The feasibility of using acoustic emissions for monitoring of fluidized bed granulation

Fluidized bed granulation has a complex and multi-variate nature and it must be handled in a multivariate way. Until recently, granulation has been investigated by monitoring a few process parameters separately, such as in- and outlet air humidity and temperature. These individual parameters carry information about the process based on previous experiments utilizing the same process conditions and the granulation end-point has been estimated by taking advantage of existing experience that has led to desired end product. However, this univariate data management approach, i.e., separate parameter detection, does not provide any information about the intercorrelation structure or collinearity of the process parameters. In this paper we introduce a method for the implementation of real-time water content and granule size determination during granulation from acoustic emission spectra. Both these attributes are widely used in the detection of the quality by characterizing the desired end product. Our results show the potential of acoustic emission spectra to contain relevant physical information about the process.