Design and implementation of a data acquisition system with filter quality evaluation

Abstract: Particulate matter is a growing health concern that is considered to contribute to many diseases. To develop appropriate air filtration systems, we need to understand how particulate matter affects air filters. In this project, we implement an automated data acquisition system for an air filter test rig. The data acquisition system allows us to gather empirical data on how particle matter affects air filters over time. Although the quality of the air filters does not reach critical levels, there is a measurable degradation. The collected data is used to train and validate a machine learning model that can evaluate air filter quality. This machine learning proved to be a powerful tool in air filter evaluation and performs with 99% accuracy on test data. The result of this project is a fully functioning data acquisition system along with a user interface that considerably reduces the number of man-hours needed to perform tests of filters. In addition, the automated data acquisition system can notify the operator when the rig needs a change of filter or when certain faults occur. Unfortunately, the project did not reach its original goal of being able to automatically determine when the test rig needs maintenance or re-calibration.