Gas-liquid two-phase flow measuring method based on bent pipe and acoustic emission technology

Abstract

The invention discloses a gas-liquid two-phase flow measurement method based on a bent pipe and an acoustic emission technology, and belongs to the field of gas-liquid two-phase flow measurement. The method comprises the steps that the average liquid holdup HL of two-phase flow in a pipeline is measured, and the instantaneous speed Vc of a gas phase and a liquid phase passing through the 45-degree section of the bent pipe is calculated; identifying a gas-liquid two-phase flow pattern in the pipeline; taking the Vc, the HL, the apparent liquid velocity VSL and the apparent gas velocity VSG as variables, and adopting a dimensional analysis method to propose a correlation between the Vc and the VSL and between the HL and the VSL and between the VSL and the VSG according to different flow patterns; calculating an effective voltage value RMS of the acoustic emission signal, and establishing a correlation between the RMS and the VSL and VSG; fitting parameters in the correlation formula; newly collected Vc, HL and RMS are input into the fitted correlation formula, VSL and VSG are output, and therefore measurement of the gas-liquid two-phase flow is completed. Compared with a single-phase metering method after separation commonly used in the oil and gas industry, the device has the advantages of being simple in structure, low in measurement cost, capable of achieving real-time measurement, non-invasive and the like.

Description

Technical Field

[0001] This invention relates to a gas-liquid two-phase flow measurement method that combines a bent tube flowmeter with acoustic emission technology—a gas-liquid two-phase flow measurement method based on bent tube and acoustic emission technology, belonging to the field of gas-liquid two-phase flow measurement. Background Technology

[0002] In petroleum engineering, produced fluids at the wellhead are mostly gas-liquid two-phase flows, and wellhead metering is of great significance for production guidance. Currently, with the further development of offshore oil and gas fields and the limited space on offshore oil and gas platforms, accurate, real-time, space-saving, and low-cost measurement of gas-liquid two-phase flows is of great importance in engineering and production.

[0003] Flow patterns, as an important tool for judging and classifying the structure of gas-liquid two-phase flows, have been applied in the energy industry. Methods for measuring the flow rate of gas-liquid two-phase flows include separate measurement methods and non-separate measurement methods. Separate measurement methods have the advantage of accuracy, but also have disadvantages such as large footprint and inability to measure in real time. Non-separate measurement methods mainly include electrical methods (including conductivity and capacitance methods), optical methods, and X-ray methods. While each of these methods has its own advantages, it also has its limitations. For example, electrical methods are susceptible to interference, optical methods have strict requirements on the measurement medium, and X-ray methods are expensive and pose a radioactive hazard.

[0004] The elbow flow meter, as a differential pressure flow meter, has been initially applied in the flow measurement of single-phase flow and has achieved good results. Acoustic emission, as a passive acoustic measurement method, has the advantages of low energy consumption, no pollution, and simple operation, and can obtain rich flow information in fluid measurement. Summary of the Invention

[0005] In view of the problems existing in the current gas-liquid two-phase flow measurement, and based on the existing research in our laboratory, this invention provides a gas-liquid two-phase flow measurement method based on bend pipe and acoustic emission technology.

[0006] A method for measuring the flow rate of a gas-liquid two-phase flow based on a bend in the pipe and acoustic emission technology, characterized by the following steps: Step 0: Install a differential pressure transmitter and an acoustic emission sensor at the 90° bend of the horizontally positioned gas-liquid two-phase flow pipe, and collect the differential pressure signal and acoustic emission signal. The cross-section of the bend can be circular or elliptical. Install two sets of dual parallel conductivity probes at a distance of 15cm before and after the two-phase flow straight pipe to measure the average liquid holdup. Step 1: Measure the average liquid holdup of the gas-liquid two-phase flow in the pipeline. H L Calculate the mixing density of gas-liquid two-phase flow ρ m ; Step 2: Measure the pressure difference between the inside and outside of the bend, the radius of curvature of the bend, and the straight-line distance between the pressure taps on the inside and outside of the bend, and calculate the parameters. V C : In the formula, V C Δ is the instantaneous velocity of the gas and liquid phases at the 45° section of the bend in the pipe, in m / s; P The pressure difference between the inside and outside of the bend, in Pa; R Let be the radius of curvature at the axis of the bend, in meters (m). l e The straight-line distance between the pressure tapping points on the inside and outside of the bend, in meters; Step 3: Collect acoustic emission signals of the gas-liquid two-phase flow in the bend and identify the flow pattern of the gas-liquid two-phase flow in the pipe; Step 4: Set parameters V C Average liquid holdup H L Apparent fluid velocity V SL and apparent gas velocity V SG As variables, dimensional analysis is used to propose parameters based on different flow patterns. V C Average liquid holdup H L With apparent fluid velocity V SL Apparent gas velocity V SG The correlations, where the measurement correlations for bubbly flow and annular flow are shown in the following equations: The correlation for intermittent flow measurement is shown in the following equation:

[0007] After performing a 50kHz low-pass filter on the acquired acoustic emission signal, the effective voltage value of the acoustic emission signal is calculated. RMS ,Establish RMS With apparent fluid velocity V SL and apparent gas velocity V SG Relationship: In the formula, k , b , x , y , z It is a dimensionless number; GFor correction terms, m / s; B For intermittent flow correction, m / s; a , C , m , n These are the acoustic emission correlation coefficients; By inputting multiple sets of the following variables: instantaneous velocity V C Average liquid holdup H L Apparent fluid velocity V SL Apparent gas velocity V SG The root mean square (RMS) of the acoustic emission signal is used to fit the three formulas, thereby obtaining the parameters. k , b , G , x , y , z , B , a , C , m , n The value; Step 5: After obtaining the parameters through fitting in Step 4, input the instantaneous velocity. V C Average liquid holdup H L The root mean square (RMS) of the acoustic emission signal is used to output the apparent liquid velocity. V SL Apparent gas velocity V SG This allows for the measurement of the flow rate of the gas-liquid two-phase flow.

[0008] In step 0, the differential pressure transmitter taps pressure on both the inside and outside sides of the 45° section of the bend.

[0009] In step 0, the acoustic emission sensor is installed on the upper and lower sides of the 45° section of the bend.

[0010] In step 1, the process of calculating the gas-liquid two-phase flow mixing density is as follows: the average liquid holdup of the gas-liquid two-phase flow in the pipeline is measured, and the gas-liquid two-phase flow mixing density is calculated using the following formula: In the formula, ρ m -Gas-liquid two-phase flow mixing density, kg / m³ 3 ; H L -Average liquid holdup during the sampling period, dimensionless; ρ L -Liquid phase density in two-phase flow, kg / m3; ρ G - Gas phase density in two-phase flow, kg / m3.

[0011] Step 3 involves using existing technology to collect acoustic emission signals of the gas-liquid two-phase flow in the bend and to identify the flow pattern of the gas-liquid two-phase flow in the pipe. The existing technology includes the gas-liquid two-phase flow pattern identification method based on acoustic emission-BP neural network described in our laboratory's early patent CN111896616B.

[0012] In step 3, the gas-liquid two-phase flow patterns include bubbly flow, annular flow, intermittent flow, and slug flow, among which slug flow can be classified as intermittent flow.

[0013] In step 4, the acoustic emission signal of the gas-liquid two-phase flow in the bend is acquired by the acoustic emission signal acquisition system, and after being low-pass filtered at 50kHz, the effective voltage value is calculated according to the following formula. RMS : Where N is the number of acoustic emission signals collected, and V i This represents the i-th acoustic emission signal collected.

[0014] In step 4, the parameters in the formula flow calculation model are used. k , b , G , x , y , z , B , a , C , m , n The values ​​are obtained through experimental measurements and calculations. Specifically, in the experiment, a bent pipe with given dimensional parameters is installed to obtain R and l. e Controlling the gas-liquid two-phase flow pattern to the desired flow pattern involves changing the known apparent velocity of the gas phase. V SG and apparent velocity of liquid phase V SL The differential pressure, acoustic emission signal, and average liquid holdup at a 45° cross-section of the bend were obtained using a differential pressure transmitter, acoustic emission sensor, and dual parallel conductivity probes. The results were then calculated. V C and RMS Multiple data points are collected, and the values ​​of the parameters in the formula are obtained through fitting. If the size parameters of the measuring bend are changed, the above experiment is repeated, and the coefficients of the correlation formula in the flow calculation model corresponding to this type of bend are obtained based on the measurement data. For bend flow measurement in marine oil and gas industrial sites, large-scale oil and gas multiphase flow metering equipment, both domestic and international, can be used for experimental calibration.

[0015] In step 4, the apparent fluid velocity is calculated.V SL Apparent gas velocity V SG When the problem is solved, the iterative solution method should be used first. When the iterative solution diverges, the distance solution method should be used.

[0016] Traditional multiphase flow measurement methods involve using separators for separation and measurement. However, separators are large and bulky, occupying significant space on space-constrained offshore platforms. Furthermore, separators are expensive and difficult to measure in real-time. Our laboratory's patent CN115683250A presents a method for measuring the mixed flow rate of gas-liquid two-phase flow using a bent tube and acoustic emission technology combined with an ELMAN neural network, demonstrating the feasibility of the bent tube-acoustic emission technology in multiphase flow measurement.

[0017] When a fluid passes through a bend, centrifugal force creates a radial pressure difference between the inner and outer walls of the bend. The magnitude of this pressure difference is related to the fluid's properties and the gas-liquid phase motion state. Based on this, this invention further investigates the relationship between the pressure difference between the inner and outer sides of the bend, liquid holdup, acoustic emission signal, and apparent flow velocity, proposing an online two-phase flow measurement method that measures the gas and liquid flow rates separately without separating the gas and liquid phases. Compared to the commonly used single-phase metering method after separation in the oil and gas industry, this invention has advantages such as simple structure, low measurement cost, real-time measurement capability, and non-invasiveness. By installing an acoustic emission sensor on the bend flowmeter, the differential pressure between the inner and outer sides of the bend, the liquid holdup at the cross-section, and the acoustic emission signal generated when the gas-liquid two-phase flow passes through the bend are measured. Combined with the flow measurement model proposed in this invention, the gas and liquid phase flow rates are measured. Attached Figure Description

[0018] Figure 1 A schematic diagram of a curved pipe acoustic emission measurement system, wherein: 1-differential pressure transmitter; 2-acoustic emission sensor; 3-preamplifier; 4-acoustic emission signal acquisition box; 5-computer; 6-two sets of dual parallel conductivity probes installed with a front-to-back spacing of 15cm; 7-horizontally installed pipeline.

[0019] Figure 2 Schematic diagram of the installation positions of differential pressure and acoustic emission measuring elements, where 1-differential pressure transmitter; 2-acoustic emission sensor; 8-horizontally installed 90° bend.

[0020] Figure 3 The program flow chart of the iterative solution method for the flow rate measurement model of gas-liquid two-phase flow with acoustic emission through a bent pipe.

[0021] Figure 4 Flowchart of the distance calculation method for the flow rate measurement model of gas-liquid two-phase flow in a curved tube acoustic emission system. Detailed Implementation

[0022] like Figure 1As shown, the hardware facilities used in this invention include 1-differential pressure transmitter, 2-acoustic emission sensor, 3-preamplifier, 4-acoustic emission signal acquisition box, 5-computer, and 6-two sets of dual parallel conductivity probes installed with a front-to-back spacing of 15cm.

[0023] like Figure 2 As shown, differential pressure transmitter 1 is installed on both the inner and outer sides of the 45° section of a horizontally installed 90° bend, with the high-pressure side on the outer side and the low-pressure side on the inner side. Acoustic emission sensor 2 is installed at the top and bottom of the 45° section of the horizontally installed 90° bend. The bend and experimental piping are made of stainless steel. The acoustic emission signal, differential pressure signal, and conductivity probe signal are acquired by a data acquisition card and input into a computer. The flow rate is calculated by a program written inside the computer. The specific method is as follows: Step 0: Install a differential pressure transmitter and an acoustic emission sensor at the 90° bend of the horizontally set gas-liquid two-phase flow pipe, and collect the differential pressure signal and acoustic emission signal. The cross-section of the bend can be circular or elliptical. Install two sets of dual parallel conductivity probes at a distance of 15cm before and after the two-phase flow straight pipe to measure the average liquid holdup.

[0024] Step 1: Measure the average liquid holdup of the gas-liquid two-phase flow in the pipeline and calculate the gas-liquid two-phase flow mixing density using equation (1).

[0025] (1) In the formula, ρ m -Gas-liquid two-phase flow mixing density, kg / m³ 3 ; H L -Average liquid holdup during the sampling period, dimensionless; ρ L -Liquid phase density in two-phase flow, kg / m3; ρ G - Gas phase density in two-phase flow, kg / m3.

[0026] Step 2: Measure the pressure difference between the inside and outside of the bend, the radius of curvature of the bend, and the straight-line distance between the pressure tapping points on the inside and outside of the bend. Calculate the parameters using equation (2). V C :

[0027] (2) In the formula, V C Δ is the instantaneous velocity of the gas and liquid phases at the 45° section of the bend in the pipe, in m / s; P The pressure difference between the inside and outside of the bend, in Pa; R Let be the radius of curvature at the axis of the bend, in meters (m). l e The straight-line distance between the pressure tapping points on the inside and outside of the bend is in meters (m).

[0028] Step 3: Collect acoustic emission signals of the gas-liquid two-phase flow in the bend and identify the flow pattern of the gas-liquid two-phase flow in the pipeline.

[0029] Step 4: Set parameters V C Average liquid holdup H L Apparent fluid velocity V SL and apparent gas velocity V SG As variables, dimensional analysis is used to propose parameters based on different flow patterns. V C Average liquid holdup H L With apparent fluid velocity V SL Apparent gas velocity V SG The correlations are as follows: the measurement correlations for bubbly flow and annular flow are shown in equation (3), and the measurement correlations for intermittent flow are shown in equation (4).

[0030] (3) (4) After performing a 50kHz low-pass filter on the acquired acoustic emission signal, the effective voltage value of the acoustic emission signal is calculated. RMS ,Establish RMS With apparent fluid velocity V SL and apparent gas velocity V SG The correlation is shown in equation (5): (5) In equations (3) to (5), k , b , x , y , z It is a dimensionless number, and its value varies depending on the pipe structure, the flow medium, and the flow pattern when fitting the solution. G The correction term, in m / s, is subject to zero drift of the instrument during the measurement process; B This is an intermittent flow correction term, in m / s, which is subject to the zero drift of the instrument during the measurement process and the fluctuation of the average liquid hold-up rate during the sampling time. a , C , m , n The coefficients for acoustic emission correlation vary depending on the fluid medium, sampling time, pipe parameters, and acoustic emission sensor parameters during fitting and solving.

[0031] By inputting multiple sets of the following variables: instantaneous velocity V C Average liquid holdup H L Apparent fluid velocity V SL Apparent gas velocity V SG The root mean square (RMS) of the acoustic emission signal is used to fit equations (3) to (5) to obtain the parameters. k , b , G , x , y , z , B , a , C , m , n The value of .

[0032] When the flow inside the pipe is a bubble flow or annular flow, the program calls equations (3) and (5) to form a system of equations for simultaneous solution; when the flow inside the pipe is an intermittent flow, the program calls equations (4) and (5) to form a system of equations for simultaneous solution.

[0033] Step 5: After obtaining the parameters through fitting in Step 4, input the instantaneous velocity. V C Average liquid holdup H L The root mean square (RMS) of the acoustic emission signal is used to output the apparent liquid velocity. V SL Apparent gas velocity V SG This allows for the measurement of the flow rate of the gas-liquid two-phase flow.

[0034] In the above process, multiple sets of experiments are needed to fit and calibrate the model parameters to determine the parameters. k , b , G , x , y , z , B , a , C , m , n The value; combining equations (3) and (5) yields the flow measurement models for bubbly and annular flow, and combining equations (4) and (5) yields the flow measurement model for intermittent flow. The solution method for the measurement model is set, with iterative solution preferred, such as... Figure 3 As shown. When the iterative solution diverges, a distance-based solution method is used, such as... Figure 4 As shown.

[0035] Pressure difference Δ between the inside and outside of the input bend P Average liquid holdup H L Root mean square (RMS) of acoustic emission signal, correlation parameters k , b , G , x , y , z , B , a , C , m , n The values ​​are entered into the computer, and the flow patterns are divided using the acoustic emission-BP neural network method. Then, the apparent liquid velocity is calculated using formulas (3) to (5). V SL Apparent gas velocity V SG Solve the equation and output the apparent gas velocity and apparent liquid velocity.

[0036] Example This embodiment uses two types of stainless steel bends, and the specific structural parameters of the two bends are shown in Table 1.

[0037] After determining the structural parameters of the bend, first calibrate the parameters in equations (3) to (5). k , b , G , x , y , z , B , a , C , m , n The value of . The specific method is as follows: 1) Using the acoustic emission-BP neural network flow pattern recognition system (CN111896616B) of this laboratory, experiments were conducted on various flow patterns such as bubbly flow, annular flow, and intermittent flow. 2) Measure the average liquid holdup in the pipeline using dual parallel conductivity probes. H L The pressure difference Δ between the inside and outside of the bend is measured using a differential pressure sensor. P The acoustic emission sensor is used to measure the acoustic signal of fluid flowing through a bend in the pipe. 3) Calculate the mixing density using equation (1) ρ m The cross-sectional velocity is calculated using equation (2). V C The acoustic emission signal is low-pass filtered at 50kHz, and the effective value is calculated according to equation (6). RMS: (6) 4) Based on the experimental data, the bivariate fitting method was used to adjust the parameters in equations (3) to (5). k , b , G , x , y , z , B , a , C , m , n The value is then calculated. 5) Measure the parameters k , b , G , x , y , z , B , a , C , m , n Substituting the values ​​into equations (3) to (5), numerical methods are used to determine the apparent gas velocity. V SG and apparent fluid velocity V SL The solution is performed. When the flow inside the pipe is a bubbly or annular flow, the program calls equations (3) and (5) to form a system of equations for simultaneous solution; when the flow inside the pipe is an intermittent flow, the program calls equations (4) and (5) to form a system of equations for simultaneous solution. Experimental calculations show that the relative errors of the prediction results of the two algorithms are basically the same, while the Newton iteration method is relatively simple to calculate. Therefore, the Newton iteration method is preferred for solution. When the iteration diverges, the distance solution method is used.

[0038] The specific structural parameters of the two types of bends used in the embodiments are shown in Table 1.

[0039] Table 1 Structural parameters of two types of bends pipe bend code radius of curvature R / mm 45° cross-sectional shape Pipe inner diameter d / mm Pressure tapping distance l e / mm]] R39 39 round 26 29 R64 64 oval 26 25.4 Example 4) Parameters measured for two types of bends k , b , G , x , y , z , B , a , C , m , n The values ​​are shown in Tables 2-4.

[0040] Table 2 Parameter values ​​of the correlation equation for bubbly flow measurement R39 0.7972 0.2739 0.1315 7.05 16.89 1.16 1.68 R64 0.9775 0.3603 0.196 2.55 2.82 1.15 1.56 Table 3 Parameter values ​​of the measurement correlation in annular flow R39 13.26 4.93 x 10 -3 ]]> -0.1542 0.17 -37.5 2.59 0.57 R64 11.49 9.06 x 10 -3 ]]> 0.2361 0.91 -38.2 1.70 0.32 Table 4 Parameter values ​​of the correlation equation for intermittent flow time measurement R39 0.9461 -0.1278 0.5 0.4 19.3 -1.20 0.90 0.65 R64 0.6797 -0.1639 0.6 0.2 5.91 -0.83 0.94 0.65 The measurement results are shown in Tables 5-10. Due to space limitations, only some of the measurement results are listed in this example. Table 11 shows the average absolute percentage measurement error of the measurement model in the two bends and three flow patterns, which shows that all have good measurement effects. The measurement effect of the R64 bend is slightly worse than that of the R39 bend. The reasons may include: first, the radius of curvature of the bend is larger, which reduces the centrifugal force experienced by the fluid when passing through the bend; second, the 45° cross-sectional shape is elliptical due to the pipe manufacturing process, which weakens the secondary flow in the bend. The measurement effect of intermittent flow is slightly worse because intermittent flow has strong randomness, resulting in strong fluctuations in liquid holdup. Therefore, the measurement effect depends on the sampling time. This experiment is limited by the hardware performance of the acoustic emission system, and the 15-second sampling time used is relatively short. When the liquid plug frequency is low, the data lacks representativeness. It is expected that extending the sampling time can improve the measurement accuracy. In practical industrial applications, by employing high-precision pipe bending technology and high-performance acoustic emission acquisition instruments, and extending the sampling time, the technology of this invention can achieve high measurement accuracy, meeting the needs of compact space and low cost oil and gas flow measurement in offshore oil and gas fields.

[0041] Table 5. Measurement results of bubbly flow in R39 bend pipe Serial Number V SG-real , m / s ​ V SG-cal , m / s ​ δV SG , %]]> ​ V SL-real , m / s ​ V SL-cal , m / s ​ δV SL , %]]> ​ 1 0.32 0.38 15.51 1.46 1.45 -0.69 2 0.32 0.33 3.03 2.09 2.11 0.94 3 0.54 0.46 -17.75 1.27 1.25 -1.60 4 0.54 0.72 25.97 1.67 1.56 -7.05 5 0.54 0.58 6.89 2.09 2.05 -1.95 6 0.69 0.95 27.95 1.46 1.32 -10.61 7 0.71 0.6 -18.33 2.09 2.05 -1.95 8 1.03 1.2 14.17 1.69 1.63 -3.68 9 1.039 1.046 0.67 1.878 1.89 0.63 10 2.033 1.896 -7.22 1.657 1.70 2.53 Note: In the above table, V SG-real - Apparent gas velocity as determined experimentally; V SG-cal – Apparent gas velocity calculated by the model; V SL-real - Apparent liquid velocity determined experimentally; V SL-cal – Apparent fluid velocity calculated by the model; δV SG - Relative error in gas phase flow measurement; δ V SL - Relative error in liquid phase flow measurement. (The same applies below.)

[0042] Table 6. Measurement results of bubbly flow in R64 bend pipe Serial Number V SG-real , m / s ​ V SG-cal , m / s ​ δV SG , %]]> ​ V SL-real , m / s ​ V SL-cal , m / s ​ δV SL , %]]> ​ 1 0.28 0.23 -21.74 1.24 1.28 3.13 2 0.28 0.33 15.15 1.46 1.42 -2.82 3 0.53 0.64 17.19 1.27 1.24 -2.42 4 0.53 0.54 1.85 1.47 1.46 -0.68 5 0.72 0.73 1.37 1.47 1.42 -3.52 6 0.71 0.79 10.13 1.67 1.64 -1.83 7 1.02 1.03 0.97 1.66 1.69 1.78 8 1.02 1.14 10.53 1.88 1.86 -1.08 9 2.04 1.82 -12.09 1.67 1.82 8.24 10 2.02 2.04 0.98 1.89 1.83 -3.28 Table 7 Measurement results of annular flow in R39 bend pipe Serial Number V SG-real , m / s ​ V SG-cal , m / s ​ δV SG , %]]> ​ V SL-real , m / s ​ V SL-cal , m / s ​ δV SL ,% ​ 1 17.23 16.75 -2.87 0.10 0.12 16.67 2 17.07 17.39 1.84 0.21 0.19 -10.53 3 21.18 20.35 -4.08 0.22 0.24 8.33 4 21.11 20.42 -3.38 0.33 0.35 5.71 5 21.29 20.96 -1.57 0.41 0.43 4.65 6 21.24 21.40 0.75 0.52 0.50 -4.00 7 25.56 23.87 -7.08 0.10 0.13 23.08 8 25.41 25.06 -1.40 0.21 0.21 -5.00 9 25.64 25.57 -0.27 0.32 0.33 3.03 10 25.44 25.81 1.43 0.42 0.41 -2.44 Table 8 Measurement results of annular flow in R64 bend pipe Serial Number V SG-real , m / s ​ V SG-cal , m / s ​ δV SG , %]]> ​ <![CDATA[ V SL-real ,m / s]]> <![CDATA[ V SL-cal ,m / s]]> <![CDATA[ δV SL ,%]]> 1 16.84 17.15 1.81 0.32 0.3 -6.67 2 16.99 17.17 1.05 0.42 0.44 4.55 3 21.26 21.78 2.39 0.22 0.19 -15.79 4 21.33 21.74 1.89 0.32 0.31 -3.23 5 21.35 21.48 0.61 0.44 0.45 2.22 6 21.31 21.03 -1.33 0.52 0.55 5.45 7 25.42 24.71 -2.87 0.11 0.13 15.38 8 24.53 26.81 8.50 0.21 0.18 -16.67 9 25.43 27.22 6.58 0.32 0.27 -18.52 10 25.35 25.94 2.27 0.43 0.42 -2.38 Table 9. Intermittent Flow Measurement Results of R39 Bend Serial Number <![CDATA[ V SG-real ,m / s]]> <![CDATA[ V SG-cal ,m / s]]> <![CDATA[ δV SG ,%]]> <![CDATA[ V SL-real ,m / s]]> <![CDATA[ V SL-cal ,m / s]]> <![CDATA[ δV SL ,%]]> 1 1.05 1.03 -1.94 0.31 0.29 -6.90 2 1.02 0.91 -12.09 0.43 0.57 24.56 3 2.01 2.42 16.94 0.21 0.17 -23.53 4 2.02 1.76 -14.77 0.31 0.44 29.55 5 3.05 4.21 27.55 0.22 0.18 -22.22 6 3.05 4.01 23.94 0.31 0.25 -24.00 7 5.01 4.21 -19.00 0.43 0.58 25.86 8 4.99 6.23 19.90 0.52 0.43 -20.93 9 6.98 7.71 9.47 1.06 0.89 -19.10 10 6.94 8.8 21.14 1.21 0.94 -28.72 Table 10 Results of Intermittent Flow Measurements on R64 Bends Serial Number <![CDATA[ V SG-real ,m / s]]> <![CDATA[ V SG-cal ,m / s]]> <![CDATA[ δV SG ,%]]> <![CDATA[ V SL-real ,m / s]]> <![CDATA[ V SL-cal ,m / s]]> <![CDATA[ δV SL ,%]]> 1 1.00 0.86 -16.28 0.21 0.29 27.59 2 1.00 0.83 -20.48 0.33 0.38 13.16 3 2.03 2.87 29.27 0.41 0.34 -20.59 4 2.00 1.72 -16.28 0.54 0.77 29.87 5 3.12 3.84 18.75 0.53 0.42 -26.19 6 3.15 3.81 17.32 0.63 0.55 -14.55 7 5.11 4.66 -9.66 0.63 0.71 11.27 8 5.08 4.52 -12.39 0.84 0.76 -10.53 9 7.14 8.03 11.08 0.84 0.71 -18.31 10 7.11 7.24 1.80 1.05 1.1 4.55 Table 11 Mean Absolute Percentage Error of the Bend Pipe Acoustic Emission Gas-Liquid Two-Phase Flow Measurement Model

Claims

1. A method for measuring the flow rate of a gas-liquid two-phase flow based on a bent pipe and acoustic emission technology, characterized in that: Includes the following steps: Step 0: Install a differential pressure transmitter and an acoustic emission sensor at the 90° bend of the horizontally set gas-liquid two-phase flow pipe, and collect the differential pressure signal and acoustic emission signal. Install two sets of double parallel conductivity probes at a distance of 15cm before and after the two-phase flow straight pipe to measure the average liquid holdup. Step 1: Measure the average liquid holdup of the gas-liquid two-phase flow in the pipeline. H L Calculate the mixing density of gas-liquid two-phase flow ρ m ; Step 2: Measure the pressure difference between the inside and outside of the bend, the radius of curvature of the bend, and the straight-line distance between the pressure taps on the inside and outside of the bend, and calculate the parameters. V C : In the formula, V C Δ is the instantaneous velocity of the gas and liquid phases at the 45° section of the bend in the pipe, in m / s; P The pressure difference between the inside and outside of the bend, in Pa; R Let be the radius of curvature at the axis of the bend, in meters (m). l e The straight-line distance between the pressure tapping points on the inside and outside of the bend, in meters; Step 3: Collect acoustic emission signals of the gas-liquid two-phase flow in the bend and identify the flow pattern of the gas-liquid two-phase flow in the pipe; Step 4: Set parameters V C Average liquid holdup H L Apparent fluid velocity V SL and apparent gas velocity V SG As variables, dimensional analysis is used to propose parameters based on different flow patterns. V C Average liquid holdup H L With apparent fluid velocity V SL Apparent gas velocity V SG The correlations, where the measurement correlations for bubbly flow and annular flow are shown in the following equations: The correlation for intermittent flow measurement is shown in the following equation: After performing a 50kHz low-pass filter on the acquired acoustic emission signal, the effective voltage value of the acoustic emission signal is calculated. RMS ,Establish RMS With apparent fluid velocity V SL and apparent gas velocity V SG Relationship: In the formula, k , b , x , y , z It is a dimensionless number; G For correction terms, m / s; B For intermittent flow correction, m / s; a , C , m , n These are the acoustic emission correlation coefficients; By inputting multiple sets of the following variables: instantaneous velocity V C Average liquid holdup H L Apparent fluid velocity V SL Apparent gas velocity V SG The root mean square (RMS) of the acoustic emission signal is used to fit the three formulas to obtain the parameters. k , b , G , x , y , z , B , a , C , m , n The value; Step 5: After obtaining the parameters through fitting in Step 4, input the instantaneous velocity. V C Average liquid holdup H L The root mean square (RMS) of the acoustic emission signal is used to output the apparent liquid velocity. V SL Apparent gas velocity V SG This allows for the measurement of the flow rate of the gas-liquid two-phase flow.

2. The method as described in claim 1, characterized in that: In step 0, the differential pressure transmitter taps pressure on both the inside and outside sides of the 45° section of the bend.

3. The method as described in claim 1, characterized in that: In step 0, the acoustic emission sensor is installed on the upper and lower sides of the 45° section of the bend.

4. The method as described in claim 1, characterized in that: In step 0, the cross-section at the bend includes a circular cross-section or an elliptical cross-section.

5. The method as described in claim 1, characterized in that: In step 1, the process of calculating the gas-liquid two-phase flow mixing density is as follows: the average liquid holdup of the gas-liquid two-phase flow in the pipeline is measured, and the gas-liquid two-phase flow mixing density is calculated using the following formula: In the formula, ρ m -Gas-liquid two-phase flow mixing density, kg / m³ 3 ; H L -Average liquid holdup during the sampling period, dimensionless; ρ L -Liquid phase density in two-phase flow, kg / m3; ρ G - Gas phase density in two-phase flow, kg / m3.

6. The method as described in claim 1, characterized in that: Step 3 involves using existing technology to collect acoustic emission signals of the gas-liquid two-phase flow in the bend and to identify the flow pattern of the gas-liquid two-phase flow in the pipeline.

7. The method of claim 6, characterized in that: In step 3, the gas-liquid two-phase flow patterns include bubbly flow, annular flow, intermittent flow, and slug flow, among which slug flow can be classified as intermittent flow.

8. The method as described in claim 1, characterized in that: In step 4, the effective voltage value is calculated. RMS The method is as follows: Acoustic emission signals of the gas-liquid two-phase flow in the bend are acquired using an acoustic emission signal acquisition system. After 50kHz low-pass filtering, the effective voltage value is calculated according to the following formula. RMS : , Where N is the number of acoustic emission signals collected, and V i This represents the i-th acoustic emission signal collected.

9. The method as described in claim 1, characterized in that: In step 4, the parameters in the flow calculation model composed of the formula are... k , b , G , x , y , z , B , a , C , m , n The values ​​are obtained through experimental measurements and calculations. Specifically, in the experiment, a bent pipe with given dimensional parameters is installed to obtain R and l. e Controlling the gas-liquid two-phase flow pattern to the desired flow pattern involves changing the known apparent velocity of the gas phase. V SG and apparent velocity of liquid phase V SL The differential pressure, acoustic emission signal, and average liquid holdup at a 45° cross-section of the bend were obtained using a differential pressure transmitter, acoustic emission sensor, and dual parallel conductivity probes. The results were then calculated. V C and RMS Multiple data points were collected, and the values ​​of the parameters in the formula were obtained by fitting the data.

10. The method of claim 9, characterized in that: In step 4, the apparent fluid velocity is calculated. V SL Apparent gas velocity V SG When the problem is solved, the iterative solution method should be used first. When the iterative solution diverges, the distance solution method should be used.

Images