Cap bubbly-to-slug flow regime transition in a vertical annulus

The transition of the cap bubbly-to-slug flow regime was studied by using impedance void meters in a vertical annular flow channel. The characteristics of various flow regimes were demonstrated by the acquired impedance signals. The statistical parameters from the impedance signals were then fed into a self-organizing neural network for pattern categorization. Based on flow visualization, the classified patterns were translated into corresponding flow regimes. In addition, an analytical model was developed to predict the regime transition from cap bubbly to slug flows. Good agreement was obtained between the results by the neural network and the analytical model. Furthermore, comparisons of the model with other researchers’ experimental data also showed satisfactory results.