Electronic International Standard Serial Number (EISSN)
1873-4405
abstract
This paper is addressed to introduce a methodology based on low frequency accelerometry that can be used for gas-solid fluidized bed monitoring, and for dynamic diagnosis purposes. The proposed methodology consists on extracting the low frequency information encoded within an accelerometry signal, by means of the Hilbert transform method. The time and the frequency domain analysis show how this low frequency information is directly related to the conventional pressure fluctuation measurements, providing useful information on bulk and bubble dynamics. The cross-correlation and the coherence function analysis between the pressure and the envelope process of the measured accelerometer signals "E", exhibit values approaching unity for frequencies ranging between 2 and 4 Hz. This reveals that pressure signals and accelerometry envelope are related processes. The results from the Coherent Output Power, COP, and the Incoherent Output Power, IOP, analysis confirms that both pressure and envelope time series exhibit the same global and local features open the possibility of using low frequency accelerometers instead of conventional pressure transducers for monitoring and for dynamic diagnosis purposes.