Multi-Scale Time Asymmetry for Detecting the Breakage of Slug Flow Structure

Abstract

We first employ multi-scale time asymmetry (MSA) to analyze typical chaotic signals from Schuster maps and indicate that the MSA method can characterize the distinct time asymmetry of chaotic time series. Then we propose a modified MSA method, i.e., multi-scale weighted time asymmetry, and a novel time asymmetry index to investigate fractal Brownian motion signals and demonstrate its effects on discriminating between fractal signals with different Hurst exponents. Considering that the dynamic behavior of slug flow exhibits chaotic features, we apply the MSA method to analyze experimental signals from a gas-liquid two-phase flow and find that slug flow presents a unique time asymmetric structure. In addition, we further explore the mechanism leading to the formation of time asymmetry in terms of adaptive optimal kernel time-frequency representation. The results suggest that the MSA method can be a useful tool for detecting the complex flow structure underlying a gas-liquid two-phase flow.