, resulting in either agglomeration or rebound. Model validations were conducted using vertical experimental setups from literature and horizontal laboratory tests from this research. The pimAAFoam model was proven to be valid for both the vertical and horizontal scenarios. The simulation results are within the measurement range from the mixed-air injection in a horizontal duct case. The difference between the measured values and the simulation values for each of the particle sizes was within 25% of the measured values.

With the validated pimAAFoam, sensitivity analysis was performed on variables related to acoustic agglomeration mechanisms. The base-case of the sensitivity test was a 4-liter space with a concentration of 25 parcels per one cm^3 over a 30-second duration. The simulation time cost for the base-case was approximately 4 days using a 6-core, 3.7 GHz CPU.

Through the sensitivity analysis, it was observed that high sound intensity and initial particle concentration led to an increase in acoustic agglomeration. Unlike bag-filter methods, acoustic agglomeration is valuable in a condition of high initial particle concentration. Additionally, slower air velocity resulted in a higher rate of acoustic agglomeration. Furthermore, the utilization of low-frequency sound waves led to a higher rate of acoustic agglomeration unless altering the frequency significantly affected the entrainment factor difference between particles. Particle removal using acoustic agglomeration mechanisms is focused on small particles with the larger particles acting as collector particles.

The study concludes that the OpenFOAM acoustic agglomeration model, pimAAFoam, offers practical simulation time costs with reliability. Improving accuracy of the model, including primary and secondary refilling mechanisms could be suitable for future study.

Creator