Identification of gas-liquid two-phase flow patterns using non-return valve sound signals

Because of the complexity of crude oil flow, identification of gas-liquid two-phase flow pattern has not been completely solved nowadays. This paper proposed a new classification method of gas-liquid two-phase flow patterns using the sound signals produced by the ball impacting on the valve wall with fluid movement in a non-return valve. A visual detection system of gas-liquid two-phase flow was built to study four typical flow patterns including liquid flow, gas flow, slug flow and churn flow pattern. The sound signals from non-return valve under different conditions were captured and processed with Empirical Mode Decomposition (EMD) and Hilbert-Huang Transform (HHT). Flow-pattern maps were constructed with the coordinates of the energy of Intrinsic Mode Functions (IMF) and Hilbert marginal spectrum. Verification tests demonstrated that the correct identification rates are better than 98.1% with the developed flow pattern maps based on sound signal analysis of non-return valve.