Experimental study and evaluation of the performance of single phase flow meters in wet gas flows

Abstract

Wet gas, a mixture of gas and small liquid fractions, poses significant challenges for accurate flow measurement, which is critical in the energy, chemical, and power industries. While Coriolis meters are widely used, their application in wet gas environments faces limitations, including significant over-reading (OR) and a lack of optimal installation guidelines. These issues are believed to be part of a fundamental question: how are the meter behaviours affected by multiphase flow dynamics, such as flow patterns, flow conditions, and phase distributions? This research evaluates the performance of three Coriolis meters in vertical and horizontal 2” pipelines using air, water, and oil to simulate wet gas conditions. Various sensor orientations (0°, 40°, 90°, and 180°) were tested in the horizontal pipeline, with 40° and 90° proposed as novel configurations. The experiments generated a new dataset, enabling detailed analysis of Coriolis behaviour under wet gas conditions. A novel wet gas model was developed based on internal Coriolis parameters, demonstrating robustness and predictive accuracy. The 40° orientation was identified as optimal for horizontal wet gas flow, minimising OR and improving measurement precision. Further investigations revealed the significant influence of flow patterns, with slug flow causing the greatest deviations. A novel correlation was established between the damping factor, XLM , and total mass flow rate, enabling the accurate prediction of gas and liquid flow rates. In high Froude number and annular flow conditions, the model achieved Mean Absolute Percentage Errors (MAPE) of 3.9% for gas flow rate and 4.3% for liquid flow rate, with an uncertainty of 2.7%. The study also innovatively employed the damping factor to estimate liquid holdup, validated against conductance ring measurements. In vertical annular flow, a robust correlation was established between the damping factor and liquid holdup (𝜀!), achieving percentage errors below ±10% at higher gas flow rates. This dual functionality—simultaneously measuring flow rates and void fraction with a single-phase meter—represents a significant advancement for reducing equipment needs and enhancing efficiency. For the first time, frequency data were extracted to characterize the medium, introducing a "distance" parameter from frequency waveform signals. Combined with damping factor data, this enabled a medium detection index with classification accuracy exceeding 98%.