Identification of gas-liquid two-phase flow patterns in a horizontal pipe based on ultrasonic echoes and RBF neural network

This paper proposes a novel flow pattern identification method using ultrasonic echo signals within the pipe wall. A two-dimensional acoustic pressure numerical model is established to investigate the ultrasonic pulse transmission behavior between the wall-gas and wall-liquid interface. Experiments were also carried out at a horizontal air-water two-phase flow loop to measure the ultrasonic echo pulse signals of stratified flow, slug flow, and annular flow. It is interesting to find that the attenuation of the ultrasonic pulse at the wall-liquid interface is faster than the attenuation at the wall-gas interface. An RBF neural network is constructed for online flow pattern identification. The normalized envelop area and the area ratios of the echo spectrum are selected as the input parameters. The results show that the stratified flow, slug flow, and annular flow can be identified with an accuracy of 94.0%.     

Backlight imaging tomography for slug flows in straight and helical tubes

Backlight imaging tomography is used to experimentally investigate interfacial structures of gas–liquid two-phase flow in circular tubes. The tomography method is based on the attenuation of visible light that causes the inside of the liquid phases to be colored with dye. Increasing the number of light projections provides accurate phase distributions to be reconstructed by a linear backward projection scheme. After the reconstruction performance is examined with numerical simulations for several test cases, the method is applied to slug flows that have complicated 3D interfaces from turbulence. Interfacial structures are compared between straight and helical tubes to determine the effect of centrifugal acceleration. The result demonstrates that centrifugal acceleration provides a liquid-clinging layer on the inner wall against gravity while a high-speed collision of liquid with the top wall happens in a straight tube.     

A reference conductivity fitting method for ERT system to achieve void fraction of gas/liquid flow

The void fraction is one of the key parameters in the measurement of gas/liquid two-phase flow. It can be derived from the absolute conductivity distribution based on Maxwell׳s theory. With Electrical Resistance Tomography (ERT) technology, the absolute conductivity distribution is obtained by multiplying the relative conductivity image with the reference conductivity which is conventionally the liquid conductivity of a gas/liquid flow. Unfortunately the liquid conductivity is not always available. Therefore, a conductivity fitting method is proposed in this paper, to find an optimal reference conductivity, which will be used in substituting the liquid conductivity to reconstruct the quasi-absolute conductivity image. The optimal reference conductivity fitting method is proposed and validated by simulation and experiments under certain flow regimes, e.g. slug flow, annular flow and bubbly flow. The simulation and experimental results show that, independent from prior-knowledge, the fitted quasi-homogenous conductivity is close to the average conductivity of the sensing field. It also leads to a much more accurate estimation of void fraction than the conventional method using liquid conductivity as the reference. With the proposed method, the ERT technique can play a more significant role in the measurement of multiphase flow (MPF).     

Wet gas pressure drop across orifice plate in horizontal pipes in the region of flow pattern transition

As a basis for measuring the mass flow rate of wet gas using differential pressure meters, predicting the pressure drop of a wet gas flowing through orifice plates is important; however, this has not yet been solved satisfactorily, although many studies have reported on that subject. In this study, the pressure drop of wet gas across sharp-edged orifice plates was experimentally investigated in the region of flow pattern transition using air and water as the two phases, and the prediction performance of the available pressure drop models was compared based on the experimental data. The results show that the homogenous flow models overestimate the pressure drop, whereas those models based on the separated flow model often present underestimations. The models reported for wet gas are also incapable of predicting the pressure drop in this region with acceptable accuracy. Through an analysis of the prediction deviations, it is found that the Froude number of the liquid phase has a significant influence on the pressure drop of the wet gas, besides the Froude number of the gas phase. Then, three new correlations that are based on the homogeneous flow, Chisholm model, and Murdock model, respectively, were proposed based on the experimental result.     

Multi-range sensors for the measurement of liquid film thickness distributions based on electrical conductance

The paper presents an approach toward an enhancement of the measuring range of high-speed sensors for the measurement of liquid film thickness distributions based on electrical conductance. This type of sensors consists of electrodes mounted flush to the wall. The sampling of the current generated between a pair of neighboring electrode is used as a measure of the film thickness. Such sensors have a limited measuring range, which is proportional to the lateral distance between the electrodes. The range is therefore coupled to the spatial resolution. The proposed new design allows an extension of the film thickness range by combining electrode matrices of different resolution in one and the same sensor. In this way, a high spatial resolution is reached with a small thickness range, whereas a film thickness that exceeds the range of the high resolution measurement can still be acquired even though on the costs of a lower spatial resolution. A simultaneous signal acquisition with a sampling frequency of 3.2 kHz combines three measuring ranges for the characterization of a two-dimensional film thickness distribution: (1) thickness range 0–600 µm, lateral resolution 2×2 mm², (2) thickness range 400–1300 µm, lateral resolution 4×4 mm², and (3) thickness range 1000–3500 µm, lateral resolution 12×12 mm². The functionality of this concept sensor is demonstrated by tests in a horizontal wavy stratified air–water flow at ambient conditions. Using flexible printed circuit board technology to manufacture the sensor makes it possible to place the sensor at the inner surface of a circular pipe.     

Identification of gas/liquid two-phase flow regime through ERT-based measurement and feature extraction

Gas/liquid two-phase flow is of great importance in various industrial processes. As the most important characteristic of a two-phase flow, the flow regime not only characterizes the flow condition in an explicit way, but also determines the measurement model in each measuring method. Based on the application of Electrical Resistance Tomography (ERT) to a gas/liquid two-phase flow on a vertical pipe, features reflecting the characteristics of gas/liquid two-phase flow are extracted directly from the measured data without reconstruction of the cross-sectional images. The statistical features are derived through time series statistical analysis. Meanwhile features in the wavelet-scale domain are derived through both one-dimensional and two-dimensional wavelet transform. All extracted features are considered as the input of a Support Vector Machine (SVM) algorithm to recognize the flow regime. The preliminary results show that the feature extraction methods of multi-feature fusion and high-dimensional wavelet transform are suitable for the identification of gas/liquid two-phase flow regimes.     

Identification of gas-liquid two-phase flow regime in pipelines with low liquid holdup based on ResNet1D-34

Flow pattern identification is an important topic in multiphase flow research. To overcome the subjectivity of manual identification, intelligent identification of flow patterns has attracted much attention in recent years. Both traditional machine learning methods and deep learning methods have been utilized in this field. However, traditional machine learning methods lack accuracy, and existing deep learning methods mostly rely on artificial feature extraction or complex preprocessing. In this paper, we propose a new method with high accuracy and low preprocessing dependency to solve these issues. We modify ResNet, which has proven high performance in computer vision, to fit the data collected by the wire-mesh sensor system (WMS). Due to its outstanding feature extraction ability, the new model can reach high accuracy with simple normalization as the preprocessing step. Additionally, the model can directly process data at various scales without retraining or rebuilding, which gives it high usability and economic value. The experimental results show that the accuracy of this method can reach 99.58% on our dataset.     

Two-phase flow pattern classification based on void fraction time series and machine learning

Two-phase flow is a complex phenomenon present in several industrial applications such as chemical reactors, power generation, and in the exploration, production, and transport of oil and natural gas. The classification of the flow pattern is a fundamental step in such applications, as it influences several derived parameters and sub-processes such as flow rate, void fraction, and pressure drop estimation. In this paper, we propose an objective approach for classifying flow patterns using time series of void fraction (from a wire-mesh sensor), signal processing and machine learning. As novel approach, the time series is modeled as a stochastic process of independent and identically distributed samples with probability density function described by a Gaussian mixture model. The estimated parameters of the mixture are then fed into a Support Vector Machine (SVM), yielding the flow pattern classification. Tests were performed with a vertical liquid–gas flow database from a 52.3-mm-diameter pipe and the results indicate a great potential for application in real systems. The average accuracy and F-score obtained was higher than 0.94 for different test sets, with standard deviation lower than 0.08 for accuracy a lower than 0.11 for F-Score, demonstrating the efficiency and generalization of the proposed method.     

Effects of flow patterns and salinity on water holdup measurement of oil-water two-phase flow using a conductance method

This research investigates the effects of flow pattern and salinity of oil-water two-phase flow on water holdup measurement using a conductance method. Firstly, vertical upward oil-water two-phase flow experiment is conducted in a 20 mm inner diameter (ID) pipe, in which the salinities of aqueous solutions are set as 151 ppm, 1003 ppm, 2494 ppm and 4991 ppm respectively. Experimental water-cut and mixture velocity are set as 80–100% and 0.0184–0.2576 m/s. In the experiment, three different flow patterns, i.e., dispersed oil-in-water slug flow (D OS/W), dispersed oil-in-water flow (D O/W) and very fine dispersed oil-in-water flow (VFD O/W) are observed and recorded by a high speed camera. Meanwhile, we collect the response of Vertical Multiple Electrode Array (VMEA) conductance sensor excited by a sine voltage signal. The result shows that, for VFD O/W, the water holdup from VMEA sensor shows a satisfied agreement with that of quick closing valve (QCV) method under certain salinities, i.e., 1003 ppm as well as 2494 ppm. For D OS/W flow and D O/W flow characterized by dispersed oil droplets with various sizes, considerable deviations of water holdup between VMEA sensor and QCV method under four kinds of salinity aforementioned are presented. Afterward, according to experimental analysis along with theoretical deviation, it is concluded that the deviation of the measurement system reaches its minimum when reference resistance in the measurement circuit and salinity of the aqueous solution satisfy constraint conditions, and the accuracy of water holdup using the conductance method can be improved through adjusting reference resistance to match the salinity of water phase. Finally, the recurrence plot algorithm is utilized to identify typical flow patterns mentioned above and it shows satisfied results on comprehending the discrepancies among different flow patterns, demonstrating that the recurrence plot algorithm can be effectively applied in flow pattern identification regarding oil-water flows.     

基于多频带谱熵的水平气液两相流结构复杂性分析

针对传统功率谱熵只能刻画总体系统结构复杂性问题,提出既能从宏观角度又能从微观角度反映系统结构复杂性的多频带谱熵的分析方法。分析几种典型信号的多频带谱熵特征,验证多频带谱熵方法的可行性及抗干扰能力;利用环形电导传感器阵列获取动态实验测试数据,计算水平气液两相流波状流、塞状流和弹状流含水率波动的多频带谱熵值,分析3种流型结构复杂性随频带因子变化的演变规律。实验结果表明:波状流的多频带谱熵最低,弹状流多频带谱熵值最高;塞状流的多频带谱熵居于波状流与弹状流之间。3种流型在分析频率为0~8.3 Hz时,呈现近似线性变化的结构复杂性特征以及较强的流型区分度,可以作为流型识别的准则。     

LDV measurement,flow visualization and numerical analysis of flow distribution in a close-coupled catalytic converter

Results from an experimental study of flow distribution in a close-coupled catalytic converter (CCC) are presented. The experiments were carried out with a flow measurement system specially designed for this study under steady and transient flow conditions. A pitot tube was a tool for measuring flow distribution at the exit of the first monolith. The flow distribution of the CCC was also measured by LDV system and flow visualization. Results from numerical analysis are also presented. Experimental results showed that the flow uniformity index decreases as flow Reynolds number increases. In steady flow conditions, the flow through each exhaust pipe made some flow concentrations on a specific region of the CCC inlet. The transient test results showed that the flow through each exhaust pipe in the engine firing order, interacted with each other to ensure that the flow distribution was uniform. The results of numerical analysis were quali-tatively accepted with experimental results. They supported and helped explain the flow in the entry region of CCC.     

Simulation and experimental research on the hydraulically powered fast switching 3-way valve for pVTt gas flow standard

pVTt has been employed as the primary gas flow standard in many countries and the performance of the inlet valve significantly influences the verification. This paper not only proposes a hydraulically powered fast switching 3-way valve but also introduces its structure and working principle. Furthermore, it builds up the flow field simulation model and conducts the dynamic simulation of its working process. As shown by the results, the designed fast switching 3-way valve could achieve the fast switching of the critical flow. In addition, the “overlapping” mode and the “non-overlapping” mode of the 3-way valve are analysed contrastively. The result reveals that the flow field of the “non-overlapping” mode is more stable. Finally, the device is tested according to the pVTt verification procedure. As indicated by the results, the switching time of two fast switching 3-way valves corresponding to DN150mm and DN100mm caliber could be controlled within 55 ms, which is consistent with the simulation results.     

A flow rate measurement method for horizontal oil-gas-water three-phase flows based on Venturi meter,blind tee,and gamma-ray attenuation

Online horizontal oil-gas-water three-phase flow rate monitoring is essential for reliable operations during industrial production. A flow rate measurement method is developed in horizontal oil-gas-water three-phase flows by combining a blind tee, a Venturi meter, and a gamma-ray densitometer. The blind tee is installed at the test section entrance to homogenize the mixture by transforming the horizontal flow to a vertical upward flow. The Venturi meter is used for the total flow rate measurement. The dual-energy gamma-ray densitometer is used for phase holdup measurement. In the present method, blind-tee mixing effects and oil-water mixture slip behavior is essential, which were experimentally analyzed in this work. The phase inversion was found in the oil-water mixture with the increasing of the oil volume fraction. Besides, the addition of the gas reduces the oil-water slip ratio. For the range of 0–35% and 65–100% oil fraction in the oil-water liquid, the oil-water mixture can be well treated as a pseudo homogenous liquid with a slip ratio of 0.9–1.1. A three-phase flow rate model is then established for these conditions. The method was validated by horizontal oil-gas-water three-phase flows with average relative errors of 3.2% for the total flow rates, 4.3% for the gas flow rates, 11.5% for the oil flow rates, and 7.8% for the water flow rates.     

Performance model evaluation of turbine flow meter in vertical gas-liquid two-phase flows

Aiming at the need for flow measurement of gas-liquid flows in domestic gas well production, this paper proposes a measurement method based on the combination of the turbine flow meter (TFM) and a rotating electric field conductance sensor (REFCS). In experiments, the REFCS is used for the measurement of the gas holdup. To verify the applicability of the TFM models investigated in the previous study, for the modeling part, the mass, momentum and torque models are evaluated in vertical upward gas-liquid two-phase flows. In our model test, the meter factor model of TFM considers the effects of the slip ratio between the gas and liquid phases and flow patterns. In particular, the gas holdup involved in calculating the slip ratio in the model evaluation is obtained from the REFCS measurements. Model test results show the torque model has better volumetric flow rate prediction accuracy than the mass and momentum models. In the present study, the ranges of the liquid and gas phases are Q_w = 2–30 m³/d and Q_g = 1–16 m³/d, it was found that the average absolute deviation (AAD) in the predicted volume flow rate is equal to 1.23 m³/d and the average absolute percentage deviation (AAPD) is equal to 7.69%. The evaluated results presented in this paper will allow better estimates of the volumetric flow rates of gas-liquid flows based on the combined TMF and REFCS measurements during the monitoring of gas well production.     

热、气、固耦合流动数值模拟在管式膨胀系统中的研究

应用相似理论和合理的相似比,得出6400ks／h高温管式膨胀设备的结构模型。采用“小滑移”模型,并利用ANSYS／FLOTRAN CFD对其进行了数值模拟,得出整个三维模型流场的温度、速度及气固两相的分布,与2400ks／h高温管式膨胀设备实测值吻合较好。     

对JJG1003—2005《流量积算仪检定规程》一般气体流量温压补偿密度问题的一点探讨

JJG 1003—2005《流量积算仪检定规程》附录B中计算气体标准体积流量的表达式是两个,在制作流量积算仪或流量积算模块时按照这个思路建立数学模型,会增加一般气体流量温压补偿在具体实施中的复杂性。在建立流量积算仪数学模型时可以简化模型;在使用较复杂数学模型的流量积算仪时可以对有关参数的设定进行灵活地处理。