Research On Gas-liquid Two-phase Flow Pattern Analysis Method Based On Complex Network

Applied in a wide range of industrial production, the two-phase flow process is dynamic, non-equilibrium, and complex, making it impossible to obtain a clear understanding of its evolutionary dynamics via the research approach based on near-equilibrium hypothesis and linear processing. Therefore, new methods should be introduced to study the dynamic characteristics and their evolutionary patterns of two-phase flow. As a method and tool for studying complex systems, complex network can be used to not only explore the important information features embedded in the time series of two-phase flow, but also investigate the complex, nonlinear dynamic systems of two-phase flow, which cannot be accurately described by the theoretical model. In this paper, the gas liquid two-phase flow in vertical upward pipe was taken as research object, and the flow pattern, evolution, and bubble coalescence mechanism of the gas liquid two-phase flow were studied with the complex network theory, by collecting the time series of the pressure-difference fluctuation of gas liquid two-phase flow. On this basis, the dynamic characteristics of gas liquid two-phase flow were explored. The findings achieved in this study were as follows:1. For the dynamic characteristics of the gas liquid two-phase flow in a vertical upward pipe, we proposed a method of establishing the complex network based on the similarity in time series of the pressure-difference fluctuation. By exacting the energy features of time series of the pressure-difference fluctuation with the empirical mode decomposition, the energy distribution of various flow patterns in different time scales was obtained. By establishing the complex network of air-water two-phase flow pattern in a vertical upward pipe, a community structure dividing method based on AP clusters was proposed and used to analyze the community structure of the complex network of air-water two-phase flow pattern in a vertical upward pipe, thereby obtaining the correspondence between the community and different flow patterns in the complex network of flow pattern and thus identifying 5 patterns of flow patterns in a vertical upward pipe, including the transitional flow.2. For the unstable, non-equilibrium and complex characteristics of the evolutionary process of gas liquid two-phase flow patterns, we proposed a method of establishing the complex network of the pattern evolution in a vertical upward pipe. With the air water two-phase flow in a vertical upward pipe as the research object, based on the similar dynamic migration of the pattern evolution of gas liquid two-phase flow, the complex network corresponding to the pattern evolution of various flow patterns was established and analyzed. The complex network of pattern evolution of gas liquid two-phase flow in a vertical upward pipe was found with obvious scaleless and small world properties. The corresponding network of different flow patterns showed different community structures, and the statistical magnitude of the network was consistent with the evolution of flow patterns, clearly revealing the dynamic characteristics of pattern evolution of gas liquid two-phase flow in a vertical upward pipe.3. In order to reveal the bubble coalescence mechanism in the gas liquid two-phase flow and its dynamic characteristic in the phase space, we proposed a method of establishing the complex network of phase space based on correlation dimension by reconstructing the phase space. A phase space complex network of flow patterns was therefore established. Through analyzing the dynamic characteristics of the phase space complex network corresponding to different dynamic systems, it was found that, attracted by the unstable cyclic track, the phase space complex network corresponding to chaotic dynamic system showed obvious small world property, with its network structure and statistical parameters closely related to the phase space dynamic characteristic or the original system. On this basis, by analyzing the phase space complex networks corresponding to the bubble flow, slug flow, and mixed flow of gas liquid two-phase flow in a vertical upward pipe, it was found that the network structure of the phase space complex network was inherently related to the flow structure of the gas liquid two-phase flow in a vertical upward pipe, and the network density was sensitive to bubble coalescence. The structure and density of the phase space complex network could effectively analyze the attracting property of unstable cyclic track in the phase space of different flow patterns, providing a new perspective for revealing the bubble coalescence mechanism in the gas liquid two-phase flow.