(17c) Comparison Between Two-Phase Flow Models and Wire Mesh Sensor Measurements in Medium and Large Diameter Pipes

Wire mesh sensors (WMS) are recently developed fast imaging instruments that have been used to investigate two-phase flows. WMS  measure either the electrical conductivity or capacitance of the two phases, allowing  the local void fraction to be inferred [1, 2]. The  WMS have been benchmarked against the gamma densitometer, a high-resolution  instrument widely used in the industry for two-phase flow metering; therefore providing confidence with  experimental data [3]. In this work, we compare the results of an extensive WMS experimental campaign,with a number of two-phase flow models based on the drift-flux approach [4]. Void fraction is a critical parameter in two-phase flows. Drift-flux models are semi-mechanistic models that compare to fully mechanistic models use an explicit equation for the void fraction, thus reducing computational time significantly. Moreover, most mechanistic models cannot capture all the flow patterns and pipe inclinations within a single framework without adding complexity and computational overhead. A number of authors have developed drift-flux models for a range of flow patterns and inclinations which seem suitable for bubbly, slug and churn flows [5, 6].

The experiments consisted of WMS measurements using water/air and oil/air in medium and large diameter pipes with pipe inclinations ranging from vertical to horizontal. We compare the performance of several drift-flux closure equations developed for a range of flow patterns and inclinations. The models have been ranked and the empirical parameters of the best performers have been optimized. Additional comparisons between the final optimized models, the experimental data and the mechanistic model OLGA-S are also presented in this study.

[1] Prasser et al. (1998) A new electrode-mesh tomograph for gas-liquid flows, Flow Meas. Instrum.,9, 111-119

[2] Da Silva et al. (2007) Capacitance wire-mesh sensor for fast measurement of phase fraction distribution, Meas. Sci. Technol., 18, 2245-2251

[3] Sharaf et al. (2011) Comparison between wire mesh sensor and gamma densitometry void measurements in two-phase flow, Meas. Sci. Technol., 22, 1-13

[4] Ishii and Hibiki (2006) Thermo-fluid Dynamics of Two-Phase Flow, Springer

[5] Shi et al. (2005) Drift-flux Modeling of Two-Phase Flow in Wellbores, SPE J., 24-33

[6] Choi et al. (2012) An Efficient Drift-flux Closure Relationship to Estimate Liquid Holdups of Gas-Liquid Two-Phase Flow in Pipes, Energies, 5, 5294-5306