| Many industrial gas-fluidized beds are operated under conditions in which the bed material tends to agglomerate due to the cohesive nature of the particles comprising the bed. This cohesiveness may be an inherent property of the bed particles at some temperature, or may be due to the addition of a liquid to the bed forming a sticky coating on the particles.;In many gas-fluidized processes, the particulate bed material is contacted with a liquid feed stream. If the liquid feed rate is sufficiently high, the liquid coated particles may stick together resulting in process upset due to bed agglomeration and defluidization, a condition commonly known in industry as "bogging". Bogging is an important concern in the operation of fluid coking units for upgrading oil sands bitumen. The consequences of bogging can be severe, possibly requiring complete unit shut-down and clean-out. It is therefore imperative that the onset of bogging be detected and eliminated in order to prevent unit upset and shut-down.;A fundamental experimental study was undertaken in order to investigate the bogging phenomena and to find an effective method of detecting the onset of bogging prior to a complete loss of fluidization. Experiments were carried out in an air fluidized bed. Different sizes of Geldart group A and B materials were studied. The effect of bed agglomeration and bogging on fluctuations in the bed pressure were measured.;The bed pressure data were analyzed using several common signal analysis techniques including the standard deviation function, the power spectral density function and the autocovariance function. Of these techniques, the standard deviation function was found to be the best method for detecting the onset of bogging. The results of this work may be applied to the development of a process sensor for detecting bogging in industrial scale units. |
