Vortex wet gas phase separation flow measurement method based on liquid film flow parameter modeling

Abstract

The invention relates to a vortex street wet gas phase separation flow measurement method based on liquid film flow parameter modeling. The method uses signals collected by a vortex street piezoelectric sensor and a liquid film flow parameter sensor, combines dimensionless liquid film thickness and disturbance wave velocity to establish an overreading prediction model, predicts the phase separation flow according to an overreading coefficient OR, and realizes wet gas phase separation flow measurement. The method comprises the following steps: collecting two-phase pressure, two-phase temperature, a timing signal output by a vortex street piezoelectric sensor, and an upstream and downstream liquid film thickness sequence output by a liquid film flow parameter sensor; calculating gas density and liquid density; obtaining a vortex street shedding frequency; obtaining an average liquid film thickness; performing cross-correlation calculation to obtain a transit time; obtaining a dimensionless liquid film thickness and a dimensionless disturbance wave velocity; calculating an indication result of a gas phase volume flow; calculating a vortex street overreading coefficient OR; calculating a real gas flow; and calculating a liquid phase flow.

Description

Technical Field

[0001] The invention belongs to the field of wet gas phase separation flow measurement, and relates to a vortex street wet gas phase separation flow measurement method based on liquid film flow parameter modeling. Background Art

[0002] Wet gas is widely used in engineering applications such as natural gas, steam boilers, and nuclear reactors. [1] , both gas and liquid flow rates in wet gas are important, reflecting the output of a single well, especially the measurement of liquid flow, which is often overlooked [1] In recent years, the rapid development of industries such as petroleum, natural gas, power, chemical industry, water conservancy, and aerospace has led to a gradual increase in the requirements for two-phase flow measurement. [2][3] The use of traditional single-phase instruments to measure gas-liquid two-phase flow has gradually become unable to meet the needs of industrial production, so the online measurement technology of two-phase flow has come into being.

[0003] In the measurement of wet gas phase flow, vortex flowmeter is widely used for online measurement of wet gas phase flow due to its robustness, economy, high range ratio and low pressure loss. However, when using traditional single-phase vortex flowmeter for fluid flow measurement, a small amount of liquid in the fluid will affect the output signal, resulting in a positive deviation (overreading, OR) in the predicted result of gas flow. [4] The maximum measurement error caused by over-reading can reach 20% [4] If necessary corrections are not made, the measurement accuracy of the vortex flowmeter will be seriously affected. In order to improve the measurement accuracy of the vortex flowmeter in wet gas, overreading needs to be compensated. The liquid phase in wet gas exists in two forms: droplets and liquid films. Vortex overreading is closely related to the two. Due to the complexity of two-phase flow, there is currently no unified vortex overreading prediction formula. Zhang Jinjing [7] Based on the vortex-induced flow field theory, the conversion relationship between vortex street frequency and gas phase volume flow rate was studied, and the theoretical results were verified in the gas-liquid two-phase flow experiment at room temperature. [8]The stability and instability of the two-phase vortex wake were explored from the perspective of combining simulation and experiment, and the vortex over-reading prediction model was obtained using the two-phase flow parameters. The above model obtains the intrinsic influence mechanism between droplets and vortex streets from the perspective of theoretical research, and achieves good prediction results. However, the droplet parameters (droplet loading, droplet diameter) cannot be truly measured online, and most of the liquid phase of the wet gas exists in the form of a liquid film. Therefore, it is very necessary to explore the influence of the liquid film in the wet gas on the vortex over-reading from the perspective of liquid film flow parameters. Among them, patent CN202110129896.3 explores the influence of liquid film thickness on vortex over-reading. However, only the liquid film thickness information cannot fully characterize the liquid film flow state. Therefore, it is urgent to establish a complete set of liquid film flow parameter measurement system to accurately obtain more liquid film information, and then explore the relationship between liquid film flow parameters (liquid film thickness, disturbance wave velocity) and vortex street signals, and finally realize the online and accurate measurement of phase separation flow in annular mist flow.

[0004] References

[0005] [1].D.He,S.Chen,B.Bai,Online measurement of liquid flow rate in wetgas based on pressure loss ratio of V-cone flowmeter[J].Chinese Journal ofScientific Instrument,39(7):235-244,2018.

[0006] [2].DGStewart,The evaluation of dry gas meters in wet gasconditions[J].DTI NMSD Flow Programme,London,UK,NEL Rep.No 2002-100,November 2002.

[0007] [3]. Fu Yingjie, Wei Yingjie, Zhang Jiazhong, et al. Computational analysis of the gas-liquid two-phase flow field in the nozzle[J]. Journal of Harbin Institute of Technology, 2010, 42(9): 1363-1368.

[0008] [4]. Sun Xianghao. Study on the influence of steam dryness on steam flooding effect[J]. Inner Mongolia Petrochemical, 2012, 38(24): 39-40.

[0009] [5]. Lu Jia, Shi Yunxiao, Zheng Yonghui. Measuring the mass flow rate of steam using vortex flowmeter[J]. Petrochemical Automation, 2006(4):79-80.

[0010] [6]. Yu Xiaoyin, Xia Weilin. Effect of severe water content in steam on vortex flowmeter[J]. Chemical Engineering and Pharmaceutical Engineering, 2008, 29(4): 51-54.

[0011] [7].Wang C,Liu Q.Influence of droplet diameter on vortex flow meterin steam flow measurement[C].2011Second International Conference on MechanicAutomation and Control Engineering.IEEE,2011:4994-4997.

[0012] [8]. Li Jinxia. Research on the measurement characteristics and stability of wet gas vortex street[D]. Tianjin: Tianjin University, 2020.

[0013] [9]. Zhang Jinjing. Research on the characteristics of vortex flowmeter in gas-liquid two-phase flow[D]. Tianjin: Tianjin University, 2015. Summary of the invention

[0014] The present invention provides a more accurate and easy-to-implement online vortex wet gas phase flow measurement method. The present invention uses a vortex flowmeter and a liquid film flow parameter measurement system, combines dimensionless liquid film thickness and disturbance wave velocity to establish an over-reading prediction model, predicts the phase flow rate according to the over-reading coefficient OR, and realizes wet gas phase flow measurement. The technical solution is as follows:

[0015] A vortex wet gas phase flow measurement method based on liquid film flow parameter modeling, using the signals collected by the vortex piezoelectric sensor and the liquid film flow parameter sensor, combined with the dimensionless liquid film thickness and disturbance wave velocity to establish an over-reading prediction model, predicting the phase flow according to the over-reading coefficient OR, and realizing wet gas phase flow measurement, including the following steps:

[0016] 1) Collect the two-phase pressure P, two-phase temperature T, the time series signal s(t) output by the vortex piezoelectric sensor, and the upstream and downstream liquid film thickness series δ1(t) and δ2(t) output by the liquid film flow parameter sensor;

[0017] 2) Calculate the gas density ρ by the two-phase pressure P and the two-phase temperature T g and liquid density ρ l ; Perform fast Fourier transform on the timing signal s(t) output by the vortex piezoelectric sensor to obtain the vortex shedding frequency f VS; The upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) output by the liquid film flow parameter sensor are averaged to obtain the average liquid film thickness δ1 and δ2;

[0018] 3) Given the center distance L of the upstream and downstream liquid film flow parameter sensors, the upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) are cross-correlated to obtain the transit time τ0, and the disturbance wave velocity is obtained by solving formula (1):

[0019]

[0020] 4) According to formula (2) and formula (3), the characteristic parameters of the liquid film are dimensionless to obtain the dimensionless liquid film thickness and dimensionless disturbance wave velocity:

[0021]

[0022] v w * =v w / u sg,max (3)

[0023] Where D is the pipe diameter; δ1 is the average liquid film thickness measured by the upstream conductivity ring sensor; δ * is the dimensionless liquid film thickness, v w * is the dimensionless disturbance wave velocity, v w is the real disturbance wave velocity, u sg,max is the maximum gas superficial velocity within the experimental range;

[0024] 5) Calculate the gas phase volume flow rate indication result Q according to formula (4): g,tp :

[0025]

[0026] Among them, f VS is the vortex shedding frequency, K v is the instrument factor of the vortex flowmeter in single-phase gas;

[0027] 6) Calculate the vortex street over-reading coefficient OR according to formula (5):

[0028]

[0029] Wherein, k1 is a constant coefficient, n1 and n2 are constant power exponents, and their specific values ​​are obtained by fitting the function form described in formula (5);

[0030] 7) Calculate the actual gas flow rate Q by formula (6): g

[0031]

[0032] 8) Calculate Q according to formula (7) g The corresponding gas superficial velocity u sg

[0033]

[0034] 9) Based on the experimental data, a mathematical model between OR and the Weber number of gas and liquid is established. The parameterized expression of the model is:

[0035]

[0036] Where k2 is a constant coefficient, n3 and n4 are constant power exponents; the Weber number of a gas is in the form of The Weber number of a liquid is defined as Δρ is the density difference between gas and liquid; u sg Indicates the superficial velocity of gas; u sl is the apparent velocity of the liquid; D is the nominal diameter of the pipeline, σ is the surface tension;

[0037] 10) From the gas superficial velocity u sg Calculate the Weber number of the gas We g , and then use formula (8) to infer the liquid phase Weber number We l

[0038]

[0039] 11) Liquid flow rate Q l Calculated by formula (10):

[0040]

[0041] At this point, the phase separation flow model established using liquid film thickness and disturbance wave velocity has been solved. And this method has the following advantages:

[0042] 1. Simple modeling, simple calculation, low cost, and online measurement;

[0043] The wet gas phase flow measurement can be achieved by measuring relevant parameters using a vortex flowmeter, a liquid film flow parameter sensor, a pressure sensor and a temperature sensor. This method is simple to operate, low in cost, and can achieve online measurement.

[0044] 2. High prediction accuracy;

[0045] This method was used to predict gas and liquid flow rates under wet gas conditions. In the example, the relative error of the gas volume flow rate prediction was within ±1.5%; the relative error of the liquid volume flow rate prediction was within ±6%.

[0046] 3. It can realize wet gas phase flow measurement.

[0047] This method uses dimensionless liquid film thickness as the main parameter and dimensionless disturbance wave velocity as the auxiliary parameter to establish the prediction formula of OR over-reading, compensate the uncorrected gas phase flow rate, and realize the measurement of gas phase flow rate. According to the experimental data, a mathematical model between OR and the Weber number of gas and liquid is established to solve the liquid phase flow rate, and finally realize the measurement of wet gas phase flow rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 :Schematic diagram of liquid film flow parameter measurement system

[0049] Figure 2 :Signal acquisition flow chart

[0050] Figure 3 : Flowchart of cross-correlation velocity calculation

[0051] Figure 4 :Flow chart of vortex wet gas phase separation flow measurement based on liquid film flow parameter modeling

[0052] Figure 5 :Relationship diagram between overreading coefficient OR and dimensionless liquid film thickness and disturbance wave velocity

[0053] Figure 6 :The relationship between dimensionless disturbance wave velocity and gas-liquid phase Weber number

[0054] Figure 7 :Relationship diagram between overreading coefficient OR and gas-liquid phase Weber number

[0055] Figure 8 :Schematic diagram of gas phase volume flow prediction error

[0056] Fig. 9 :Schematic diagram of liquid phase volume flow prediction error DETAILED DESCRIPTION

[0057] In order to further understand the features, technical means, specific objectives and functions of the present invention, the specific implementation methods of the present invention are described in detail below with reference to the accompanying drawings.

[0058] The present invention utilizes a vortex flowmeter and a liquid film flow parameter measurement system to measure the average liquid film signals, disturbance wave signals and uncorrected gas phase flow rates in the upstream and downstream, respectively. In view of the vortex over-reading problem, an over-reading prediction model is established in combination with the dimensionless liquid film thickness and disturbance wave velocity, and the uncorrected gas phase flow rate is compensated for by the over-reading coefficient OR, thereby realizing the measurement of the gas phase flow rate. In order to solve the liquid phase flow rate, a mathematical model between OR and the gas-liquid Weber number is established based on experimental data, and the liquid phase flow rate is solved to obtain the liquid phase flow rate, thereby realizing the measurement of the phase flow rate under wet conditions. The specific solution steps are as follows:

[0059] 1) Collect the two-phase pressure P, two-phase temperature T, the time series signal s(t) output by the vortex piezoelectric sensor, and the upstream and downstream liquid film thickness series δ1(t) and δ2(t) output by the liquid film flow parameter sensor;

[0060] 2) The gas density ρ can be calculated by the two-phase pressure P and the two-phase temperature T g and liquid density ρ l ; Perform fast Fourier transform (FFT) on the timing signal s(t) output by the vortex piezoelectric sensor to obtain the vortex shedding frequency f VS ; The upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) output by the liquid film flow parameter sensor are averaged to obtain the average liquid film thickness δ1 and δ2;

[0061] 3) Given that the center distance L between the upstream and downstream sensors is 50 mm, which is converted into the international unit of 0.05 m, the upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) are cross-correlated to obtain the transit time τ0. Substitute the transit time τ0 and the center distance L between the upstream and downstream sensors into formula (1) to obtain the disturbance wave velocity;

[0062]

[0063] 4) According to formula (2) and formula (3), the characteristic parameters of the liquid film are converted into dimensionless parameters. The pipeline diameter D = 15 mm, which is converted into 0.015 m in international units. δ1 is the average liquid film thickness measured by the upstream conductivity ring sensor. Substituting the pipeline diameter D and δ1 into formula (2) yields the dimensionless liquid film thickness δ * ;v w is the real disturbance wave velocity, u sg,max is the maximum gas apparent velocity within the experimental range, which is set to 40 m / s in this paper. The real disturbance wave velocity v w and the maximum gas superficial velocity u sg,max Substituting into formula (3) we get the dimensionless disturbance wave velocity;

[0064]

[0065]

[0066] Among them, δ * is the dimensionless liquid film thickness, v w * is the dimensionless disturbance wave velocity.

[0067] 5) The vortex shedding frequency f VS and the instrument factor K in single-phase gas v Substituting into formula (4) we get the gas phase volume flow rate indication result Qg,tp

[0068]

[0069] Among them, K v is the instrument factor of the vortex flowmeter in single-phase gas (m -3 );

[0070] 6) According to the experimental data, a mathematical model between the dimensionless liquid film thickness and the dimensionless disturbance wave velocity and OR is established, and the vortex street over-reading coefficient is obtained by substituting it into formula (5):

[0071]

[0072] 7) The gas phase volume flow rate indication result Q g,tp Substituting the vortex street over-reading coefficient OR into formula (6) yields the actual gas flow rate Q g

[0073]

[0074] 8) The actual gas flow Q g Substituting into formula (7) we get Q g The corresponding gas superficial velocity u sg

[0075]

[0076] 9) Based on the experimental data, a mathematical model between OR and the Weber number of gas and liquid is established. The parameterized expression of the model is:

[0077]

[0078] 10) The gas superficial velocity u sg Substituting into formula (9) we get the gas Weber number We g , where the surface tension of water is 0.072N / m

[0079]

[0080] 11) According to formula (8), the liquid phase Weber number We is obtained by reverse calculation l

[0081]

[0082] 12) The liquid phase Weber number We l Substituting into formula (11) we get the final liquid phase flow rate Q l

[0083]

[0084] At this point, the phase separation flow model established using liquid film thickness and disturbance wave velocity has been solved.

[0085] This example is a vortex street wet gas phase flow measurement method based on liquid film flow parameter modeling. The following is a specific implementation in wet gas measurement. Using the multi-parameter adjustable mist flow experimental system in patent 201810644726.7, a wet gas measurement model is proposed, and its experimental range is: normal temperature: 25~35℃, pressure: 150~350kPa, gas volume flow: 12~24m 3 / h, gas density: 1.79~4.24kg / m 3 , liquid volume flow rate: 0.5~4.5ml / s, gas apparent velocity: 18.86~37.73m / s, liquid apparent velocity: 0.0028~0.028m / s. The liquid film flow parameter measurement system is as attached Figure 1 As shown, it consists of two upstream and downstream conductivity loops, FPGA, IV amplifier module and host computer.

[0086] The sensor signals are collected, including the two-phase pressure P output by the pressure sensor, the two-phase temperature T output by the temperature sensor, the timing signal s(t) output by the vortex flowmeter, and the upstream and downstream liquid film thickness sequences δ1(t) and δ2(t) output by the liquid film flow parameter sensor. The vortex signal is collected by the NI-USB-6353 data acquisition card. The sampling frequency of the vortex signal is 100kHz, and the sampling experiment time for each set of data is 10s. The signal acquisition flow chart is as shown in the attached figure. Figure 2 As shown. The upstream and downstream liquid film thickness sequences δ1(t) and δ2(t) are measured simultaneously by two conductivity ring sensors, with a sampling frequency of 32MHz and a sinusoidal excitation signal frequency of 500KHz. 64 samples can be taken in each sinusoidal cycle, and the liquid film signal is demodulated after sampling 512 sampling points in 8 sinusoidal cycles. Then the gas density ρ can be calculated by the two-phase pressure P and the two-phase temperature T. g and liquid density ρ l ; Perform fast Fourier transform (FFT) on the time series signal s(t) output by the vortex piezoelectric sensor to obtain the vortex shedding frequency f VS ; The upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) output by the liquid film flow parameter sensor are averaged respectively to obtain the average liquid film thickness δ1 and δ2.

[0087] The velocity of the disturbance wave is closely related to the transformation between the liquid film, so it is necessary to accurately obtain the velocity of the disturbance wave and make a prediction. It is known that the center distance L of the upstream and downstream sensors is 50 mm. The upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) are cross-correlated to obtain the transit time τ0. The flowchart of the cross-correlation calculation is shown in the attached figure. Figure 3As shown, the disturbance wave velocity can be obtained by solving formula (1). The characteristic parameters of liquid film flow are mainly used to characterize the flow characteristics of liquid film in wet gas. The main parameters include the disturbance wave velocity v w , disturbance wave frequency f w and the thickness of the liquid film δ. The characteristic parameters of the liquid film are often converted into dimensionless parameters for consideration. The dimensionless liquid film thickness and dimensionless disturbance wave velocity can be obtained from formula (2) and formula (3). The dimensionless disturbance wave velocity is the ratio of the disturbance wave velocity to the maximum gas apparent velocity, u sg,max It is the maximum gas superficial velocity within the experimental range, and its value is 40m / s.

[0088] Among the above sensors, the vortex flowmeter is a velocity sensor that measures the vortex shedding frequency f VS The volume flow rate is obtained, and the indicated result of the gas phase volume flow rate Q is obtained by formula (4): g,tp In the wet gas flow, the presence of a small amount of liquid phase makes the measured gas phase volume flow rate Q g,tp Higher than the actual gas flow rate Q g , which is called "over-reading". Therefore, the indicated result of gas phase volume flow rate Q g,tp The actual flow rate Q g The relationship between them is shown in formula (6).

[0089] In order to improve the measurement accuracy of the vortex flowmeter in wet gas and wet steam, it is necessary to establish an OR prediction model to correct the indication of the vortex flowmeter. Among them, patent CN202110129896.3 explores the influence of liquid film thickness on vortex over-reading. However, only the liquid film thickness information cannot fully characterize the flow state of the liquid film. Therefore, it is necessary to establish a complete set of liquid film flow parameter measurement systems to accurately obtain more liquid film information, and then explore the relationship between liquid film flow parameters (liquid film thickness, disturbance wave velocity) and vortex signals, so as to achieve online accurate measurement of phase flow in annular mist flow. Next, the dimensionless liquid film thickness is used as the main parameter, and the dimensionless disturbance wave velocity is used as an auxiliary parameter to establish the prediction formula for OR over-reading.

[0090] The dimensionless liquid film thickness, dimensionless disturbance wave velocity and OR are fitted by the least squares method. The prediction model between the overreading coefficient OR and the dimensionless liquid film thickness and dimensionless disturbance wave velocity is shown in formula (12):

[0091] OR=1+2.846δ *1.282 v w *-0.3207 (12)

[0092] Figure 5The prediction model between the overreading coefficient OR and the dimensionless liquid film thickness and the dimensionless disturbance wave velocity is given. It can be seen from the figure that the prediction accuracy is within ±1.1%, and the correlation coefficient R 2 It can reach 0.934, and the relative root mean square error rRMSE = 0.608%, and the prediction effect is good.

[0093] It is known that the main factors affecting the thickness of the liquid film are surface tension and inertia, and the gas-liquid phase Weber number can characterize the surface tension and inertia in the wet gas flow. The relationship between the gas phase and liquid phase Weber number and the average liquid film thickness has been established in patent CN202110129896.3. Through experimental data analysis, there is also a close correlation between the dimensionless disturbance wave velocity and the gas-liquid Weber number. Therefore, the experimental data is fitted by the least squares method, and the mathematical model between the gas-liquid Weber number and the dimensionless disturbance wave velocity is obtained as shown in formula (13):

[0094]

[0095] The fitting curve is as follows Figure 6 As shown, the correlation coefficient R 2 =0.9735, mean absolute percentage error MAPE = 2.798%, relative root mean square error rRMSE = 3.205%, 98% of the data points are within ±5% error, the above data show that the fitting effect is good.

[0096] Similarly, the vortex over-reading coefficient OR is affected by multiple factors such as liquid phase flow rate, gas phase flow rate and gas-liquid density. Following the dimensionless disturbance wave velocity modeling process, under the experimental conditions, the least squares method was used to fit the experimental data, and it was found that there was a linear relationship between the over-reading coefficient OR and the power product of the gas-liquid Weber number. The gas-liquid Weber number was used to establish the vortex over-reading coefficient OR prediction model:

[0097] OR=1+50.47We g -0.788 We l 0.2427 (14)

[0098] Figure 7 The prediction model between the overreading coefficient OR and the gas-liquid Weber number is given. It can be seen from the figure that the prediction accuracy is within ±1%, and the correlation coefficient R 2 It can reach 0.9414, and the relative root mean square error rRMSE = 0.575%. The above error data show that the prediction results of the OR mathematical model are good.

[0099] Based on the above modeling and calibration results, the wet gas phase flow measurement is carried out. The implementation process is as shown in the attached Figure 4 As shown, the method is as follows:

[0100] 1) Collect the two-phase pressure P, two-phase temperature T, the time series signal s(t) output by the vortex piezoelectric sensor, and the upstream and downstream liquid film thickness series δ1(t) and δ2(t) output by the liquid film flow parameter sensor;

[0101] 2) The gas density ρ can be calculated by the two-phase pressure P and the two-phase temperature T g and liquid density ρ l ; Perform fast Fourier transform (FFT) on the timing signal s(t) output by the vortex piezoelectric sensor to obtain the vortex shedding frequency f VS ; The upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) output by the liquid film flow parameter sensor are averaged to obtain the average liquid film thickness δ1 and δ2;

[0102] 3) Given the center distance L between the upstream and downstream sensors, the upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) are cross-correlated to obtain the transit time τ0, and the disturbance wave velocity is obtained by solving it, thus converting the liquid film characteristic parameters into dimensionless parameters;

[0103] 4) Calculate the indicated result Q of gas phase volume flow g,tp ;

[0104] 5) Calculate the vortex street over-reading coefficient OR based on the dimensionless liquid film thickness and the dimensionless disturbance wave velocity;

[0105] 6) Calculate the actual gas flow Q g ;

[0106] 7) Calculate Q g The corresponding gas superficial velocity u sg ;

[0107] 8) According to the experimental data, a mathematical model between OR and the Weber number of gas and liquid is established, and the gas superficial velocity u sg Calculate the Weber number of the gas We g , and then inversely calculate the liquid phase Weber number We l ;

[0108] 9) Finally, the liquid phase flow rate Q is obtained l .

[0109] In this example, according to the above method, the wet gas phase flow measurement is finally achieved.

[0110] The gas-liquid phase volume flow rate obtained by solving the above model is used for error analysis to evaluate the prediction effect. The percentage error (PE) of the gas phase volume flow rate is within ±1.5%, and the mean absolute percentage error (MAPE) is 0.684%. PE is the percentage error of the actual flow rate, LVF is the liquid volume fraction, and the calculation formula is LVF = Q l / (Q l +Qg ). The relative error PE (%) = (predicted value - true value) / true value × 100, and the percentage error of gas flow is as follows Figure 8 As shown. The liquid phase flow prediction, percentage error (PE) is within ±6%, the mean absolute percentage error (MAPE) is 3.06%, and the full scale percentage error (FSPE) is within ±4%. Among them, FSPE (%) = (predicted value - true value) / maximum true value × 100, the full scale percentage error of the liquid phase flow is as follows Fig. 9 .

[0111] The present invention proposes a vortex street wet gas phase flow measurement method based on liquid film flow parameter modeling by modeling disturbance wave velocity, liquid film thickness and vortex street over-reading coefficient OR. The method is simple to operate, low in cost, and can achieve high-precision online measurement.

Claims

1. A vortex street wet gas phase flow measurement method based on liquid film flow parameter modeling, using the signals collected by the vortex street piezoelectric sensor and the liquid film flow parameter sensor, combined with the dimensionless liquid film thickness and disturbance wave velocity to establish an over-reading prediction model, predicting the phase flow according to the over-reading coefficient OR, and realizing wet gas phase flow measurement, including the following steps: 1) Collect the two-phase pressure P, two-phase temperature T, the time series signal s(t) output by the vortex piezoelectric sensor, and the upstream and downstream liquid film thickness series δ1(t) and δ2(t) output by the liquid film flow parameter sensor; 2) Calculate the gas density ρ by the two-phase pressure P and the two-phase temperature T g and liquid density ρ l ; Perform fast Fourier transform on the timing signal s(t) output by the vortex piezoelectric sensor to obtain the vortex shedding frequency f VS ; The upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) output by the liquid film flow parameter sensor are averaged to obtain the average liquid film thickness δ1 and δ2; 3) Given the center distance L of the upstream and downstream liquid film flow parameter sensors, the upstream and downstream liquid film thickness sequence signals δ1(t) and δ2(t) are cross-correlated to obtain the transit time τ0, and the disturbance wave velocity is obtained by solving formula (1): 4) According to formula (2) and formula (3), the characteristic parameters of the liquid film are dimensionless to obtain the dimensionless liquid film thickness and dimensionless disturbance wave velocity: v w * =v w / u sg,max (3) in, D is the pipe diameter; δ1 is the average liquid film thickness measured by the upstream conductivity ring sensor; δ * is the dimensionless liquid film thickness, v w * is the dimensionless disturbance wave velocity, v w is the real disturbance wave velocity, u sg,max is the maximum gas superficial velocity within the experimental range; 5) Calculate the gas phase volume flow rate indication result Q according to formula (4): g,tp : Among them, f VS is the vortex shedding frequency, K v is the instrument factor of the vortex flowmeter in single-phase gas; 6) Calculate the vortex street over-reading coefficient OR according to formula (5): Wherein, k1 is a constant coefficient, n1 and n2 are constant power exponents, and their specific values ​​are obtained by fitting the function form described in formula (5); 7) Calculate the actual gas flow rate Q by formula (6): g 8) Calculate Q according to formula (7) g The corresponding gas superficial velocity u sg 9) Based on the experimental data, a mathematical model between OR and the Weber number of gas and liquid is established. The parameterized expression of the model is: Where k2 is a constant coefficient, n3 and n4 are constant power exponents; the Weber number of a gas is in the form of The Weber number of a liquid is defined as Δρ is the density difference between gas and liquid; u sg Indicates the superficial velocity of gas; u sl is the apparent velocity of the liquid; D is the nominal diameter of the pipeline, σ is the surface tension; 10) From the gas superficial velocity u sg Calculate the Weber number of the gas We g , and then use formula (8) to infer the liquid phase Weber number We l 11) Liquid flow rate Q l Calculated by formula (10):

Images