An investigation of oil–water two-phase flow instability using multivariate multi-scale weighted permutation entropy

Abstract

The present study is devoted to the investigation of oil–water two-phase flow instability using multivariate multi-scale weighted permutation entropy (MWMPE). Four typical multivariate multi-scale entropies, namely MWMPE, multivariate multi-scale permutation entropy (MMPE), multivariate multi-scale sample entropy (MMSE) and multivariate multi-scale approximate entropy (MMAE) are applied in Lorenz system with the addition of different kinds of noises. The comparison results indicate that MWMPE presents the superiorities of being sensitive to the variation in scale, showing a monotonous increasing trend as well as the best anti-noise ability. Accordingly, with the fluctuating signals from an eight-electrode rotating electrical field conductance sensor used for the measurement of oil–water flows, we extract MWMPE for the whole experimental flow conditions, in terms of which the effects of mixture velocity and water-cut on oil–water two-phase flow instability are illuminated. Our research provides an effective method for uncovering the underlying evolution instability of the flow structures in oil–water flows.