Experimental and theoretical analysis of heat transfer mechanism in circulating fluidized bed

Fluidized bed and particularly circulating fluidized bed combustion technology is the appropriate technology to directly burn low quality lignite because of the mentioned combustion technologies ability to hold internally SO2 and NOx emissions. With this combustion technology and keeping the bed temperature between 850-950 ºC and also with the addition of absorbents emissions of SO2 and NOx are reduced. As a result, construction of expensive emission filtering systems can be avoided. In circulating fluidized beds highly turbulent bed fills the combustion chamber and this mechanism tremendously increases heat transfer. As a consequence of increased heat transfer from the sizing point of view overall dimensions of the combustion chamber heat transfer area and investment coast as well can be reduced. In circulating fluidized beds, heat transfer to the heating surfaces can be determined from the combination of several different dynamics interaction mechanisms including between bed core flow region and surface, particularly cluster near flow region of heating surface and dispersed phase. In this study, heat transfer characteristics between fluidized bed and heating surface on the wall and between bed and immerse surface perpendicular to bed axis have been experimentally investigated, on a setup designed for this purpose. Heat transfer characteristics have been determined for different bed materials and test conditions. To generalize the experimental results and to be able to use them for out of range values ii correlations have been obtained. In this study, experimental design and working parameters have been obtained in dimensionless form. Power law functions which govern the model have been obtained through regression analysis to correlate test results according to a ?cluster renewal model? with minimum deviation. Correlations can be used for heat transfer calculations (rating) and sizing problems for industrial scale applications. In the developed model uncertainties including measuring errors were around 10 %. In the literature, previous models were able to predict heat transfer with 30 % standard deviation for the low density case (silica sand, sand, aluminum) and 43 % percent standard deviation for high density case (copper, steal). With the ?cluster renewal model? developed from this study low density case heat transfer can be correlated with 19 % standard deviation and for high density case devotion is within 27 %. With this new model, in low density bed 11 % and in high density bed 16 % improvement in the prediction have been obtained compared to previous available model. Key Words : Fluidized bed, heat transfer, experimental and theoretical analysis, cluster renewal model, scale-up