Predictions of flow transitions in a bubble column by chaotic time series analysis of pressure fluctuation signals

摘要: Abstract In this study, predictions of flow transitions a bubble column are carried out by chaotic time series analysis pressure fluctuation signals. Pressure signals measured for four different distributor plates with the superficial gas velocity varying from 0 to approximately 0.22 m / s . Methodology is presented, including procedures time-delayed phase-space reconstruction and calculation invariants. Four invariants, which include largest Lyapunov exponent, metric entropy, correlation dimension mutual information, all computed. The critical velocities determined sharp increase or decrease obtained invariants at regime transitions. This simple criterion consistent experimental observations. various measures quantitatively in good agreement results holdup data.