Monitoring a lab-scale fluidized bed dryer: A comparison between pressure transducers, passive acoustic emissions and vibration measurements

Abstract

The potential of both passive acoustic emission and vibration measurements for monitoring gradual process changes in comparison to pressure fluctuation measurements is investigated. Fluidized bed drying experiments of wetted pharmaceutical placebo granule were conducted and monitored by measuring the temperature-, humidity-, and pressure fluctuations. Next to that, the granule moisture content was determined. To test the proposed monitoring technique based on passive acoustic emission and vibration measurements, two types of microphones and an accelerometer were used. Consequently, several data analysis techniques were used to investigate the analogies between the recorded signals and the granule moisture content. All signals were recorded at a frequency of 400 Hz. The results demonstrate that one type of microphone and the accelerometer have some potential as non-intrusive alternatives for the purpose of monitoring transient processes compared to pressure fluctuation measurements. The recordings of another type of microphone did not show any correlation to changing process hydrodynamics. Time- and frequency domain analyses on the measured signals were found to be not suitable for monitoring the process hydrodynamics. The S-statistic, calculated by the attractor comparison method, was demonstrated to be capable of detecting gradual changes in the hydrodynamics of the fluidized bed. However, it is also shown that the investigated alternative measurement techniques are not as robust as pressure fluctuation measurements, which were found to be much better reproducible.