Premium
Early warning of agglomeration in fluidized beds by attractor comparison
Author(s) -
van Ommen J. Ruud,
Coppens MarcOlivier,
van den Bleek Cor M.,
Schouten Jaap C.
Publication year - 2000
Publication title -
aiche journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.958
H-Index - 167
eISSN - 1547-5905
pISSN - 0001-1541
DOI - 10.1002/aic.690461111
Subject(s) - fluidization , attractor , fluidized bed , economies of agglomeration , mechanics , particle (ecology) , series (stratigraphy) , position (finance) , warning system , signal (programming language) , control theory (sociology) , mathematics , computer science , physics , engineering , thermodynamics , mathematical analysis , geology , artificial intelligence , chemical engineering , paleontology , oceanography , control (management) , finance , economics , telecommunications , programming language
An enhanced monitoring method, based on pressure fluctuation measurements, for observing nonstationarities in fluidized‐bed hydrodynamics is presented. Experiments show that it can detect small changes in the particle‐size distribution. Such a monitoring method is useful to give an early warning of the onset of agglomeration in a fluidized bed. In contrast to earlier methods, this method is insensitive to small changes in superficial gas velocity and can handle multiple signals, making it relevant to industrial application. By carefully choosing the measurement position, the method becomes also insensitive to small bed mass variations. It uses the attractor reconstructed from a measured pressure signal, which is a “fingerprint” of the hydrodynamics of the fluidized bed for a certain set of conditions. Using this method statistically the reconstructed attractor of a reference time series of pressure fluctuations (representing the desired fluidization behavior) is compared with that of successive time series measured during the bed operation.