Gas-water two-phase flow pattern recognition based on ERT and ultrasound Doppler

Abstract

Two phase flow is widely encountered and of high importance in the manufacturing process and related scientific research. The two-phase flow process is complex, and the flow patterns are jointly determined by the phase fraction and the flow rate. The traditional way of studying the flow patterns uses a single sensor, generally sensing one point, to discover and identify the flow patterns. However, these methods lack a comprehensive description from both the phase distribution and flow velocity. Therefore, it is essential to fuse different detection mechanism sensors to investigate the flow patterns in a more comprehensive way. In this work, an electrical resistance tomography (ERT) sensor measures phase fraction and a continuous wave ultrasound Doppler (CWUD) sensor measures velocity of two-phase flow. For the measurement, data of ERT is treated as a multivariate time-series and a method of graph signal processing named Modular Dirichlet Energy (MDE) is adopted to extract features. The results show that the combination of these two sensors can distinguish horizontal gas-water flow well and lays the foundation for the identification of oil-gas-water three-phase flow.