Three-phase gas-oil-water flow is an important type of flow present in petroleum extraction and processing. This paper reports a novel threshold-based method to visualize and estimate the crosssectional phase fraction of gas-oil-water mixtures. A 16x16 dual-modality wire-mesh sensor (WMS) was employed to simultaneously determine the conductive and capacitive components of the impedance of fluid. Then, both electrical parameters are used to classify readings of WMS into either pure substance (gas, oil or water) or two-phase oil-water mixtures (foam is neglected in this work). Since the wire-mesh sensor interrogates small regions of the flow domain, we assume that the three-phase mixture can be segmented according to the spatial sensor resolution (typically 2-3 mm). Hence, the proposed method simplifies a complex three-phase system in several segments of single or two-phase mixtures. In addition to flow visualization, the novel approach can also be applied to estimate quantitative volume fractions of flowing gas-oil-water mixtures. The proposed method was tested in a horizontal air-oil-water flow loop in different flow conditions. Experimental results suggest that the threshold-based method is able to capture transient three-phase flows with high temporal and spatial resolution even in the presence of water-oil dispersion regardless of the continuous phase.

No abstract

In this paper, the front-end circuit of a capacitance wire-mesh sensor (WMS) is analyzed in detail and a new methodology to tune its feedback gains is reported. This allows, for the first time, a capacitance WMS to be able to provide linear measurements of multiphase fluids with electrical conductivity greater than 100 𝜇S/cm, which is particularly important for tap water, where the conductivity is typically in between 100 S/cm and 500 𝜇S/cm. Experimental and numerical results show that the selected gains using the proposed methodology contribute to suppress cross-talk and energy losses, which in turn, reduces considerably the deviation of the conductivity measurement and the estimation of derived flow parameters, such as local and average phase fraction.

Wire-mesh sensors are well-established scientific instruments for measuring the spatio-temporal phase distribution of two-phase flows based on different electrical conductivities of the phases. Presently, these instruments are also applied in industrial processes and need to cope with dynamic operating conditions increasingly. However, since the quantification of phase fractions is achieved by normalizing signals with respect to a separately recorded reference measurement, the results are sensitive to temperature differences in any application. Therefore, the present study aims at proposing a method to compensate temperature effects in the data processing procedure. Firstly, a general approach is theoretically derived from the underlying measurement principle and compensation procedures for the electrical conductivity from literature models. Additionally, a novel semi-empirical model is developed on the basis of electrochemical fundamentals. Experimental investigations are performed using a single-phase water loop with adjustable fluid temperature in order to verify the theoretical approach for wire-mesh sensor applications and to compare the different compensation models by means of real data. Finally, the preferred model is used to demonstrate the effect of temperature compensation with selected sets of experimental two-phase data from a previous study. The results are discussed in detail and show that temperature effects need to be handled carefully—not merely in industrial applications, but particularly in laboratory experiments.