Substance measurement device and substance measurement method

Abstract

The embodiment of the invention provides a substance measuring device and a substance measuring method, the substance measuring device comprises a pressure difference forming pipe, and the apertures of at least two positions in the pressure difference forming pipe are different; the pressure sensor is arranged at the upstream of the pressure difference forming pipe and is used for detecting pressure data in the pressure difference forming pipe; the temperature sensor is arranged on the pressure difference forming pipe and is used for detecting temperature data of a multiphase object to be detected; the microwave resonance sensor is connected with the pressure difference forming pipe, the microwave resonance sensor comprises a channel, and the channel is communicated with the pressure difference forming pipe and used for collecting resonance data caused by the multiphase matter to be measured; and the pressure difference sensor is arranged on the pressure difference forming pipe and is used for detecting the pressure difference of at least two positions of the pressure difference forming pipe.

Description

Technical Field

[0001] The present application relates to the field of measurement technology, and in particular to a material measuring device and a material measuring method. Background Art

[0002] In oil fields, oil and gas fields, gas fields and other fields, it is difficult to measure the content or flow of each phase in a mixed object, especially in some scenarios with relatively low water content, which is difficult to measure accurately. Summary of the Invention

[0003] The purpose of the present application is to provide a material measuring device and a material measuring method, which can improve the accuracy of measuring the content of a multiphase substance to be measured.

[0004] In a first aspect, a material measuring device comprises: a pressure differential forming tube, wherein the apertures of at least two positions in the pressure differential forming tube are different; a pressure sensor installed upstream of the pressure differential forming tube, for detecting pressure data in the pressure differential forming tube; a temperature sensor installed in the pressure differential forming tube, for detecting temperature data of a multiphase substance to be measured; a microwave resonant sensor connected to the pressure differential forming tube, the microwave resonant sensor comprising a channel, the channel being connected to the pressure differential forming tube, for collecting resonance data caused by the multiphase substance to be measured; and a pressure differential sensor installed in the pressure differential forming tube, for detecting the pressure difference at at least two positions in the pressure differential forming tube.

[0005] In the above implementation, the differential pressure data can be obtained by using the differential pressure forming tube in combination with the differential pressure sensor, and the resonance data of the microwave resonance sensor can be used to calculate the content and flow rate of each phase in the multiphase material to be measured.

[0006] In an optional embodiment, the pressure differential sensor also includes: a first pressure differential sensor and a second pressure differential sensor; the first pressure differential sensor is used to detect the pressure difference between two positions with different apertures of the pressure differential forming tube; the second differential sensor is used to detect the pressure difference between the inlet section and the outlet section of the material measuring device.

[0007] In an optional embodiment, the method further includes: a velocity sensor installed in the pressure difference forming tube, for checking velocity-related data of the multiphase substance to be measured in the pressure difference forming tube.

[0008] In the above implementation, a speed sensor may be further provided. The speed-related data measured by the speed sensor may provide a more comprehensive understanding of the multiphase material to be measured, and more algorithm logic may be combined to understand the situation of the multiphase material to be measured.

[0009] In an optional embodiment, it further includes: a flow pattern adjustment device for adjusting the flow pattern of the multiphase material to be measured entering the material measuring device.

[0010] In an optional embodiment, the flow pattern adjustment device includes one or more of a guide vane, a rotator, and a rectifier.

[0011] In an optional embodiment, it further includes: a wire mesh sensor for collecting image data of the multiphase substance to be measured flowing in the pressure difference forming tube.

[0012] In a second aspect, the present invention provides a material measurement method, which is applied to the material measuring device provided in the above embodiment, and the material measurement method includes: detecting the pressure difference data of the multiphase material to be measured in the pressure difference forming tube based on the material measuring device; calculating the initial air flow rate in the multiphase material to be measured based on the pressure difference data; obtaining the temperature data and pressure data of the multiphase material to be measured based on the detection of the material measuring device; determining the proportion data of a specified phase in the multiphase material to be measured in each phase based on the temperature data, the pressure data and the component parameters of the multiphase material to be measured; wherein the component parameters of the multiphase material to be measured are pre-parameters; and calculating the flow rate of each phase of the multiphase material to be measured based on the proportion data and the initial air flow rate.

[0013] In the above implementation, the flow state of the multiphase material under test can be better characterized by combining it with the fluid state parameters of the multiphase material under test. Based on this state, the flow rate of each phase in the multiphase material under test can be more accurately determined. Specifically, the fluid state parameters include resonance data caused by the multiphase material under test; the cross-sectional moisture content of the multiphase material under test can be calculated by combining the resonance data; and the cross-sectional moisture content can be combined with the ratio data to obtain the flow rate of each phase in the multiphase material under test.

[0014] In an optional embodiment, the flow rate of each phase of the multiphase material to be measured is calculated based on the proportional data and the initial air flow rate, including: detecting resonance data formed under the action of the multiphase material to be measured based on the microwave resonance sensor of the material measuring device; calculating the cross-sectional moisture content in the multiphase material to be measured based on the resonance data; calculating the flow rate of each phase of the multiphase material to be measured based on the proportional data, the initial air flow rate and the cross-sectional moisture content.

[0015] In an optional embodiment, the flow rate of each phase of the multiphase material to be measured is calculated based on the proportional data, the initial air flow rate and the cross-sectional moisture content, including: using an ultra-virtual height model to calculate the flow rate of each phase of the multiphase material to be measured based on the proportional data, the initial air flow rate and the cross-sectional moisture content; wherein the ultra-virtual height model is determined based on a deformation of the Venturi virtual height model.

[0016] In the above implementation method, for cases where the content of some specific substances is small, the super-virtual height model can be combined for measurement and calculation. Since the super-virtual height model is based on modeling data to correct the parameters of the Venturi virtual height model, the influence of data such as gas phase flow rate, liquid phase content, gas-liquid density ratio, gas-liquid viscosity ratio, and pressure difference on the virtual height model is combined to correct the parameters in the model, which can make the values ​​obtained by the super-virtual height model more reliable and accurate.

[0017] In an optional embodiment, the method for determining the super-virtual height model includes: performing parameter correction on the Venturi virtual height model based on modeling data to obtain the super-virtual height model; wherein the modeling data includes one or more of the gas phase flow rate, liquid phase content, gas-liquid density ratio, gas-liquid viscosity ratio, and pressure difference of the sample in the material measuring device, and the sample is a mixture to which the multi-phase material to be measured belongs.

[0018] In an optional embodiment, the method further includes: detecting wire mesh cross-sectional image data of the flow cross-section of the multiphase material to be measured based on the wire mesh sensor of the material measuring device; identifying the flow type of the multiphase material to be measured according to the wire mesh cross-sectional image data to obtain an identification result; when the identification result characterizes that the flow type of the multiphase material to be measured is a specified flow type, then executing the use of the super-virtual height model to calculate the flow rate of each phase of the multiphase material to be measured according to the proportional data, the initial air flow rate and the cross-sectional moisture content.

[0019] In an optional embodiment, identifying the flow type of the multiphase material to be tested based on the cross-sectional image data to obtain a recognition result includes: using a flow type recognition model to identify the flow type of the multiphase material to be tested based on the cross-sectional image data to obtain a recognition result; wherein, the flow type recognition model includes a long short-term memory network, and the long short-term memory network includes a forgetting gate, an input gate, and an output gate.

[0020] In the above-mentioned implementation method, the method of identifying the substance content can also be selected in combination with the flow type of the multiphase material to be measured, so that the determined model can be more suitable for the measurement of the flow type of the current multiphase material to be measured, and the flow rate results of each phase of the multiphase material to be measured can also be more reliable.

[0021] In an optional embodiment, the method further includes: obtaining velocity-related data of the multiphase material to be measured in the pressure difference forming tube based on the velocity sensor of the material measuring device; inputting the velocity-related data into a first network model for identification to identify the flow rate of each phase in the multiphase material to be measured; wherein, the first network model is obtained by supervised training based on the velocity-related data of the multiphase material in the material measuring device.

[0022] In the above implementation, a neural network model can also be used to identify the multiphase material to be measured, which can reduce the data resources required for collection. The recognition based on the neural network model can also further improve the accuracy of the measurement.

[0023] In an optional embodiment, the method further includes: inputting the velocity-related data into a second network model for identification to identify the content of each phase in the multiphase material to be measured; wherein the second network model is obtained by supervised training based on the velocity-related data of the multiphase material in the material measuring device.

[0024] In an optional embodiment, the speed-related data includes vector acceleration modulus and precession frequency; the method for determining the first network model includes: using training data to perform supervised training on the initial model to obtain the first network model; wherein the training data includes a vector acceleration modulus sample set, a precession frequency sample set, and a phase flow label set, and the initial model includes a two-layer fully connected neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A schematic structural diagram of a material measuring device provided in an embodiment of the present application;

[0027] Figure 2 A flow chart of a material measurement method provided in an embodiment of the present application;

[0028] Figure 3 A partial flow chart of the substance measurement method provided in an embodiment of the present application;

[0029] Figure 4 Another partial flow chart of the substance measurement method provided in an embodiment of the present application;

[0030] Figure 5 A deviation diagram of the predicted value and actual value of the gas phase volume flow rate provided by the first network model in an embodiment of the present application;

[0031] Figure 6 A training residual graph of the first network model provided in an embodiment of the present application;

[0032] Figure 7 A deviation diagram of the predicted value and actual value of liquid phase content of the second network model provided in the embodiment of the present application;

[0033] Figure 8 This is a training residual graph of the second network model provided in an embodiment of the present application.

[0034] Icons: 110-pressure difference forming tube; 120-pressure sensor; 130-microwave resonance sensor; 140-pressure difference sensor; 150-speed sensor; 160-flow pattern adjustment device; 170-wire mesh sensor. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings in the embodiments of the present application.

[0036] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0037] In the description of this application, it should be noted that the terms "upper", "lower", "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or are the orientations or positional relationships in which the inventive products are usually placed when in use. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be interpreted as a limitation of this application.

[0038] It should also be noted that, in the description of this application, unless otherwise expressly specified or limited, the terms "disposed," "installed," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; mechanical connections, electrical connections; direct connections, indirect connections through an intermediate medium, and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0039] Figure 1 This is a schematic diagram of the structure of the material measuring device provided in the embodiment of the present application. Figure 1 As shown, the substance measuring device may include: a pressure difference forming tube 110, a pressure sensor 120, a microwave resonance sensor 130, a pressure difference sensor 140 and a temperature sensor.

[0040] Alternatively, the pressure differential forming tube 110 may be a venturi tube. Objects of different flow rates flowing through the venturi tube generate different pressure losses, thereby forming a pressure differential effect. The pressure differential forming tube 110 may also be other structures capable of forming a pressure differential.

[0041] Optionally, the pressure difference forming tube 110 may also include multiple sections of pipes with different pore diameters.

[0042] The pressure sensor 120 can be installed in the pressure differential forming tube 110 to detect pressure data in the pressure differential forming tube 110. Alternatively, the pressure sensor 120 can be installed in the upstream section of the pressure differential forming tube 110 to detect pressure data in the upstream section.

[0043] A temperature sensor may also be installed on the pressure differential forming tube 110 to measure the temperature data of the substance to be measured. Optionally, the pressure sensor 120 and the temperature sensor may be two independent sensors or a combined temperature and pressure sensor.

[0044] In this embodiment, the microwave resonant sensor 130 is connected to the pressure differential forming tube 110. For example, the microwave resonant sensor 130 can be connected to one end of the pressure differential forming tube 110. For example, the pressure differential forming tube 110 can be a venturi tube, and the microwave resonant sensor 130 can be installed at the throat of the venturi tube.

[0045] Exemplarily, the microwave resonant sensor 130 may include a resonant cavity and a microwave detector. The inner annular surface of the resonant cavity may form a channel of the microwave resonant sensor 130 , and the channel is in communication with the pressure difference forming tube 110 .

[0046] The microwave detector can transmit or receive signals to the resonant cavity. When the channel of the resonant cavity passes through different materials, the resonance data formed will also be different.

[0047] The microwave resonance sensor 130 is used to collect resonance data when the multiphase to be measured flows through the channel of the resonant cavity.

[0048] The pressure difference sensor 140 may be installed in the pressure difference forming tube 110 to detect the pressure difference between at least two positions of the pressure difference forming tube 110 .

[0049] The substance measurement device may be equipped with multiple differential pressure sensors 140 to measure the pressure differential between multiple groups of locations. Optionally, the differential pressure sensors 140 may include a first differential pressure sensor 140 and a second differential pressure sensor 140. The first differential pressure sensor 140 is used to detect the pressure differential between two locations with different pore sizes in the pressure differential forming tube 110; the second differential pressure sensor is used to detect the pressure differential between the inlet and outlet sections of the substance measurement device.

[0050] Optionally, the material measuring device may further include a velocity sensor 150. When a multiphase flow passes through the pressure difference forming pipe 110, the velocity sensor 150 is triggered to detect data related to the flow velocity of the multiphase in the pipe.

[0051] The velocity sensor 150 can be used to detect velocity data related to the multiphase material to be measured in the pressure difference forming tube 110. The velocity data can include accelerations in three axial directions and the angular velocity of the multiphase material to be measured.

[0052] Optionally, the velocity sensor 150 may be an inertial measurement unit (IMU), which can measure data such as acceleration and angular velocity of a multiphase substance flowing in a pipe.

[0053] For example, the inertial measurement unit may be fixedly installed in the pressure difference forming tube 110 . When a multiphase substance flows in the pressure difference forming tube 110 , the inertial measurement unit may rotate under the action of the flow force of the multiphase substance.

[0054] Optionally, there may be multiple speed sensors 150 . The speed sensor 150 may be installed at a position between the center of the pressure differential forming tube 110 and the tube wall. The installation position of the speed sensor does not coincide with the center of the pressure differential forming tube 110 .

[0055] For example, the velocity sensor 150 may be installed in a container, which is fixed in the pressure difference forming tube 110. For example, the container may be a straight tube.

[0056] For some multiphase fluids, especially those containing multiple immiscible substances, the flow pattern of the fluid can take on various forms. Measuring some flow patterns of fluids is more difficult, and measurement deviations can be significant. To address this, the material measuring device may further include a flow pattern adjustment device 160 for adjusting the flow pattern of the multiphase material entering the material measuring device.

[0057] Optionally, the flow pattern adjustment device 160 may include one or more of a guide vane, a rotator, and a rectifier.

[0058] For example, the guide vane can be installed near the inlet end of the material measuring device, and the vortexer can also be installed near the inlet end. The guide vane can be installed on the vortexer. The rectifier can be installed near the outlet end of the material measuring device.

[0059] Depending on the actual usage scenario, the material measurement device may also include additional detection components. For example, to facilitate identification of the flow pattern of the multiphase material under test, a wire mesh sensor 170 (WMS) may be provided to capture cross-sectional image data of the flow cross section of the multiphase material under test. For example, the wire mesh sensor 170 may be installed at one of the locations within the channel of the pressure differential forming tube 110 of the material measurement device.

[0060] For example, the wire mesh sensor can be a resistive wire mesh imaging sensor. The wire mesh imaging results obtained by this sensor can reflect the spatiotemporal distribution of gas and liquid within the pipeline. In one example, the wire mesh sensor can adopt a 16*16 orthogonal layout, achieving a maximum resolution of 3.125mm in a 50mm pipe diameter. The wire mesh sensor system can achieve an imaging rate of 250 frames per second. Of course, in other examples, with different wire mesh sensor layouts, it can be applied to pipes of different diameters, potentially achieving different resolutions and imaging rates.

[0061] Taking a 16*16 orthogonal layout wire mesh sensor as an example, the wire mesh sensor adopts a cyclic scanning excitation method when working. The 16 excitation electrodes share the same signal source and need to be matched with a suitable electronic analog switch. For example, the typical value of the on-resistance can be 4Ω, allowing a bipolar signal of ±5V-±15V to pass through. The maximum current allowed to pass continuously through a single channel is 115mA, and the typical switching time is 140ns. In order to prevent mutual influence between different excitation electrodes, during the excitation cycle of a certain electrode, the other excitation electrodes are forced to be grounded. In the physical context, the change in the local liquid content at the intersection of the excitation electrode and the receiving electrode reflects the size of the resistance. Under the condition of constant voltage source excitation, the change in resistance is reflected as a change in current. Since it is difficult to directly measure the size of the current signal, in circuit design, the changing current signal is usually converted into a voltage signal, but the converted voltage signal is still a square wave.

[0062] In some scenarios, the flow rates of each phase of the multiphase material to be measured can also be identified using the vortex precession frequency. Based on this, the material measurement device can also include a vortex flowmeter to measure the precession frequency. The precession frequency can also be measured using an inertial measurement unit.

[0063] The above-mentioned material measuring device can realize the measurement of the flow rate of each phase in the multi-phase material to be measured in various scenarios. The following describes the measurement that can be realized by the material measuring device in conjunction with the embodiment of the material measurement method.

[0064] like Figure 2 As shown, Figure 2 The flow chart of a substance measurement method provided by the embodiment of the present application is shown. The steps of the substance measurement method in the embodiment of the present application can be performed by the above-mentioned substance measurement device. Figure 2 The specific process shown is explained in detail.

[0065] Step 210: Detecting pressure difference data of the multi-phase substance to be measured in the pressure difference forming tube based on the substance measuring device.

[0066] like Figure 1 As shown, two differential pressure sensors can be used to measure two sets of differential pressure data. One set of differential pressure data represents the pressure difference at two locations of different pipe diameters upstream of the microwave resonant sensor. The other set of differential pressure data can represent the pressure difference between the inlet and outlet sections of the material measuring device.

[0067] Step 230: Calculate the initial gas flow rate in the multiphase material to be measured based on the pressure difference data.

[0068] Step 250: obtaining temperature data and pressure data of the multiphase material to be measured based on detection by the material measuring device.

[0069] Optionally, the material measuring device may be provided with an integrated temperature and pressure sensor, and the temperature data and pressure data may be detected based on the integrated temperature and pressure sensor provided with the material measuring device. Optionally, the material measuring device may be provided with a temperature sensor and a pressure sensor, and the temperature data may be measured by the temperature sensor, and the pressure data may be measured by the pressure sensor.

[0070] Step 270 , determining the proportion data of the designated phase in each phase of the multiphase material to be measured based on the temperature data, the pressure data, and the component parameters of the multiphase material to be measured.

[0071] The component parameters of the multiphase material to be measured are pre-parameters, and the component parameters of the multiphase material to be measured may include the components contained in the multiphase material to be measured.

[0072] The designated phase mentioned above may be the ratio data of the gas phase in each phase of the multiphase substance to be measured.

[0073] For example, the ratio data may include a characteristic ratio of substances in different phases. The characteristic ratio may be a gas-liquid viscosity ratio μ l / μ g , gas-liquid density ratio ρ l / ρ g Alternatively, the ratio data of each phase in the multiphase material to be measured can be calibrated in advance based on the research and analysis of the multiphase material to be measured.

[0074] Optionally, the above-mentioned ratio data may also include the mass ratio of the gas phase in each phase.

[0075] Optionally, the above gas-liquid viscosity ratio μ l / μ g , gas-liquid density ratio ρ l / ρ g It can also be a value set in advance. Based on the actual measured temperature data and pressure data, the gas-liquid viscosity ratio μ l / μ g , gas-liquid density ratio ρ l / ρg Make adaptive corrections.

[0076] Step 290 , calculating the flow rate of each phase of the multiphase material to be measured based on the ratio data and the initial gas flow rate.

[0077] Optionally, a Venturi false height model may be used to determine the flow rate of each phase of the multiphase material to be measured based on the above-mentioned ratio data.

[0078] Optionally, the fluid state parameters may also include resonance data caused by the multiphase material to be measured. Exemplarily, the resonance data may be a microwave signal generated when the multiphase material to be measured passes through a channel of a microwave resonant sensor. Exemplarily, the microwave resonant sensor may include a microwave signal transmitter and a receiver, and the resonance data may be a signal received by the microwave signal receiver when the multiphase material to be measured passes through the channel of the microwave resonant sensor.

[0079] Optionally, the Venturi virtual height model can be shown in the following formula:

[0080]

[0081] Among them, OR represents an artificial high value; C1, C2, and C3 represent constants; μ l / μ g represents the gas-liquid viscosity ratio; ρ l / ρ g Indicates the gas-liquid density ratio; Fr g represents the gas phase Froude number; β l Indicates the liquid volume fraction.

[0082] Optionally, an initial gas phase virtual high flow rate may be calculated based on the density of each substance in the multiphase substance to be measured; the gas phase virtual high flow rate and the Venturi virtual high flow rate are used to determine the gas phase flow rate.

[0083] Optionally, step 290 may include step 291 and step 292 .

[0084] Step 291 : Using a microwave resonance sensor of a material measuring device, resonance data generated under the action of the multiphase material to be measured is detected.

[0085] Step 292: Calculate the cross-sectional moisture content in the multiphase material to be measured based on the resonance data.

[0086] Step 293 , calculating the flow rate of each phase of the multiphase material to be measured based on the ratio data, the initial gas flow rate, and the cross-sectional moisture content.

[0087] By combining the above-mentioned Venturi false height model and the determined water content in the multiphase material to be measured, the flow rate of each phase in the multiphase material to be measured can be analyzed.

[0088] In this embodiment, after determining the gas phase flow rate of the multiphase substance to be measured, the flow rates of substances such as the water flow rate and the oil flow rate are determined in combination with the above-mentioned cross-sectional water content.

[0089] In one example, the multiphase substance to be measured may be a mixture of oil, gas, and water. After determining the gas phase flow rate and cross-sectional water content of the mixture, the oil flow rate can be determined.

[0090] Of course, the multiphase substance to be measured may also be other mixtures. Based on the water flow rate and the air flow rate, the flow rate of other substances can be determined.

[0091] Optionally, the above step 293 may include: using an over-virtual height model to calculate the flow rate of each phase of the multiphase material to be measured according to the cross-sectional moisture content and ratio data.

[0092] Among them, the super virtual height model is determined based on the deformation of the Venturi virtual height model.

[0093] Optionally, the method for determining the super-virtual height model includes: performing parameter correction on the Venturi super-virtual height model based on modeling data to obtain the super-virtual height model. The modeling data includes one or more of the following: gas phase flow rate, liquid phase holdup, gas-liquid density ratio, gas-liquid viscosity ratio, and pressure difference of the sample in the material measurement device. The sample may be a multiphase mixture used to construct the super-virtual height model.

[0094] In this embodiment, complex flow patterns, such as slug flow, are more difficult to detect. Therefore, a swirler can be installed on the pressure differential forming tube of the material measuring device to adjust the flow pattern. This swirler can be a precession-type swirler, thereby changing the throttling pattern. This change in throttling pattern affects both the differential pressure of the pressure differential forming tube and the total pressure loss differential pressure.

[0095] Based on this, the present embodiment updates the Venturi virtual height model to a super virtual height model. The rules of this super virtual height model differ from those of the original virtual height model. Based on this, we study the effects of key parameters such as gas velocity, liquid holdup, gas-liquid density ratio, and gas-liquid viscosity ratio on the virtual height model.

[0096] Alternatively, fluid simulation can be used to change the above parameters, obtain the information law of the differential pressure of the special-shaped Venturi, and use dimensional analysis to model and study the high-liquid-containing super-virtual high model. For example, the super-virtual high model can be expressed as:

[0097]

[0098] Among them, β lrepresents the liquid volume fraction; β represents the throttling ratio of the special-shaped Venturi; Resg represents the gas phase apparent Reynolds number, H is the dimensionless number characterizing the liquid phase surface tension σl, Frg is the gas phase Froude number; P is the system pressure.

[0099] Among them, the throttling ratio β of the special-shaped Venturi and the surface tension H of the liquid phase are fixed values, the gas phase apparent Reynolds number Resg can be expressed by the gas phase Froude number Frg, and the density ratio reflects the pressure P. The functional relationship of the super-virtual height model can be simplified as follows:

[0100]

[0101] By combining the super-virtual high model described above, it is possible to better adapt to more complex scenarios of multi-phase objects to be measured, and the accuracy of the determined flow rates of each phase can also be improved.

[0102] In this embodiment, for some flow types that are relatively difficult to identify, a method with higher accuracy and recognition range can be used to achieve it.

[0103] Before step 270, Figure 3 As shown, the material measurement method may further include step 261 and step 262 .

[0104] Step 261 : Detecting the screen cross-sectional image data of the flow cross-section of the multi-phase material to be measured based on the screen sensor of the material measuring device.

[0105] Optionally, the cross-sectional image data may be acquired using a wire mesh sensor.

[0106] Step 262 : Identify the flow pattern of the multiphase material to be tested based on the screen cross-sectional image data to obtain an identification result.

[0107] When the identification result indicates that the flow pattern of the multiphase material to be measured is the specified flow pattern, the step of using the super-high model to calculate the flow rate of each phase of the multiphase material to be measured according to the proportion data, the initial gas flow rate and the cross-sectional moisture content is performed.

[0108] Exemplarily, the designated flow pattern may be a flow pattern that is relatively difficult to measure, for example, the designated flow pattern may be a slug flow.

[0109] Optionally, when the identified flow pattern is a common flow pattern that is easy to identify and measure, such as a stratified wavy flow, a wavy flow, annular flow, etc., other calculation methods can be used to calculate the flow rate of each phase of the multiphase material to be measured.

[0110] Optionally, the above-mentioned step 262 may include: using a flow pattern recognition model to identify the flow pattern of the multiphase material to be measured according to the cross-sectional image data to obtain an identification result.

[0111] Long Short-Term Memory (LSTM) networks are commonly used to process time series prediction problems. The flow pattern recognition model of this embodiment may include a LSTM network, which includes a forget gate, an input gate, and an output gate.

[0112] Exemplarily, the long short-term memory network includes a forget gate, an input gate, and an output gate. t It can represent the input value of the network at the current moment, h t-1 and h t are the output values ​​of the network at the previous moment and the current moment respectively, C t-1 and C t are the unit states of the previous moment and the current moment respectively, x t 、h t-1 and C t-1 is the input, h t and C t is the output. t To determine C t-1 How much of it is retained in C t , input gate i t To determine x t How much of it is retained in C t , output gate o t To determine C t Whether as h t The calculation formulas for each part are as follows:

[0113] Forget Gate f t It can be expressed as: t =sigmoid(x t W xf +h t-1 W hf +b f );

[0114] Input gate i t It can be expressed as: t =sigmoid(x t W xi +h t-1 W hi +b i )and

[0115] Unit Status C t The update can be achieved by the following formula:

[0116] Output gate o t It can be expressed as: t =sigmoid(xt W xo +h t-1 W ho +b c ) and h t =o t tanh(C t ).

[0117] Among them, W and b in the above formulas are weight parameters and bias parameters respectively, is a candidate memory unit, and the activation function sigmoid converts f t 、i t and o t The value of is controlled in the range of 0 to 1, and the output value of the activation function tanh can range from -1 to 1.

[0118] For example, two normalized phase time series signals from the horizontal and vertical electrodes collected by the wire mesh sensor can be used as network input, and the flow type can be used as the network output. Optionally, the output can include four categories: stratified wavy flow, wavy flow, slug flow, and annular flow. The sample data was partitioned using a sliding window segmentation method, with a window length of 2000 and a sliding step size of 100. A total of 10,062 sample data were obtained, with 70% of the data used as the training set and 30% as the test set.

[0119] Optionally, network training and testing can be implemented based on PyTorch, using the cross-entropy loss function and Adam optimizer, with a learning rate of 0.001, a training batch size of 64, and an epoch of 30.

[0120] It should be understood that the parameter settings used in the above model are merely illustrative. Based on different accuracy and other requirements, the above values ​​can be adaptively adjusted. For example, the ratio of the training set to the test set can be adjusted. Parameters such as the learning rate, batch size, and epoch number can also be adjusted appropriately based on actual needs.

[0121] Based on the above flow pattern identification, appropriate calculation logic can be selected in combination with the flow pattern identification result, and a measurement calculation model that is more suitable for the current flow pattern can be selected, so that the flow rate of each phase in the multiphase material to be measured can be more accurately identified.

[0122] In the above various embodiments, the flow rate of each phase of the multiphase material to be measured can be identified. In some scenarios, when it is inconvenient to collect some data, the flow rate can also be measured and calculated based on the following method.

[0123] In an optional embodiment, the material measuring device can also measure velocity-related data of the multiphase material to be measured, and the flow rate of the gas phase in the multiphase material to be measured can be determined in combination with the velocity-related data.

[0124] For example, the velocity-related data may include acceleration, angular velocity, and other data. The velocity-related data may also include information such as vector acceleration modulus and precession frequency. The vector acceleration modulus may be calculated based on basic velocity data, such as acceleration and angular velocity.

[0125] Optionally, the above-mentioned acceleration, angular velocity and other data can be measured using an inertial measurement unit.

[0126] Optionally, the precession frequency can be measured using a vortex flowmeter, which can be installed in the pressure difference forming tube.

[0127] For example, the vector acceleration modulus can be expressed as:

[0128]

[0129] Among them, a i Indicates the acceleration of each axis; Represents the vector acceleration modulus.

[0130] like Figure 4 As shown, the material measurement method provided in this embodiment may further include: step 2010, inputting velocity-related data into the first network model for identification to identify the flow rate of each phase in the multi-phase material to be measured.

[0131] The first network model is obtained by supervised training based on velocity-related data of multiphase matter in a material measurement device.

[0132] The velocity-related data may be velocity-related data of the multiphase material to be measured in the pressure difference forming tube obtained by a velocity sensor based on the material measuring device.

[0133] For example, the velocity sensor may be an inertial measurement unit, which may be used to obtain data such as acceleration and angular velocity of the multiphase material to be measured while flowing in the pressure difference forming tube.

[0134] The speed-related data input into the first network model may include a vector acceleration modulus and a precession frequency. The vector acceleration modulus may be calculated based on the acceleration of each axis measured by a speed sensor. The precession frequency may be data measured using a vortex flowmeter.

[0135] Exemplarily, the first network model may output the gas phase volume flow rate of the multiphase substance to be measured.

[0136] like Figure 4 As shown, the material measurement method provided in this embodiment may further include: step 2011, inputting velocity-related data into a second network model for identification, so as to identify the content of each phase in the multi-phase material to be measured.

[0137] in, Figure 4 The order shown is only exemplary. The execution order of step 2010 and step 2011 can be as follows: Figure 4 The example execution shown can also be done with Figure 4 Execute in reverse order.

[0138] For example, the second network model can be used to predict the liquid volume fraction in the multiphase material to be measured.

[0139] The second network model is obtained by supervised training based on velocity-related data of multiphase matter in a material measurement device.

[0140] The velocity-related data input to the second network model may include vector acceleration modulus and precession frequency. For example, the second network model may output the liquid phase volume fraction of the multiphase material to be measured.

[0141] In this embodiment, the method for determining the first network model includes: using training data to perform supervised training on the initial model to obtain the first network model.

[0142] The training data may include a vector acceleration modulus sample set, a precession frequency sample set, and a flow label set of each phase, and the initial model of the first network model includes a two-layer fully connected neural network.

[0143] In this embodiment, the vector acceleration modulus and precession frequency can be used as input data of the first network model, and the output data can include the flow rate of each phase in the multiphase material to be measured. The output data can also include the volume flow rate of the gas phase in the multiphase material.

[0144] The following combination Figure 5 and Figure 6 Let’s introduce the prediction of the first network model mentioned above. Figure 5 and Figure 6 The example shown is the result of training a model with a training time of 37.17 seconds and 10 nodes per layer. The average absolute error (MAE) is 1.1184, the R-square is 0.99, the mean square error (MSE) is 4.2613, and the root mean square error (RMSE) is 2.0643.

[0145] in, Figure 5 This is a deviation diagram of the predicted value and actual value of the gas phase volume flow rate provided by the first network model in the embodiment of the present application, Figure 6This is the training residual graph of the first network model provided in the embodiment of the present application. In the actual value and predicted value deviation graph, when the gas phase volume flow rate is 49.48m 3 / h、70.68m 3 / h、84.82m 3 / h、106.03m 3 / h, the corresponding gas phase velocity is 7m / s, 10m / s, 12m / s, 15m / s respectively. The regression model prediction value at this time is Figure 5 The predicted values ​​of the first network model are evenly distributed near the diagonal of , and most of the predicted values ​​of the first network model fall on the diagonal represented by the actual observed values, with a very small error.

[0146] The training residual graph shows the residual distribution of the first network model, that is, the difference between the actual value and the predicted value of the first network model. In the training residual graph, it can be seen that when the gas phase volume flow rate is 49.48m 3 / h、70.68m 3 / h、84.82m 3 / h、106.03m 3 / h, the corresponding gas phase flow rates are 7m / s, 10m / s, 12m / s, and 15m / s respectively. At this time, the residuals are distributed between ±5, and the residual values ​​are relatively small. Figure 5 and Figure 6 As shown in the figure, the error between the predicted value of the first network model and the actual value of the experiment is small, and the residual error of the model is small. The first network model has a high prediction accuracy for the gas phase volume content.

[0147] In this embodiment, the second network model is determined by performing supervised training on the initial model using training data to obtain the second network model.

[0148] The training data may include a vector acceleration modulus sample set, a precession frequency sample set, and a flow label set for each phase. The initial model of the second network model may also include a two-layer fully connected neural network.

[0149] In this embodiment, the vector acceleration modulus and precession frequency may be used as input data of the second network model, and the output data may also be the liquid volume fraction in the multiphase material.

[0150] The following combination Figure 7 and Figure 8 Let’s introduce the prediction of the second network model mentioned above. Figure 7 and Figure 8The example shown is the result of training a model with a training time of 37.17 seconds and 10 nodes per layer. The average absolute error (MAE) is 0.0069876, the R-square is 0.97, the mean square error (MSE) is 0.00019, and the root mean square error (RMSE) is 0.013846.

[0151] in, Figure 7 This is a deviation diagram of the predicted value and actual value of the liquid phase content of the second network model provided in the embodiment of the present application, Figure 8 This is the training residual graph of the second network model provided in the embodiment of the present application. In the actual value and predicted value deviation graph, when the gas phase volume flow rate is 49.48m 3 / h、70.68m 3 / h、84.82m 3 / h、106.03m 3 / h, the corresponding gas phase velocity is 7m / s, 10m / s, 12m / s, 15m / s respectively. The regression model prediction value at this time is Figure 7 The predicted values ​​of the second network model are evenly distributed near the diagonal of , and most of the predicted values ​​of the second network model fall on the diagonal represented by the actual observed values, with a very small error.

[0152] The training residual graph shows the residual distribution of the second network model, that is, the difference between the actual value and the predicted value of the second network model. In the training residual graph, it can be seen that when the gas phase volume flow rate is 49.48m 3 / h、70.68m 3 / h、84.82m 3 / h、106.03m 3 / h, the corresponding gas phase velocity is 7m / s, 10m / s, 12m / s, 15m / s, and the residuals are distributed between ±0.01, which is a small residual value. Figure 7 and Figure 8 As shown in the figure, the error between the predicted value of the second network model and the actual value of the experiment is small, and the residual error of the model is small. Therefore, it can be determined that the second network model has a high prediction accuracy for liquid volume fraction.

[0153] The two network-based models provided in the embodiments of this application provide relatively high accuracy in identifying gas phase volume flow and liquid phase volume fraction. Based on the above-described predictive logic, the flow rate and liquid phase volume fraction of the multiphase material to be measured can be identified while collecting less data and requiring less testing equipment.

[0154] The foregoing is merely an optional embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included within the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined or explained in subsequent figures.

[0155] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A material measuring device, characterized in that: The substance measuring device comprises: a pressure difference forming tube, wherein at least two positions of the pressure difference forming tube have different pore diameters; a pressure sensor installed upstream of the pressure difference forming pipe, for detecting pressure data in the pressure difference forming pipe; A temperature sensor installed in the pressure difference forming pipe is used to detect the temperature data of the multiphase material to be measured; a microwave resonance sensor connected to the pressure difference forming tube, the microwave resonance sensor comprising a channel, the channel being in communication with the pressure difference forming tube and being used to collect resonance data caused by the multiphase substance to be measured; The pressure difference sensor installed on the pressure difference forming tube is used to detect the pressure difference at at least two positions of the pressure difference forming tube.

2. The device according to claim 1, characterized in that The differential pressure sensor further includes: a first differential pressure sensor and a second differential pressure sensor; The first differential pressure sensor is used to detect the pressure difference between two positions of the pressure differential forming tube with different apertures; The second differential sensor is used to detect the pressure difference between the inlet section and the outlet section of the material measuring device.

3. The device according to claim 1, characterized in that Also includes: The velocity sensor installed in the pressure difference forming tube is used to check the velocity-related data of the multiphase material to be measured in the pressure difference forming tube.

4. The device according to claim 1, characterized in that Also includes: The flow pattern adjustment device is used to adjust the flow pattern of the multiphase material to be measured entering the material measuring device.

5. The device according to claim 4, characterized in that Also includes: The flow pattern adjustment equipment includes one or more of: guide vanes, rotators, and rectifiers.

6. The device according to any one of claims 1 to 5, characterized in that include: The wire mesh sensor is used to collect image data of the multiphase substance to be measured flowing in the pressure difference forming tube.

7. A method for measuring a substance, characterized in that: Applicable to the material measuring device according to any one of claims 1 to 7, the material measuring method comprising: The material measuring device detects pressure difference data of the multiphase material to be measured in the pressure difference forming tube; Calculating the initial gas flow rate in the multiphase substance to be measured based on the pressure difference data; Obtaining temperature data and pressure data of the multiphase substance to be measured based on detection by the substance measuring device; Determining the proportion data of the designated phase in each phase of the multiphase material to be measured based on the temperature data, the pressure data, and the component parameters of the multiphase material to be measured; wherein the component parameters of the multiphase material to be measured are pre-parameters; The flow rates of each phase of the multiphase material to be measured are calculated according to the ratio data and the initial gas flow rate.

8. The method according to claim 7, characterized in that The calculating the flow rate of each phase of the multiphase material to be measured according to the ratio data and the initial gas flow rate includes: The microwave resonance sensor of the material measuring device detects resonance data formed under the action of the multiphase material to be measured; Calculating the cross-sectional moisture content of the multiphase material to be measured based on the resonance data; The flow rates of each phase of the multiphase material to be measured are calculated according to the ratio data, the initial gas flow rate and the cross-sectional moisture content.

9. The method according to claim 8, characterized in that Calculating the flow rate of each phase of the multiphase material to be measured based on the ratio data, the initial gas flow rate, and the cross-sectional moisture content includes: The flow rates of each phase of the multiphase material to be measured are calculated using a super-virtual height model according to the proportion data, the initial air flow rate and the cross-sectional moisture content; wherein the super-virtual height model is determined based on a deformation of a Venturi virtual height model.

10. The method according to claim 9, characterized in that The method for determining the super-high model includes: The parameters of the Venturi virtual height model are corrected based on the modeling data to obtain the super virtual height model; wherein the modeling data includes one or more of the gas phase flow rate, liquid phase content, gas-liquid density ratio, gas-liquid viscosity ratio, and pressure difference of the sample in the material measuring device, and the sample is a mixture to which the multiphase material to be measured belongs.

11. The method according to claim 9, characterized in that The method further comprises: Detecting wire mesh cross-sectional image data of the flow cross-section of the multiphase material to be measured based on the wire mesh sensor of the material measuring device; Identifying the flow pattern of the multiphase material to be measured based on the screen cross-sectional image data to obtain an identification result; When the identification result indicates that the flow pattern of the multiphase material to be measured is a specified flow pattern, the super-virtual height model is used to calculate the flow rate of each phase of the multiphase material to be measured based on the proportion data, the initial air flow rate and the cross-sectional moisture content.

12. The method according to claim 11, characterized in that The identifying the flow pattern of the multiphase material to be measured according to the cross-sectional image data to obtain an identification result includes: using a flow pattern identification model to identify the flow pattern of the multiphase material to be measured according to the cross-sectional image data to obtain an identification result; The flow pattern recognition model includes a long short-term memory network, and the long short-term memory network includes a forget gate, an input gate and an output gate.

13. The method according to claim 7, characterized in that The method further comprises: obtaining velocity-related data of the multiphase substance to be measured in the pressure difference forming tube based on the velocity sensor of the substance measuring device; The velocity-related data is input into a first network model for identification to identify the flow rate of each phase in the multiphase material to be measured; wherein the first network model is obtained by supervised training based on the velocity-related data of the multiphase material in the material measurement device.

14. The method according to claim 13, characterized in that The method further comprises: The velocity-related data is input into a second network model for identification to identify the fraction of each phase in the multiphase material to be measured; wherein the second network model is obtained by supervised training based on the velocity-related data of the multiphase material in the material measurement device.

15. The method according to claim 13, characterized in that The speed-related data includes vector acceleration modulus and precession frequency; The method for determining the first network model includes: The initial model is supervisedly trained using training data to obtain the first network model; wherein the training data includes a vector acceleration modulus sample set, a precession frequency sample set, and a flow label set for each phase, and the initial model includes a two-layer fully connected neural network.

Images