Abstract
New methods for flow regime identification were developed and applied to photon count time series measured in a bubble column (0.162 m in ID) and fluidized bed (0.438 m in ID). The signals in the bubble column (operated with an air-therminol system) were measured by means of Computed Tomography (CT), whereas the data in the fluidized bed (operated with an air-polyethylene system) were recorded by means of Nuclear Gauge Densitometry (NGD). The hidden information in the time series was extracted by means of two new parameters: entropy (bit/s) and information entropy (bit). Both of them were calculated on the basis of multiple reconstructions of the time series. In the case of the bubble column, the well-pronounced local minima were used for identification of three transition velocities (0.04, 0.08 and 0.13 m/s). They distinguished the boundaries of the bubbly flow, transition and churn-turbulent flow regimes. In the case of the fluidized bed, the minimum fluidization velocity (0.086 m/s) and minimum bubbling velocity (0.12 m/s) were also identified on the basis of the well-pronounced local minima in the profiles of the new parameters. They distinguished the boundaries of both the transition and bubbling fluidization regimes.