An early-warning-method for detecting bed agglomeration in fluidized bed combustors

In recent years, increased environmental demands has led to growing interest in co-firing biomass or waste in fluidized bed combustors. The addition of these kinds of fuel to the coal feed can enhance the formation of ash agglomerates in the bed. Because of agglomeration, the fluidization quality decreases and even (partial) defluidization may occur. When detected in an early stage, one can take measures to prevent this unwanted situation, for instance, by increased removal of bottom ash or by temporarily increasing the gas velocity. Therefore, on-line detection of agglomerate formation (i.e., changes in the particle size distribution) is a valuable tool in fluidized bed combustor operation. In this paper, a recently reported monitoring method, based on three characteristics of pressure signals (viz., short-term predictability, intensity, and time-scale), is evaluated and the influence of these characteristics on the method outcomes is investigated. The method, which showed good results in laboratory-scale agglomeration experiments, shows to be very sensitive to the signal intensity and therefore to changes in the superficial gas velocity. The authors present an improved monitoring method, based solely on the time-scale of pressure signals, which is sensitive to changes in the particle size distribution, but relatively insensitive to changes inmore » the superficial gas velocity. Because the gas supply often shows small fluctuations in fluidized bed combustors, this method will be useful as an early-warning-method for bed agglomeration in such installations.« less