An oilfield wellhead multi-parameter measuring device, control system and measuring method

Abstract

This invention provides a device, control system, and measurement method for multi-parameter measurement at oilfield wellheads, belonging to the field of petroleum production logging technology. The device includes a gas-liquid separation subsystem, a control valve group, a dynamic measurement subsystem, an cumulative conductivity fiber optic measurement subsystem, a horizontal water cut and flow rate fusion measurement subsystem, a gas volume measurement subsystem, and a turbine flow rate measurement subsystem. This invention, employing the aforementioned device, control system, and measurement method for multi-parameter measurement at oilfield wellheads, solves the problems of outdated traditional measurement methods, low measurement accuracy of existing devices, limited functionality, high maintenance costs, poor compatibility of control systems, and imperfections in measurement methods.

Description

Technical Field

[0001] This invention relates to the field of oil production logging technology, and in particular to a device, control system and measurement method for multi-parameter measurement at oilfield wellheads. Background Technology

[0002] In oil extraction, fluid parameters such as water cut, gas cut, and phase flow rate of the three-phase flow (oil, gas, and water) at the wellhead are core data reflecting the well's production status, guiding adjustments to oilfield development plans, and achieving efficient oil and gas resource extraction. The real-time performance, accuracy, and stability of these measurements directly affect the efficiency and profitability of oilfield production. Currently, there are still many problems to be solved in the measurement technology of the three-phase flow parameters at the wellhead. Traditional measurement methods often involve manual sampling followed by offline analysis at a central laboratory. This method is not only cumbersome and time-consuming, but also exhibits significant measurement lag, failing to reflect the real-time production status of the well in a timely manner. Furthermore, it requires substantial on-site manpower and resources for sampling and delivery, significantly increasing the operating costs of oilfield production. Additionally, manual operation is prone to introducing human error, making it difficult to guarantee the accuracy of the measurement data.

[0003] While existing online measurement devices and technologies have addressed the lag issue of offline analysis to some extent, they generally suffer from limitations such as single-parameter measurement and poor adaptability. Most devices can only measure single parameters such as water cut and flow rate. To obtain multi-dimensional fluid parameters, multiple independent measurement devices need to be deployed at the wellhead, resulting in cluttered equipment layout, low wellhead space utilization, and potential signal interference and data incompatibility between different devices. Some multi-parameter measurement devices suffer from narrow measurement range and insufficient accuracy. Due to the complex flow patterns and large fluctuations in phase content of the three-phase oil-gas-water flow, sensors are prone to insensitive responses and large measurement deviations under complex conditions such as high gas content and high water content, failing to meet the measurement needs of oil wells in different production states.

[0004] Furthermore, existing measurement devices often employ sensors based on a single detection principle, resulting in poor adaptability to three-phase flows of oil, gas, and water. Additionally, the housing materials of some sensors lack sufficient temperature, pressure, and corrosion resistance, making them prone to damage in the harsh working environment of oilfield wellheads. This leads to short device lifespans and high maintenance costs. Simultaneously, the control and signal transmission mechanisms of existing measurement systems are poorly designed, often relying on wired connections. This results in complex wiring, susceptibility to environmental interference, weak remote control capabilities, and an inability to achieve intelligent and centralized management of wellhead measurement devices. Moreover, the data processing algorithms are relatively simple, failing to effectively fuse signals from multiple sensors, thus limiting the accuracy of the final output parameters. At the measurement methodology level, existing methods for calculating water cut, gas cut, and flow rate are mostly based on single sensor data, neglecting the complementarity of data from different detection principles. Under conditions of changing flow patterns and abrupt changes in phase content, this can easily lead to distorted calculation results, hindering accurate characterization of parameters in three-phase flows of oil, gas, and water.

[0005] Currently, in the field of multi-parameter measurement of three-phase flow of oil, gas and water at oilfield wellheads, there is an urgent need for an integrated, intelligent, and highly adaptable measurement device, along with a stable and reliable control system and a precise and efficient measurement method. Summary of the Invention

[0006] The purpose of this invention is to provide a device, control system, and measurement method for multi-parameter measurement at oilfield wellheads, solving the problems of outdated traditional measurement methods, low measurement accuracy of existing devices, limited functionality, high maintenance costs, poor compatibility of control systems, and imperfect measurement methods. This invention enables real-time, accurate, and online measurement of multiple parameters of the three-phase flow of oil, gas, and water at oilfield wellheads, providing data support for intelligent oilfield production.

[0007] To achieve the above objectives, the present invention provides a device for multi-parameter measurement at oilfield wellheads. The device includes a gas-liquid separation subsystem, a control valve group, a dynamic measurement subsystem, an cumulative conductivity fiber optic measurement subsystem, a horizontal water cut flow fusion measurement subsystem, a gas volume measurement subsystem, and a turbine flow measurement subsystem. The gas-liquid separation subsystem includes a gas-liquid separation tank and a baffle plate. The gas-liquid separation tank is composed of a cylindrical tank and an eccentric conical tank connected together. The upper surface of the cylindrical tank is connected to the exhaust pipe, the left side of the cylindrical tank is connected to the inlet pipe, and the right side of the cylindrical tank is connected to the dynamic impedance measurement sensor. The control valve group includes a No. 1 solenoid valve and a No. 2 solenoid valve. The No. 1 solenoid valve is installed on the main transmission pipeline, and the No. 2 solenoid valve is installed on the inlet pipeline. The dynamic measurement subsystem includes a total water impedance measurement sensor and a mixed phase impedance measurement sensor. The total water impedance measurement sensor is located at the bottom of the cylindrical tank, one end of the mixed phase impedance measurement sensor is connected to the end of the eccentric conical tank, and the other end of the mixed phase impedance measurement sensor is connected to the flow-blocking pipe. The cumulative conductivity fiber optic measurement subsystem includes an integrated fiber optic conductivity probe sensor, which is installed parallel to the front and rear of the cylindrical tank wall. The horizontal moisture content flow fusion measurement subsystem includes a flow-collecting horizontal transmission pipe, a combined capacitive sensor, and a circumferential conductivity probe sensor. One end of the flow-collecting horizontal transmission pipe is connected to a flow-blocking pipe, and the other end of the flow-collecting horizontal transmission pipe is connected to an exhaust pipe. A second check valve is placed between the exhaust pipe and the flow-collecting horizontal transmission pipe. The gas volume measurement subsystem includes a float flow meter, a first check valve, and a second check valve. The gas volume measurement subsystem is located inside the exhaust pipe. The first check valve is located at the front end of the float flow meter, and the second check valve is located at the rear end of the float flow meter. The turbine flow measurement subsystem includes a turbine sensor, which is installed inside the inlet pipe and connected to the main transmission pipe. A second solenoid valve is installed in front of the turbine sensor.

[0008] Preferably, the central axis of the full water impedance measurement sensor is parallel to the bottom surface of the cylindrical tank, and the mixed phase impedance measurement sensor is perpendicular to the bottom surface of the cylindrical tank.

[0009] Preferably, the combined capacitive sensor, circumferential conductivity probe sensor and the outer diameter of the current-collecting horizontal transmission pipe are the same.

[0010] A control system for multi-parameter measurement at oilfield wellheads, comprising a circuit subsystem, a control subsystem, a signal acquisition and transmission subsystem, and a main controller module. The circuit subsystem is connected to the gas-liquid separation subsystem, control valve group, dynamic measurement subsystem, cumulative conductivity fiber optic measurement subsystem, horizontal water cut flow fusion measurement subsystem, gas volume measurement subsystem and turbine flow measurement subsystem, providing a power source for the multi-parameter measurement device at the oilfield wellhead; The control subsystem includes base stations and DL-LN3X series wireless self-organizing network modules; The signal acquisition and transmission subsystem includes an excitation module and a signal processing module; The main controller module incorporates methods for measuring moisture content, gas content, and flow rate.

[0011] Preferably, the circuit subsystem includes: The capacitor excitation module and capacitor signal processing module are connected to the combined capacitor sensor; The impedance excitation module and impedance signal processing module are connected to the full-water impedance measurement sensor and the mixed-phase impedance sensor; The turbine excitation module and turbine signal processing module are connected to the turbine sensor; The fiber optic conductivity excitation module and the fiber optic conductivity signal processing module are connected to the integrated fiber optic conductivity probe sensor. The float flow excitation module and the float flow signal processing module are connected to the float flowmeter; The circumferential conductivity excitation signal and circumferential conductivity signal processing module are connected to the circumferential conductivity probe sensor; The power module is used for power supply; Channel programmable switches and timing modules are used to control the start and stop of the measurement subsystem; ZigBee wireless self-organizing network module, used for communication.

[0012] A method for multi-parameter measurement at oilfield wellheads, comprising a water cut measurement method, a flow rate measurement method, and a gas cut measurement method; The method for measuring moisture content includes the following steps: S11. The multi-task prediction model framework combines a frequency domain module, a hybrid attention module, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, the hybrid attention module, and the convolutional layers effectively achieves feature extraction from the multiphase flow signal and uses this as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the water content y of the multiphase flow model. mw ; S12. Obtain the measurement signal from the dynamic measurement subsystem and calculate the impedance water content y using the Maxwell equation. zw ; S13. Obtain the measurement signal from the cumulative conductivity fiber optic measurement subsystem, and calculate the fiber conductivity and water content y based on the horizontal layering principle and threshold signal processing method. ozw ; S14. Acquire the measurement signal from the horizontal moisture content and flow rate fusion measurement subsystem, and calculate the capacitive moisture content y based on the capacitance measurement principle and impedance measurement principle. cw and circumferential electrical conductivity and water content y czw ; The comprehensive moisture content information Y is obtained by combining the moisture content measurements of S15, S11-S14, and S15. w =(y mw +y zw +y ozw +y cw +y czw ) / 5; The flow measurement method includes the following steps: S21. The multi-task prediction model framework combines frequency domain modules, hybrid attention modules, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, hybrid attention module, and convolutional layers effectively achieves feature extraction of the multiphase flow signal and uses it as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the multiphase flow model flow rate Q. m ; S22. Acquire the measurement signal from the cumulative conductivity fiber optic measurement subsystem, obtain the oil / water / gas fluid velocity information based on the cross-correlation measurement principle, and then calculate the oil / water / gas fluid flow rate information Q. ozg Q ozg Q ozg ; S23. Obtain the measurement signal from the dynamic measurement subsystem and use the cross-correlation algorithm to obtain the oil-water two-phase flow rate Q. zow ; S24. Obtain the measurement signal from the horizontal water cut flow fusion measurement subsystem and use the cross-correlation algorithm to obtain the circumferential oil-water two-phase flow information Q. czow ; S25. Acquire the measurement signal from the gas volume measurement subsystem to obtain the gas phase flow information Q from the float flowmeter. fg ; S26. Obtain the measurement signal from the turbine flow measurement subsystem to obtain the oil-gas-water three-phase flow information Q. togs ; S27. By integrating the flow information from S21 to S26 above, the phase flow information Q = (Q m +(Q ozg +Q ozg +Q ozg )+(Q czow +Q zow ) / (1-Y g )+(Q fg ) / (Y g )+Q togs ) / 5; Methods for measuring gas content include: S31. The multi-task prediction model framework combines a frequency domain module, a hybrid attention module, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, the hybrid attention module, and the convolutional layers effectively extracts features from the multiphase flow signal and uses these features as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the gas content y of the multiphase flow model. mg ; S32. Obtain the measurement signal from the gas volume measurement subsystem to obtain the gas phase flow information Q from the float flowmeter. fg ; S33. Obtain the measurement signal from the cumulative conductivity fiber optic measurement subsystem, and calculate the fiber conductivity and gas content y based on the horizontal layering principle and threshold signal processing method. ozg ; S34. Obtain the turbine flow measurement subsystem signal to get the oil-gas-water three-phase flow information Q. tgwo ; S35. Combining the gas phase flow rate information and the three-phase flow rate information of oil, gas and water, the gas content information Y is calculated. g =(y ozg +(Q fg ) / (Q tgwo )+y mg ) / 3.

[0013] Therefore, the present invention employs the above-mentioned device, control system, and measurement method for multi-parameter measurement at oilfield wellheads, and the technical effects are as follows: 1. Comprehensive improvement in measurement performance, adapting to complex oilfield conditions: The device integrates a gas-liquid separation subsystem and multiple types of sensor subsystems. The measurement methods combine various technologies such as impedance, fiber optic conductivity, capacitance, and cross-correlation algorithms. Furthermore, relying on Maxwell's equations, inversion of deep models to accurately calculate phase content, and fusion of phase flow based on multi-source flow data, it realizes integrated measurement of multiple parameters such as water content, gas content, and phase flow in three-phase flow of oil, gas, and water, breaking through the problems of limited measurement range and insufficient accuracy of single sensors.

[0014] 2. Real-time online measurement: Through the wireless self-organizing network module of the control system, it can link with remote computers and base stations to collect wellhead fluid parameters in real time and complete the calculation, visualization display and storage. This replaces the offline mode of "wellhead sampling-transportation-central laboratory testing" in traditional production, and accurately and in real time reflects the changes in single-well fluid parameters.

[0015] 3. Improve the accuracy of parameter calculation: The fusion strategy of multi-source data effectively offsets the error of a single measurement method, further improving the accuracy and stability of fluid parameter measurement results; at the same time, the control system equips each sensor with a dedicated excitation and signal processing module, ensuring the acquisition accuracy and transmission reliability of measurement signals. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of a multi-parameter measuring device for oilfield wellheads according to an embodiment of the present invention; Figure 2 This is a block diagram of a circuit subsystem of a multi-parameter measurement and control system for oilfield wellheads, according to an embodiment of the present invention. Figure 3 This is a block diagram of the control subsystem of a multi-parameter control and measurement system for oilfield wellheads, according to an embodiment of the present invention. Figure 4 This is a block diagram of a multi-parameter measurement method for oilfields according to an embodiment of the present invention; Figure 5 This is a schematic diagram of a turbine sensor structure according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a full-water (mixed-phase) impedance measurement sensor according to an embodiment of the present invention; Figure 7 This is a schematic diagram of a combined capacitive sensor structure according to an embodiment of the present invention; Figure 8 This is a schematic diagram of a circumferential conductivity probe sensing structure according to an embodiment of the present invention; Figure 9 This is a schematic diagram of an integrated optical fiber conductivity probe sensor structure according to an embodiment of the present invention.

[0017] Figure Labels 1. Main transmission pipeline; 2. Solenoid valve No. 1; 3. Solenoid valve No. 2; 4. Inlet pipeline; 5. Turbine sensor; 6. Gas-liquid separator; 7. Columnar tank; 8. Eccentric conical tank; 9. Baffle plate; 10. Fiber optic conductivity integrated sensor; 11. Total water impedance measurement sensor; 12. Exhaust pipeline; 13. Check valve No. 1; 14. Float flowmeter; 15. Check valve No. 2; 16. Mixed phase impedance measurement sensor; 17. Flow obstruction pipeline; 18. Collector 19. Horizontal transmission pipeline; 20. Combined capacitive sensor; 21. Circumferential conductivity probe sensor; 22. No. 3 check valve; 23. Oilfield wellhead multi-parameter measurement and control system; 24. Computer; 501. Front guide hub; 502. Front guide blade; 503. Impeller blade; 504. Shaft; 505. Magnetoelectric induction converter; 506. Rear guide blade; 507. Rear guide hub; 1004. Fiber optic probe; 1003. Excitation current. Electrode; 1002, Insulating tube; 1001, Excitation ground electrode; 1101, Excitation electrode ring 1; 1106, Excitation electrode ring 2; 1102, Measurement electrode ring 1; 1103, Measurement electrode ring 2; 1104, Measurement electrode ring 3; 1105, Measurement electrode ring 4; 1901, Metal casing; 1902, Outer insulating layer; 1903, Metal layer electrode; 1904, Inner insulating layer; 1905, Center electrode insulating layer; 1906, Center electrode rod; 2001, Excitation electrode 1; 2002, Excitation electrode 2; 2003, Excitation electrode 3; 2004, Excitation electrode 4; 2005, Excitation electrode 5; 2006, Excitation electrode 6; 2007, Measurement electrode 1; 2008, Measurement electrode 2; 2009, Measurement electrode 3; 2010, Measurement electrode 4; 2011, Measurement electrode 5; 2012, Measurement electrode 6. Detailed Implementation

[0018] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0020] Example 1 like Figure 1 As shown, the present invention provides a device for multi-parameter measurement at the wellhead of an oilfield. The device includes a gas-liquid separation subsystem, a control valve group, a dynamic measurement subsystem, an cumulative conductivity fiber optic measurement subsystem, a horizontal water cut flow fusion measurement subsystem, a gas volume measurement subsystem, and a turbine flow measurement subsystem. It is mainly used for gas-liquid separation and multi-parameter measurement of three-phase flow of oil, gas, and water at the wellhead of an oilfield.

[0021] The gas-liquid separation subsystem includes a gas-liquid separation tank 6 and a baffle plate 9. The gas-liquid separation tank 6 is composed of a cylindrical tank 7 and an eccentric conical tank 8 connected together. The central axis of the eccentric conical tank 8 forms an angle of α degrees with the horizon. The upper surface of the cylindrical tank 7 is connected to the exhaust pipe 12. The left side of the cylindrical tank 7 is connected to the inlet pipe 4, and the right side of the cylindrical tank 7 is connected to the dynamic impedance measurement sensor.

[0022] The dynamic measurement subsystem includes a total water impedance measurement sensor 11 and a mixed-phase impedance measurement sensor 16. The total water impedance measurement sensor 11 is located at the bottom of the cylindrical tank 7, and one end of the mixed-phase impedance measurement sensor 16 is connected to the end of the eccentric conical tank 8, while the other end of the mixed-phase impedance measurement sensor 16 is connected to the flow-blocking pipe 17. The central axis of the total water impedance measurement sensor 11 is parallel to the bottom surface of the cylindrical tank 7, and the mixed-phase impedance measurement sensor 16 is perpendicular to the bottom surface of the cylindrical tank 7.

[0023] The cumulative conductivity fiber optic measurement subsystem consists of n (n>1, n∈N) The system consists of n groups of integrated fiber optic conductivity probe sensors, which are arranged in a parallel front-to-back configuration, where n (n>1, n∈N) The fiber optic conductivity integrated probe sensor is located on the wall of the cylindrical tank 7. The housing of the fiber optic conductivity integrated probe sensor is made of temperature-resistant, pressure-resistant and corrosion-resistant ceramic.

[0024] The horizontal moisture content and flow rate fusion measurement subsystem is located inside the horizontal collection pipeline 18, with one end of the horizontal collection pipeline 18 connected to the flow-blocking pipeline 17 and the other end connected to the exhaust pipeline 12. A second check valve 15 is placed between the exhaust pipeline 12 and the horizontal collection pipeline 18. The combined capacitance sensor 19 and the circumferential conductivity probe sensor 20 have the same outer diameter as the horizontal collection pipeline 18. The two ends of the circumferential conductivity probe sensor 20 are the combined capacitance sensor 19 and the third check valve 21, respectively.

[0025] The gas volume measurement subsystem mainly consists of a float flowmeter 14, a first check valve 13, and a second check valve 15, all located in the exhaust pipe 12. The first check valve 13 is positioned before the float flowmeter 14, and the second check valve 15 is positioned after the float flowmeter 14 to prevent fluid backflow.

[0026] The turbine flow measurement subsystem consists of a turbine flow meter and is located inside the inlet pipe 4. The main transmission pipe 1 is connected to the inlet pipe 4. Solenoid valve 2 is located on the main transmission pipe 1, and solenoid valve 3 is located on the inlet pipe 4 and before the turbine flow meter.

[0027] Solenoid valve 2 (number one) is installed on the main transmission pipeline 1, and solenoid valve 3 (number two) is installed on the inlet pipeline 4. Inlet pipeline 4 connects to the cylindrical tank 7. A total water resistance measurement sensor 11 is installed at the bottom of the cylindrical tank 7, with its central axis parallel to the central axis of the cylindrical tank 7. It is used to obtain the total water resistance value of the fluid. (n>1, n∈N) A fiber optic conductivity integrated probe sensor is vertically mounted on the side surface of the cylindrical tank 7, and the exhaust pipe 12 is connected to the upper surface of the cylindrical tank 7. A float flowmeter 14 is mounted on the exhaust pipe 12; a mixed-phase impedance measurement sensor 16 is connected to the end of the eccentric conical tank 8 and is perpendicular to the central axis of the cylindrical tank 7, used to measure the mixed-phase value of the oil-water two-phase flow after gas-liquid separation. A combined capacitance sensor 19 and a circumferential conductivity probe sensor 20 are installed inside the flow-collecting horizontal transmission pipe 18. A turbine sensor 5 is installed inside the inlet pipe 4.

[0028] like Figure 2 As shown, the multi-parameter measurement and control system 22 at the oilfield wellhead mainly includes a circuit subsystem, a base station, a computer 23, a DL-LN3X series wireless self-organizing network module, and the circuit subsystem is connected to solenoid valve 2 (number one), solenoid valve 3 (number two), a full water impedance measurement sensor 11, a mixed-phase impedance measurement sensor 16, and n (n>1, n∈N). The system includes a fiber optic conductivity integrated probe sensor, a combined capacitance sensor 19, a circumferential conductivity probe sensor 20, a turbine sensor 5, a float flowmeter 14, and a DL-LN3X series wireless self-organizing network module. The control subsystem includes a base station and the DL-LN3X series wireless self-organizing network module. Remote control of these modules is achieved through communication with the base station, computer 23, and circuit subsystem. The water cut measurement method, gas cut measurement method, and flow rate measurement method primarily obtain wellhead parameter information by processing the signal acquisition sensor data.

[0029] The circuit subsystem includes: The capacitor excitation module is connected to the gas-liquid separation subsystem, control valve group, dynamic measurement subsystem, cumulative conductivity fiber optic measurement subsystem, horizontal water cut flow fusion measurement subsystem, gas volume measurement subsystem and turbine flow measurement subsystem, providing a power source for the multi-parameter measurement device at the oilfield wellhead; The capacitance signal processing module acquires and processes the capacitance signal, and outputs a frequency signal that reflects the dielectric constant of the fluid. The impedance excitation module is connected to the full water impedance measurement sensor 11 and the mixed-phase impedance sensor, and is used to generate an excitation current for them. An impedance signal processing module, connected to the full water impedance measurement sensor 11 and the mixed-phase impedance sensor, is used to process the measurement frequency signal; The turbine excitation module is connected to the turbine sensor 5 and is used to generate an excitation signal for the turbine sensor 5. The turbine signal processing module is connected to the turbine sensor 5 and processes the frequency signal generated by the turbine sensor 5. The fiber optic conductance excitation module, and n (n>1, n∈N) A group of fiber optic conductivity integrated sensors 10 are connected to provide fiber optic excitation voltage signals and conductivity excitation voltage signals; The fiber optic conductance signal processing module, with n (n>1, n∈N) A set of fiber optic conductivity integrated sensors (10 units) are connected to process fiber optic voltage signals and conductivity voltage signals. The float flow excitation module is connected to the float flowmeter 14 and provides voltage excitation to it; The float flow signal processing module is connected to the float flow meter 14 and is used to process the float flow signal. The circumferential conductivity excitation signal is connected to the circumferential conductivity probe sensor 20 to excite it with current; The circumferential conductivity signal processing module is connected to the circumferential conductivity probe sensor 20 and processes its measured frequency signal; The power module supplies power to the multi-parameter measurement device at the oilfield wellhead; The main controller module controls the start and stop of the measurement subsystem, and has embedded water cut measurement method, gas cut measurement method and flow rate measurement method to detect wellhead fluid parameters; The channel programmable switch establishes a channel flag position for the core sensor in the multi-parameter measurement device at the oilfield wellhead, thereby identifying the sensor's activation and deactivation. The timing module is connected to the main control module, enabling the main control module to switch the working state of the measurement subsystem at regular intervals. The ZigBee wireless self-organizing network module is connected to the main control module and enables remote control of the module by communicating with the base station, computer 23 and circuit subsystem.

[0030] like Figure 3 As shown, the control method of this control system mainly includes the following processes: (1) The computer 23 sends a signal acquisition command to the wireless base station; (2) The wireless base station sends the signal acquisition command to the oilfield wellhead multi-parameter measurement and control system 22; (3) After receiving the signal acquisition instruction, the oilfield wellhead multi-parameter measurement and control system 22 realizes the flow of fluid inside the oilfield wellhead multi-parameter measurement device by closing the first solenoid valve 2 and opening the second solenoid valve 3, and acquires the measurement signals of the dynamic measurement subsystem, the cumulative conductivity fiber optic measurement subsystem, the horizontal water cut flow fusion measurement subsystem, the gas volume measurement subsystem and the turbine flow measurement subsystem, and sends the signals to the wireless base station. (4) The wireless base station sends the measurement signal to the computer 23; (5) The computer 23 receives the measurement signal, calculates parameters such as phase content and flow rate of the signal, and performs visualization display and storage. (6) Repeat operations (1) to (5); (7) The computer 23 sends a termination signal acquisition command to the wireless base station; (8) The wireless base station sends the termination signal acquisition command to the oilfield wellhead multi-parameter measurement and control system 22; (9) After receiving the instruction to terminate signal acquisition, the oilfield wellhead multi-parameter measurement and control system 22 closes the second solenoid valve 3 and opens the first solenoid valve 2, so that the fluid is transmitted through the main transmission pipeline.

[0031] like Figure 4 As shown, a multi-parameter measurement method for oilfield wellheads is presented, which includes a water cut measurement method, a flow rate measurement method, and a gas cut measurement method.

[0032] The multi-task measurement method mainly includes a multi-task prediction model framework. In the feature extraction part, it uses a hybrid modeling approach combining frequency domain layers, a hybrid attention module, and convolutional layers. The frequency domain layer contains a Fast Fourier Transform layer, a weighted gating layer, and an Inverse Fourier Transform layer. This design allows for a complex-scale examination of the input data, capturing different frequency components to extract local frequency domain features. The hybrid attention layer employs a window-based multi-head self-attention module and a channel attention module inserted in parallel into a standard Transformer block, significantly enhancing the network's representational capabilities. Furthermore, it utilizes one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. This hybrid connection of the frequency domain layer, the hybrid attention module, and the convolutional layers effectively achieves feature extraction from oil, gas, and water flow signals. Because the tasks of water cut prediction and total flow velocity prediction are coupled, the difficulty of measuring flow parameters is greatly increased. Therefore, in the decoupling and output sections, these three related tasks are decoupled, so that each task has its own unique linear mapping layer. Using the same backbone feature extraction structure and a unique decoding head, combined with different activation functions, water cut prediction, gas cut prediction, and total flow prediction for oil, gas, and water flow are achieved. The specific implementation process of the multi-task prediction model is as follows: Preprocessed oil, gas, and water flow data is used as the model input. After positional encoding through absolute embedding and a one-dimensional convolution operation, feature 1 is obtained. Feature 1 is passed through a frequency domain layer to extract local frequency domain features from the data, resulting in feature 2. Feature 2 undergoes two hybrid attention modules for more comprehensive feature extraction, resulting in feature 3. Feature 3 is then passed through a 29×29 convolutional layer and a 3×3 convolutional layer, followed by a 5×5 average pooling layer, and then a dropout layer to obtain feature 4. Feature 4 is processed through a residual block, a 5×5 average pooling layer, and a Dropout layer to obtain Feature 5. The residual block includes a 29×29 convolutional layer. Finally, Feature 5 is flattened into one dimension by a Flatten layer. To achieve multi-task output, in the decoder and output parts, the data passes through a Dropout layer and then enters a linear layer. Each task has its own unique linear mapping layer with activation functions set to Sigmoid, Sigmoid, and Softmax, respectively, to achieve water content prediction and total flow velocity prediction tasks. Finally, the final prediction result is output based on the probability values.

[0033] The frequency domain module introduces a frequency-domain gated network, which consists of a Fast Fourier Transform (FFT) layer, a weighted gating layer, and an Inverse Fourier Transform (IFFT) layer. First, the FFT layer transforms the physical space of the data into the frequency domain to capture relevant features from the initial layers of the architecture. Then, a gating layer with learnable weight parameters determines the weights of each frequency component, thereby better capturing refined features in the data. The gating layer allows the network to update parameters through backpropagation. Subsequently, the IFFT layer restores the frequency domain space back to the physical space. Following the IFFT layer, the frequency domain layer includes a normalization layer for channel mixing and a Multilayer Perceptron (MLP), while token fusion is accomplished using frequency-domain gating techniques.

[0034] The hybrid attention module adds a channel attention-based convolutional block to the standard Transformer block to enhance the network's expressive power. After the first normalization (LayerNorm, LN) layer, the Channel Attention Block (CAB) is inserted into the standard Swin Transformer block and connected in parallel with the window-based multi-head self-attention (Window Attention, WA) module. The CAB consists of two standard convolutional layers, a GELU activation function, and a channel attention.

[0035] The method for measuring moisture content includes the following steps: The multi-task prediction model framework combines frequency domain modules, hybrid attention modules, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, hybrid attention module, and convolutional layers effectively extracts features from the multiphase flow signal and uses these features as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the water content y of the multiphase flow model. mw ; Acquire the measurement signal from the dynamic measurement subsystem and calculate the impedance water content using the Maxwell equation; Under conditions where water is the continuous phase, the voltage amplitude of the miscibility impedance measurement sensor 16 is inversely proportional to the conductivity of the fluid passing through the miscibility impedance measurement sensor 16. Let the conductivity of the measuring electrode rings (1102~1105) in the oil-water miscibility phase be... When the water is full The conductivity of the mixed phase is The electrical conductivity of water is When the phases are mixed, the sensor output frequency is (Mixability value), Total water value (Total water value), then: ; and The ratio is given by Maxwell's formula: ; In the formula, The volume fraction of the continuous conductive phase in a two-phase flow is called the impedance water-holding capacity, which is equivalent to the impedance water-holding capacity in an oil-water two-phase flow. Water holdup refers to the volume percentage of water phase at a certain point in the wellbore. The ratio of total water value to miscibility value is called the instrument relative response. The miscibility value is measured when the oil-water two-phase fluid flows through the miscibility impedance measurement sensor 16, and the total water value can be obtained through the total water impedance measurement sensor 11.

[0036] The measurement signals of the cumulative conductivity fiber optic measurement subsystem are acquired, and the fiber conductivity and water content measurement information are calculated based on the horizontal layering principle and the threshold signal processing method.

[0037] After the three-phase flow of oil, gas, and water in gas-liquid separator 6 undergoes natural stratification, the integrated fiber optic conductivity probe sensor detects the gas phase medium, outputting a high-level voltage signal from both the fiber optic cable and the conductivity sensor. When the integrated fiber optic conductivity probe sensor detects the water phase medium, it outputs a low-level voltage signal from both the fiber optic cable and the conductivity sensor. When the integrated fiber optic conductivity probe sensor detects the oil phase medium, it outputs a low-level voltage signal from both the fiber optic cable and the conductivity sensor. Based on the different responses of the oil, gas, and water phases, the fluid phase content is calculated.

[0038] The measurement signals of the horizontal moisture content and flow rate fusion measurement subsystem are acquired, and the capacitive moisture content and circumferential conductivity moisture content are calculated based on the principles of capacitance measurement and impedance measurement.

[0039] When the combined capacitive sensor 19 is placed in a fully oil-phase environment, the output frequency is: When the combined capacitive sensor 19 is placed in a fully aqueous environment, the output frequency is: When the combined capacitive sensor 19 is placed in the oil-water two-phase fluid to be measured, the output frequency is: : ; Determining the capacitive water holding capacity by fluid capacitance: .

[0040] The principle of circumferential conductivity water content measurement is that, under the condition that water is a continuous phase, the voltage amplitude of the measuring electrode (2007~2012) is inversely proportional to the conductivity of the fluid passing through the circumferential conductivity probe sensor 20. Let the conductivity of the measuring electrode (2007~2012) in the oil-water miscible phase be... When the water is full The conductivity of the mixed phase is The electrical conductivity of water is When the phases are mixed, the sensor output frequency is (Mixability value), Total water value (Total water value), then: ; and The ratio is given by Maxwell's formula: ; In the formula, The volume fraction of the continuous conductive phase in a two-phase flow is called the impedance water-holding capacity, which is equivalent to the impedance water-holding capacity in an oil-water two-phase flow. Water holdup refers to the volume percentage of water phase at a certain point in the wellbore. The ratio of total water value to miscibility value is called the instrument's relative response. The miscibility value is measured when oil and water two-phase fluids flow through the sensor, while the total water value can be obtained under static stratification conditions.

[0041] By combining the measured values ​​of impedance water content, fiber conductivity water content, capacitor water content, and circumferential conductivity water content, a comprehensive water content information is obtained.

[0042] The flow measurement method includes the following steps: The multi-task prediction model framework combines frequency domain modules, hybrid attention modules, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, hybrid attention module, and convolutional layers effectively extracts features from the multiphase flow signal and uses these features as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the multiphase flow model flow rate Q. m ; The measurement signal from the cumulative conductivity fiber optic measurement subsystem is acquired, and the flow velocity information of oil / water / gas fluids is obtained based on the cross-correlation measurement principle. Then, based on the basic information of the gas-liquid separator 6, the flow rate information Q of oil / water / gas fluids is calculated. ozg Q ozg Q ozg ; The measurement signal from the dynamic measurement subsystem is acquired, and the flow rate Q of the oil-water two-phase flow is obtained using a cross-correlation algorithm. zow ; The measurement signal from the horizontal water cut flow fusion measurement subsystem is acquired, and the circumferential oil-water two-phase flow information Q is obtained using a cross-correlation algorithm. czow ; For the all-water (mixable) impedance measurement sensor, measuring electrode ring 1102 and measuring electrode ring 1103 are the upstream detection electrode rings, and measuring electrode ring 3104 and measuring electrode ring 1105 are the downstream detection electrode rings. The time delay generated between the upstream and downstream detection electrode rings can be obtained by the cross-correlation function R of their output signals x(t) and y(t). xy (τ) is obtained as follows: ; There is a certain similarity between x(t) and y(t): ; In the formula, z(t) represents the system noise superimposed on the measurement signal that is independent of the fluid flow. Substituting it into the cross-correlation function, we get: ; Since the measured signal x(t) is independent of the system noise z(t), their cross-correlation function is zero. Therefore, the above equation can be rewritten as: ; In the above equation, the right side of the equation represents the autocorrelation function of the downstream signal. From the properties of the autocorrelation function, we know that when hour ) reaches its maximum value, while R xy (τ) also reached its maximum value, which is the time value corresponding to the peak of the cross-correlation function of x(t) and y(t). It refers to the transit time of the fluid flowing from the upstream detection electrode ring to the downstream detection electrode ring.

[0043] Performing cross-correlation calculations on the measured signals will allow us to obtain the transit time of the fluid from the upstream detection electrode ring to the downstream detection electrode ring. This leads to the relevant flow rate: ; In the formula, U cc Where L is the fluid velocity, and L is the distance between the upstream and downstream detection electrode rings. This is the transit time of the fluid from the upstream detection electrode ring to the downstream detection electrode ring.

[0044] The gas flow rate information Q of the float flowmeter is obtained by acquiring the measurement signal from the gas flow measurement subsystem. fg ; The turbine flow measurement subsystem is used to acquire the measurement signal, and the flow rate information Q of the three-phase oil-gas-water flow is obtained. togs ; The frequency of the signal output by turbine sensor 5 within a certain flow rate range and a certain fluid viscosity range. Volumetric flow rate through the turbine flow meter Proportional, that is: ; In the formula, - Instrument coefficient of turbine sensor 5 ( or Fluid velocity Based on frequency signal Similarly, we can obtain the result.

[0045] By integrating cumulative oil / water / gas flow information, dynamic oil-water two-phase flow information, circumferential oil-water two-phase flow information, float gas phase flow information, and turbine flow information, phase-separated flow information is obtained.

[0046] The specific steps for measuring gas content are as follows: The multi-task prediction model framework combines frequency domain modules, hybrid attention modules, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, hybrid attention module, and convolutional layers effectively extracts features from the multiphase flow signal and uses these features as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the gas content y of the multiphase flow model. mg ; The gas flow rate information Q of the float flowmeter is obtained by acquiring the measurement signal from the gas flow measurement subsystem. fg ; The measurement signal from the cumulative conductivity fiber optic measurement subsystem is acquired, and the fiber conductivity gas content y is calculated based on the horizontal layering principle and threshold signal processing method. ozg ; The turbine flow measurement subsystem signal is acquired to obtain the three-phase flow information Q of oil, gas and water. tgwo ; By combining gas phase flow rate information and oil-gas-water three-phase flow rate information, the gas content information Y is calculated. g =(y ozg +(Q fg ) / (Q tgwo )+y mg ) / 3.

[0047] like Figure 5As shown, the turbine sensor 5 includes a front guide vane hub 501, a front guide vane blade 502, an impeller blade 503, a rotating shaft 504, a magnetoelectric induction converter 505, a rear guide vane 506, and a rear guide vane hub 507. The impeller blade 503 is mounted on the rotating shaft 504, one end of which is connected to the front guide vane hub 501, and the other end is connected to one end of the magnetoelectric induction converter 505. The other end of the magnetoelectric induction converter 505 is connected to the rear guide vane hub 507. The front guide vane hub 501 has the front guide vane blade 502 from the guide vane assembly; the rear guide vane hub 507 has the rear guide vane blade from the guide vane assembly. The impeller blade 503 is a rotatable component and has a strong magnet.

[0048] like Figure 6 The image shows a full-water (mixed-phase) impedance measurement sensor comprising an excitation electrode ring 1101, an excitation electrode ring 1106, a measurement electrode ring 1102, a measurement electrode ring 1103, a measurement electrode ring 1104, and a measurement electrode ring 1105. Electrode rings 1101-1106 are all embedded in plexiglass and are not connected to the horizontal transmission pipe 18. The full-water (mixed-phase) impedance measurement sensor is located inside the horizontal transmission pipe 18, and its inner diameter is the same as that of the horizontal transmission pipe 18.

[0049] like Figure 7 As shown, the combined capacitive sensor 19 includes a metal housing 1901, an outer insulating layer 1902, a metal layer electrode 1903, an inner insulating layer 1904, a central electrode insulating layer 1905, and a central electrode rod 1906. The combined capacitive sensor 19 consists of an electrode rod inserted inside a cylindrical capacitive sensor. The central electrode rod 1906 is located on the central axis of the combined capacitive sensor 19 and is connected in parallel with the metal layer electrode 1903, serving as an electrode together. The central electrode insulating layer 1905 is located outside the central electrode rod 1906. From the outside to the inside, the combined capacitive sensor 19 consists of the metal housing 1901, the outer insulating layer 1902, the metal layer electrode 1903, the inner insulating layer 1904, the central electrode insulating layer 1905, and the central electrode rod 1906, with a flow channel between the central electrode insulating layer 1905 and the inner insulating layer 1904.

[0050] like Figure 8As shown, the circumferential conductivity probe sensor 20 consists of a cylindrical insulating support and 12 probe electrodes fixed on the inner wall of the insulating support. The 12 probe electrodes include 6 excitation electrodes 1003 (2001~2006) and 6 measurement electrodes (2007~2012). The excitation electrodes 1003 include excitation electrode 1 2001, excitation electrode 2002, excitation electrode 3 2003, excitation electrode 4 2004, excitation electrode 5 2005, and excitation electrode 6 2006. The measurement electrodes include measurement electrode 1 2007, measurement electrode 2008, measurement electrode 3 2009, measurement electrode 4 2010, measurement electrode 5 2011, and measurement electrode 6 2012.

[0051] Six excitation electrodes 1003 (2001~2006) are located on the circumferential direction of the upstream axial section of the inner wall of the support and are evenly spaced. The remaining six measuring electrodes (2007~2012) are located on the circumferential direction of the downstream axial section of the inner wall of the support and are arranged in parallel to the upstream excitation electrodes 1003. The six excitation electrodes 1003 (2001~2006) and the six measuring electrodes (2007~2012) constitute six sets of dynamic total water value measurement sensors. Each set of dynamic total water value measurement sensors includes one excitation electrode 1003 and one measuring electrode.

[0052] like Figure 9 As shown, the integrated conductive fiber optic probe sensor includes a fiber optic probe 1004, an excitation electrode 1003, an insulating tube 1002, and an excitation ground electrode 1001. The excitation electrode 1003 is located outside the fiber optic probe 1004. The insulating tube 1002 separates the excitation electrode 1003 and the excitation ground electrode 1001 and is located outside the fiber optic probe 1004. The excitation ground electrode 1001 has a stainless steel shell for protection.

[0053] Therefore, this invention employs the aforementioned device, control system, and measurement method for multi-parameter measurement at oilfield wellheads. The device integrates seven subsystems, including gas-liquid separation, control valve groups, and dynamic measurement. It is equipped with various types of sensors, such as all-water impedance, fiber optic conductivity integrated probes, and turbine sensors. The control system comprises circuit, control, and signal acquisition and transmission subsystems, including various excitation / signal processing modules and wireless self-organizing network modules, enabling power supply, remote control, and signal transmission for the device. The measurement method utilizes Maxwell equations, cross-correlation algorithms, and multi-source data fusion techniques to accurately calculate water cut, gas cut, and flow rate. Furthermore, the water cut measurement integrates multiple measurements such as impedance and fiber optic conductivity, the flow rate integrates data from multiple subsystems, and the gas cut... This technical solution combines gas phase and three-phase total flow calculation to achieve automated, real-time online measurement of multiple parameters of oil, gas, and water three-phase flow at the wellhead. It solves the problems of high cost, poor real-time performance, and high manual labor intensity associated with traditional measurement methods. It features a wide measurement range, high accuracy, low cost, easy processing, and good maintainability. Key sensor components are made of temperature-, pressure-, and corrosion-resistant materials, making it suitable for harsh oilfield conditions and scattered field applications. It not only reflects changes in single-well fluid parameters in real time and promptly detects fluctuations in oil well production, but also reduces on-site personnel, material, and financial inputs, saving costs for oilfield production management. Furthermore, it provides key technical support for digital oilfield networking projects, facilitating integrated management of oilfield production command and overall optimization of the production process.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A device for multi-parameter measurement at oilfield wellheads, characterized in that, The device includes a gas-liquid separation subsystem, a control valve group, a dynamic measurement subsystem, an cumulative conductivity fiber optic measurement subsystem, a horizontal water content flow fusion measurement subsystem, a gas volume measurement subsystem, and a turbine flow measurement subsystem. The gas-liquid separation subsystem includes a gas-liquid separation tank and a baffle plate. The gas-liquid separation tank is composed of a cylindrical tank and an eccentric conical tank connected together. The upper surface of the cylindrical tank is connected to the exhaust pipe, the left side of the cylindrical tank is connected to the inlet pipe, and the right side of the cylindrical tank is connected to the dynamic impedance measurement sensor. The control valve group includes a No. 1 solenoid valve and a No. 2 solenoid valve. The No. 1 solenoid valve is installed on the main transmission pipeline, and the No. 2 solenoid valve is installed on the inlet pipeline. The dynamic measurement subsystem includes a total water impedance measurement sensor and a mixed phase impedance measurement sensor. The total water impedance measurement sensor is located at the bottom of the cylindrical tank, one end of the mixed phase impedance measurement sensor is connected to the end of the eccentric conical tank, and the other end of the mixed phase impedance measurement sensor is connected to the flow-blocking pipe. The cumulative conductivity fiber optic measurement subsystem includes an integrated fiber optic conductivity probe sensor, which is installed parallel to the front and rear of the cylindrical tank wall. The horizontal moisture content flow fusion measurement subsystem includes a flow-collecting horizontal transmission pipe, a combined capacitive sensor, and a circumferential conductivity probe sensor. One end of the flow-collecting horizontal transmission pipe is connected to a flow-blocking pipe, and the other end of the flow-collecting horizontal transmission pipe is connected to an exhaust pipe. A second check valve is placed between the exhaust pipe and the flow-collecting horizontal transmission pipe. The gas volume measurement subsystem includes a float flow meter, a first check valve, and a second check valve. The gas volume measurement subsystem is located inside the exhaust pipe. The first check valve is located at the front end of the float flow meter, and the second check valve is located at the rear end of the float flow meter. The turbine flow measurement subsystem includes a turbine sensor, which is installed inside the inlet pipe and connected to the main transmission pipe. A second solenoid valve is installed in front of the turbine sensor.

2. The device for multi-parameter measurement at oilfield wellheads according to claim 1, characterized in that, The central axis of the full water impedance measurement sensor is parallel to the bottom surface of the cylindrical tank, while the mixed phase impedance measurement sensor is perpendicular to the bottom surface of the cylindrical tank.

3. The device for multi-parameter measurement at oilfield wellheads according to claim 1, characterized in that, The combined capacitive sensor, circumferential conductivity probe sensor and the collector-type horizontal transmission pipe have the same outer diameter.

4. A control system for multi-parameter measurement at oilfield wellheads, used to control the device according to any one of claims 1-3 to perform measurements, characterized in that, The control system includes a circuit subsystem, a control subsystem, a signal acquisition and transmission subsystem, and a main controller module. The circuit subsystem is connected to the gas-liquid separation subsystem, control valve group, dynamic measurement subsystem, cumulative conductivity fiber optic measurement subsystem, horizontal water cut flow fusion measurement subsystem, gas volume measurement subsystem and turbine flow measurement subsystem, providing a power source for the multi-parameter measurement device at the oilfield wellhead; The control subsystem includes base stations and DL-LN3X series wireless self-organizing network modules; The signal acquisition and transmission subsystem includes an excitation module and a signal processing module; The main controller module incorporates methods for measuring moisture content, gas content, and flow rate.

5. The control system for multi-parameter measurement at the oilfield wellhead according to claim 4, characterized in that, The circuit subsystem includes: The capacitor excitation module and capacitor signal processing module are connected to the combined capacitor sensor; The impedance excitation module and impedance signal processing module are connected to the full-water impedance measurement sensor and the mixed-phase impedance sensor; The turbine excitation module and turbine signal processing module are connected to the turbine sensor; The fiber optic conductivity excitation module and the fiber optic conductivity signal processing module are connected to the integrated fiber optic conductivity probe sensor. The float flow excitation module and the float flow signal processing module are connected to the float flowmeter; The circumferential conductivity excitation signal and circumferential conductivity signal processing module are connected to the circumferential conductivity probe sensor; The power module is used for power supply; Channel programmable switches and timing modules are used to control the start and stop of the measurement subsystem; ZigBee wireless self-organizing network module, used for communication.

6. A method for multi-parameter measurement at oilfield wellheads, comprising applying the apparatus described in any one of claims 1-3 to the measurement, characterized in that, The measurement methods include moisture content measurement methods, flow rate measurement methods, and gas content measurement methods; The method for measuring moisture content includes the following steps: S11. The multi-task prediction model framework combines a frequency domain module, a hybrid attention module, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, the hybrid attention module, and the convolutional layers effectively achieves feature extraction from the multiphase flow signal and uses this as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the water content y of the multiphase flow model. mw ; S12. Obtain the measurement signal from the dynamic measurement subsystem and calculate the impedance water content y using the Maxwell equation. zw ; S13. Obtain the measurement signal from the cumulative conductivity fiber optic measurement subsystem, and calculate the fiber conductivity and water content y based on the horizontal layering principle and threshold signal processing method. ozw ; S14. Acquire the measurement signal from the horizontal moisture content and flow rate fusion measurement subsystem, and calculate the capacitive moisture content y based on the capacitance measurement principle and impedance measurement principle. cw and circumferential electrical conductivity and water content y czw ; The comprehensive moisture content information Y is obtained by combining the moisture content measurements of S15, S11-S14, and S15. w =(y mw +y zw +y ozw +y cw +y czw ) / 5; The flow measurement method includes the following steps: S21. The multi-task prediction model framework combines frequency domain modules, hybrid attention modules, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, hybrid attention module, and convolutional layers effectively achieves feature extraction of the multiphase flow signal and uses it as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the multiphase flow model flow rate Q. m ; S22. Acquire the measurement signal from the cumulative conductivity fiber optic measurement subsystem, obtain the oil / water / gas fluid velocity information based on the cross-correlation measurement principle, and then calculate the oil / water / gas fluid flow rate information Q. ozg Q ozg Q ozg ; S23. Obtain the measurement signal from the dynamic measurement subsystem and use the cross-correlation algorithm to obtain the oil-water two-phase flow rate Q. zow ; S24. Obtain the measurement signal from the horizontal water cut flow fusion measurement subsystem and use the cross-correlation algorithm to obtain the circumferential oil-water two-phase flow information Q. czow ; S25. Acquire the measurement signal from the gas volume measurement subsystem to obtain the gas phase flow information Q from the float flowmeter. fg ; S26. Obtain the measurement signal from the turbine flow measurement subsystem to obtain the oil-gas-water three-phase flow information Q. togs ; S27. By integrating the flow information from S21 to S26 above, the phase flow information Q = (Q m +(Q ozg +Q ozg +Q ozg )+(Q czow +Q zow ) / (1-Y g )+(Q fg ) / (Y g )+Q togs ) / 5; Methods for measuring gas content include: S31. The multi-task prediction model framework combines a frequency domain module, a hybrid attention module, and convolutional layers for hybrid modeling. It employs one one-dimensional convolutional layer, three two-dimensional convolutional layers, and two average pooling layers, with a residual connection added after the first pooling layer. The hybrid connection of the frequency domain layer, the hybrid attention module, and the convolutional layers effectively extracts features from the multiphase flow signal and uses these features as input. Each task has its own unique linear mapping layer, using the same core feature extraction structure and a unique decoding head paired with different activation functions to obtain the gas content y of the multiphase flow model. mg ; S32. Obtain the measurement signal from the gas volume measurement subsystem to obtain the gas phase flow information Q from the float flowmeter. fg ; S33. Obtain the measurement signal from the cumulative conductivity fiber optic measurement subsystem, and calculate the fiber conductivity and gas content y based on the horizontal layering principle and threshold signal processing method. ozg ; S34. Obtain the turbine flow measurement subsystem signal to get the oil-gas-water three-phase flow information Q. tgwo ; S35. Combining the gas phase flow rate information and the three-phase flow rate information of oil, gas and water, the gas content information Y is calculated. g =(y ozg +(Q fg ) / (Q tgwo )+y mg ) / 3.

Images