Identifying Flow Pattern of Horizontal Oil-Water Two-Phase Flow Based on Multiscale Ordinal Transition Entropy

To comprehend the complex flow structure and elucidate the dynamic evolution mechanism of horizontal oil-water two-phase flow, a more precise and effective method for identifying and analyzing flow patterns is urgently required. In this study, we conduct the experimental simulations under various flow conditions using a small-diameter acrylic glass pipe to gather data. Tap water and 3# industrial oil are used in these simulations. Utilizing a high-speed camera along with the signal response from mini-conductance probes, we classify the standard flow patterns under different conditions. Four typical flow patterns were observed: stratified flow (ST), ST with mixing at the interface (ST&MI), dispersion of oil in water and water (D O/W&W), and dispersion of water in oil and oil in water (D W/O&O/W). Employing an ordinal network, we calculated the entropy index of ultrasonic velocity signals. The mean value and slope of entropy at different scales were combined to construct a joint plane, which yielded satisfactory results in the flow pattern map. Multiscale entropy effectively captures the complex characteristics of two-phase flow and aids in explaining the transition mechanism between flow patterns. Experimental applications confirm the practical utility of this approach. The ordinal transition entropy provides insights into fluid flow characteristics and aids in understanding the flow pattern evolution process.