Rapid On-Line and Non-Invasive Flow Rate Measurement of Gas-Liquid Slug Flow Based Only on Basic Differential Pressure Fluctuations

The gas-liquid two-phase flow is widely encountered in many industrial applications, and its on-line and non-separation flow rate measurement for each phase has plagued the industry for many years. Based on the one-to-one correspondence between differential pressure fluctuations and the movement of Taylor bubbles (TBs) and liquid slugs (LSs), a novel method is proposed to measure the flow rate of gas-liquid slug flow in this article. Theoretical correlations between flow characteristics (velocity, length, and void fraction) and differential pressure fluctuations were quantitatively analyzed by experiments, and the gas-liquid ratio was found to be linearly related to the length ratio of TB and LS. The gas flow rate and liquid flow rate were predicted based on the measured flow characteristics. Experimental tests on air-water two-phase flow show that the proposed method and correlations were effective, and the average relative errors (AREs) in the gas flow rate and liquid flow rate are 5.50% and 5.62% with maximum values of 11.46% and 12.06%, respectively, under the conditions of 0.16-0.56 m/s of air superficial velocity and 0.23-0.48 m/s of water velocity. This method is based only on the differential pressure fluctuations and is on-line, non-separation, noninvasive, and radiation-free, which provides a new cost-effective solution for the flow rate measurement of industrial two-phase flows.

Automated summary

A novel method is proposed in this article to measure the gas-liquid flow rate in slug flow based on the one-to-one correspondence between differential pressure fluctuations and the movement of Taylor bubbles and liquid slugs. The flow characteristics and the gas-liquid ratio were quantitatively analyzed by experiments, and the gas and liquid flow rates were predicted based on the measured flow characteristics. Experimental tests showed that the proposed method and correlations were effective, with average relative errors of 5.50% and 5.62% for gas and liquid flow rates, respectively.