Dielectric spectroscopic sensing of fine liquid droplets in an airstream

Abstract

Contamination of compressed air can reduce its utility and lead to costly failure of pneumatic components. Monitoring contaminants in the compressed air could help take preventive measures to maintain usefulness of the pneumatic systems. Dielectric spectroscopy has good potential as a viable commercial sensor technology for pneumatic systems as it can differentiate dielectric properties of the air with and without contaminants. It could also be used to detect the presence of oil mist, required for lubricating pneumatic components. Two tests were performed using a sensor capable of measuring the dielectric spectrum of the fluid mixture. The objective was to investigate the efficacy of dielectric spectroscopy in detecting the presence of deionised water and light lubricant oil in an airstream. These liquids were atomised using industrial spray nozzles, then entrained in an airstream and passed through the sensor. Spectroscopic measurements were acquired and multivariate classifiers were developed using principal component analysis and linear discriminant analysis to investigate the sensor’s performance in differentiating the presence and absence of liquid droplets in the airstream. The classifier was able to separate the two cases suggesting dielectric spectroscopy could be used to detect these two liquids in an airstream.