A synchronous dual-frequency excited annular gas-liquid two-phase flow electromagnetic tomography device and method

Abstract

The application provides a synchronous dual-frequency excitation annular air-liquid two-phase flow electromagnetic tomography device and method, which comprises a synchronous dual-frequency excitation module, an electromagnetic sensor array, a signal acquisition and demodulation module, an image reconstruction module and an image display module. The excitation signal generator and the power amplifier circuit constitute the synchronous dual-frequency excitation module; eight electromagnetic coils are uniformly distributed along the circumference of the annulus to constitute the electromagnetic sensor array; the data acquisition unit, the digital frequency mixer, the digital low-pass filter and the CORDIC calculation unit constitute the signal acquisition and demodulation module; the sensitivity matrix memory, the image reconstruction unit, the weighted fusion processor and the image denoising processing unit constitute the image reconstruction module. The application can improve the speed and quality of annular air-liquid two-phase flow electromagnetic tomography, realize real-time imaging and flow pattern recognition of the air-liquid two-phase flow distribution in the annular flow channel, and provide strong support for downhole air-liquid two-phase flow monitoring.

Description

Technical Field

[0001] This invention relates to the field of two-phase flow detection and electromagnetic tomography, and particularly to an electromagnetic tomography device and method for synchronous dual-frequency excitation of annular air-liquid two-phase flow. Background Technology

[0002] The annular structure, widely used in downhole oil and gas extraction and transportation systems, suffers from low imaging accuracy and poor real-time performance due to the complex distribution characteristics of its internal gas-liquid two-phase flow in dynamic measurement environments. Existing technologies have significant limitations: single-frequency excitation electromagnetic tomography (EMT) is constrained by information dimensionality, only acquiring electromagnetic responses at a single frequency and unable to simultaneously distinguish between high-conductivity and low-conductivity media; traditional analog demodulation techniques rely on complex hardware circuits, resulting in significant signal-to-noise ratio degradation in strong downhole electromagnetic interference environments, and lack the ability to simultaneously analyze multi-frequency features. To overcome the dilemma of simultaneously meeting the three major requirements of medium identification capability, dynamic measurement accuracy, and real-time imaging in annular gas-liquid two-phase flow measurement, synchronous dual-frequency excitation-based EMT technology has emerged. By simultaneously applying high-frequency and low-frequency composite electromagnetic excitation in a single measurement, integrating digital orthogonal demodulation and multi-source information analysis, high-precision image reconstruction is achieved, thereby realizing rapid real-time dynamic response and flow pattern recognition, providing a technological foundation for intelligent monitoring of downhole two-phase flows.

[0003] Fluid electromagnetic tomography commonly uses single-frequency excitation. This method typically employs a single-frequency sinusoidal AC signal as the excitation source, using a sensor array to electromagnetically excite the measured region. By measuring the electromagnetic field response signal at the boundary, the conductivity distribution image of the measured region is reconstructed using back-projection or linear inversion algorithms. The key to this technique lies in achieving the visual identification of the distribution of media with different conductivity levels through the electromagnetic field perturbation characteristics at a single frequency.

[0004] Single-frequency excitation electromagnetic tomography (EMT) has insufficient information acquisition capabilities, as it can only apply electromagnetic waves of a single frequency for excitation. In complex downhole environments, it struggles to simultaneously respond to the differentiated electromagnetic properties of high- and low-conductivity media. Because single-frequency measurements cannot provide multi-band complementary information, the system's sensitivity distribution is fixed and cannot be dynamically adjusted according to flow field changes, resulting in limited ability to identify two-phase flow components. Furthermore, due to the inherent limitation of insufficient single-frequency measurement data, the reconstructed images suffer from severe artifacts and edge blurring, failing to provide accurate flow pattern distribution information. Summary of the Invention

[0005] This invention aims to address the shortcomings of existing technologies by providing a synchronous dual-frequency excited electromagnetic tomography imaging device and method for annular air-liquid two-phase flow. This solution achieves high-precision imaging and real-time flow pattern identification of downhole annular air-liquid two-phase flow by integrating synchronous dual-frequency electromagnetic excitation, digital orthogonal demodulation, and multi-source image fusion technologies.

[0006] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a synchronous dual-frequency excited annular air-liquid two-phase flow electromagnetic tomography imaging device and method, which can effectively improve imaging speed and quality. Through this synchronous dual-frequency excited annular air-liquid two-phase flow electromagnetic tomography imaging device and method, real-time visual monitoring of the annular air-liquid two-phase flow distribution can be achieved, providing more reliable data support for flow pattern identification and process control, optimizing oil and gas extraction efficiency, and reducing downhole operation risks.

[0007] The present invention adopts the following technical solution: An electromagnetic tomography device for synchronous dual-frequency excitation of annular air-liquid two-phase flow includes a synchronous dual-frequency excitation module, an electromagnetic sensor array, a signal acquisition and demodulation module, an image reconstruction module, and an image display module.

[0008] The synchronous dual-frequency excitation module consists of an FPGA main controller and a dual DDS signal synthesis unit; The excitation signal generator and power amplifier circuit form a synchronous dual-frequency excitation module; The signal acquisition and demodulation module consists of a data acquisition unit, a digital mixer, a digital low-pass filter, and a CORDIC computing unit. The image reconstruction module consists of a sensitivity matrix memory, an image reconstruction unit, a weighted fusion processor, and an image denoising processing unit. Technical improvements were made to the traditional electromagnetic tomography (EMT) device, including synchronous dual-frequency signal output from the synchronous dual-frequency excitation module, quadrature demodulation and amplitude / phase extraction of the mixed signal from the signal acquisition and demodulation module, and weighted fusion of the dual-frequency images from the image reconstruction module. The FPGA main controller controls the excitation signal generator to produce the excitation signal, which is then amplified by a power amplifier circuit and distributed to the electromagnetic sensor array via a multiplexer. The detected signal is pre-amplified and anti-aliasing filtered before being acquired by the signal acquisition unit. It is then converted from an A / D converter and processed by a digital mixer, a digital low-pass filter, and a CORDIC computing unit. The processed signal has anti-interference capabilities. After demodulation preprocessing, the signal is transmitted to the image reconstruction module via the FPGA processor.

[0009] The electromagnetic sensor array uses a ring-shaped fixed bracket on which eight electromagnetic coils are evenly distributed. The central angle between adjacent coils is 45°, and the center of the coil faces the axis of the annular measuring pipe. The electromagnetic sensor array is arranged close to the inner ring sidewall of the annular space to generate a uniform magnetic field in the detection area.

[0010] After technical improvements to the device, image reconstruction and fusion processing are performed via an image reconstruction module to improve imaging speed and quality. The specific processing method includes the following steps: Step 1: Using the demodulated low-frequency and high-frequency amplitude data, and combining them with the sensitivity matrix pre-calculated and stored through finite element simulation, a linear back-projection algorithm is used to reconstruct the preliminary low-frequency and high-frequency images, respectively. Constructing the sensitivity matrix is ​​crucial for improving the system's imaging quality. The finite element method is used to calculate the sensitivity distribution of each element within the imaging region, establishing a mapping relationship between measured values ​​and object field parameters.

[0011] Step 2 involves processing the reconstructed image data. The main processing method is to use an adaptive weighted fusion algorithm based on regional features to improve the edge sharpness and contrast of the electromagnetic tomography image.

[0012] Step 3: Perform local variance calculation on the obtained low-frequency preliminary image and high-frequency preliminary image to reduce the impact of incomplete information in a single frequency image. The specific processing expression is as follows: In the formula, , The fusion weights are dynamically determined from the local variance. , Preliminary images of low and high frequencies. This is the merged image.

[0013] Step 4: Median filtering is used to eliminate noise points in the fused image, followed by image enhancement processing.

[0014] The beneficial effects of this invention are: 1. This invention improves upon traditional electromagnetic tomography devices by using synchronous dual-frequency excitation instead of single-frequency excitation, thereby increasing the dimensionality of object field information acquisition and the accuracy of image reconstruction.

[0015] 2. This invention optimizes the signal acquisition and demodulation process by employing digital quadrature demodulation and the CORDIC algorithm to achieve synchronous separation and amplitude-phase extraction of the mixing signal, effectively suppressing inter-channel interference and environmental noise, and improving the accuracy and reliability of the measurement data.

[0016] 3. This invention improves the image reconstruction process by using dual-frequency image weighted fusion, which solves the problem of incomplete information in single-frequency images and greatly improves the imaging quality. This is of great significance for real-time monitoring of gas-liquid two-phase flow in annular channels. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the device structure of the system of the present invention; Figure 2 This is a schematic diagram of the electromagnetic sensor array of the present invention; Figure 3 This is a block diagram of the internal functional modules of the FPGA of this invention; Figure 4 This is a flowchart of the digital orthogonal demodulation algorithm of the present invention; Figure 5 This is a flowchart of the image reconstruction and fusion algorithm of the present invention; Figure 6 This is a diagram showing the complete workflow and data flow of the system of the present invention.

[0018] In the diagram: 1-Synchronous dual-frequency excitation module, 2-Electromagnetic sensor array, 3-Signal acquisition and demodulation module, 4-Image reconstruction module, 5-Image display module, 6-Outer wall, 7-Ring fixed bracket, 8-Electromagnetic coil, 9-Inner wall. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0020] like Figure 1 As shown, the synchronous dual-frequency excited annular air-liquid two-phase flow electromagnetic tomography imaging device of the present invention includes a synchronous dual-frequency excitation module 1, an electromagnetic sensor array 2, a signal acquisition and demodulation module 3, an image reconstruction module 4, an image display module 5, an outer wall 6, an annular fixing bracket 7, an electromagnetic coil 8, and an inner wall 9. The present invention provides a complete electromagnetic tomography imaging system. Except for the image display module, all other modules can be implemented in the downhole instrument, and each module is powered by the downhole system power supply.

[0021] like Figure 2 As shown, the electromagnetic sensor array 2 consists of a single-layer ring array of eight electromagnetic coils. The coils are evenly distributed on the ring-shaped fixed bracket, with the central angle between adjacent coils being 45°. The center of the coils faces the axis of the annular measuring pipe. The coils are fixed by the ring-shaped fixed bracket, and the entire array is tightly fitted to the inner ring sidewall of the annulus, so that it generates a uniform magnetic field in the detection area.

[0022] like Figure 3As shown, the FPGA integrates an excitation generation module and a demodulation module. The excitation generation module contains two independent DDS cores and a digital adder. DDS core 1 generates a 10MHz high-frequency sine wave signal, and DDS core 2 generates a 100kHz low-frequency sine wave signal. The two signals are superimposed by the digital adder and output to an external D / A converter. The demodulation module contains a digital mixer, a digital low-pass filter, and a CORDIC calculation unit. The digital mixer receives the mixed signal acquired by the ADC and multiplies it by four quadrature reference signals. The digital low-pass filter filters out the sum-frequency components and extracts the I / Q components. The CORDIC calculation unit adopts a 12-stage pipeline structure and calculates the amplitude and phase information of each detection channel in real time through iterative shift and addition operations.

[0023] like Figure 4 As shown, signal acquisition and demodulation module 3 includes an 8-channel synchronous high-speed ADC array with a sampling rate of 10 MSPS. The acquired mixed signal is sent to the digital quadrature demodulation unit within the FPGA, which includes a digital mixer, a digital low-pass filter, and a CORDIC calculation unit. The digital mixer multiplies the input signal by four quadrature reference signals, which are synchronously generated by the excitation DDS core to ensure phase consistency. The digital low-pass filter has a cutoff frequency of 50 kHz to effectively filter out sum-frequency components. The CORDIC calculation unit calculates the amplitude and phase information of each detection channel corresponding to low and high frequencies in real time.

[0024] like Figure 5 As shown, image reconstruction module 4 runs on a host computer platform and includes a built-in sensitivity matrix memory, image reconstruction unit, weighted fusion processor, and image denoising unit. The image reconstruction unit uses a linear back-projection algorithm to reconstruct amplitude data at low frequencies of 100kHz and high frequencies of 10MHz, generating a low-frequency preliminary image reflecting the conductivity distribution and a high-frequency preliminary image reflecting the electromagnetic property distribution under high-frequency excitation. The weighted fusion processor uses an adaptive weighted fusion algorithm based on regional features to calculate the local variance of grayscale in each pixel region of the two preliminary images and dynamically allocates fusion weights according to the variance magnitude. The fusion formula is as follows: In the formula, , The fusion weights are dynamically determined from the local variance. , Preliminary images of low and high frequencies. To obtain the fused image, the image denoising unit uses a median filtering algorithm to denoise the fused image.

[0025] like Figure 6As shown, the complete system workflow is as follows: After the system is powered on, the FPGA initializes the parameters of each module; the synchronous dual-frequency excitation module generates a composite excitation signal, which is selected by the multiplexer to select the excitation coil; the detection coil array synchronously acquires the response signal, which is then converted by the ADC and sent to the digital quadrature demodulation module; the demodulated dual-frequency data is transmitted to the host computer for image reconstruction and fusion processing; the final result is displayed in real time in the image display module.

[0026] When using this device to measure gas-liquid two-phase flow in an annular channel, the key to obtaining high-quality imaging lies in synchronous dual-frequency excitation, digital quadrature demodulation, and dual-frequency image fusion. The specific implementation steps are as follows: Step 1. Applying synchronous dual-frequency excitation: Two direct digital frequency synthesis cores are started synchronously to generate high-frequency 10MHz and low-frequency 100kHz sine signals respectively. These signals are superimposed in the digital domain by an adder to form a composite synchronous dual-frequency digital signal. The signal is then converted into an analog signal by a high-speed digital-to-analog converter, amplified by a broadband power amplifier, and then selected by a multiplexer to apply synchronous dual-frequency composite electromagnetic excitation to the annular measurement area.

[0027] Step 2. Response Signal Acquisition and Digital Quadrature Demodulation: The response signals from the remaining seven detection coils in the array are synchronously sensed, amplified, and filtered against anti-aliasing before being synchronously sampled and converted into digital signals by an 8-channel high-speed ADC. The digitized mixed signal is then digitally multiplied with four quadrature reference signals. After filtering out the sum-frequency component using a digital low-pass filter, the in-phase and quadrature components corresponding to the low and high frequencies are extracted. The amplitude is then synchronously extracted from the I / Q components using a 12-stage pipelined CORDIC algorithm through parallel processing by a CORDIC computing unit. With phase In the formula, For in-phase components, For orthogonal components, For amplitude, Using phase as the basis, the amplitude and phase information of each detection channel corresponding to low and high frequencies are calculated.

[0028] Step 3. Image Reconstruction and Fusion Processing: Using the low-frequency and high-frequency amplitude data obtained from demodulation, combined with the sensitivity matrix pre-calculated and stored through finite element simulation, a linear back-projection algorithm is used to reconstruct a preliminary low-frequency image reflecting the conductivity distribution and a preliminary high-frequency image reflecting the electromagnetic property distribution under high-frequency excitation. An adaptive weighted fusion algorithm based on region features is used to fuse the two-frequency images: the local variance of grayscale in each pixel region of the two preliminary images is calculated, and fusion weights are dynamically allocated according to the variance magnitude. The fusion formula is as follows: In the formula, , The fusion weights are dynamically determined from the local variance. , Preliminary images of low and high frequencies. The fused image is then subjected to median filtering for noise reduction to generate the final fluid distribution image.

[0029] Step 4. Flow pattern identification and parameter calculation: The reconstructed final image is matched with the preset flow pattern database to identify the flow pattern category of the gas-liquid two-phase flow (such as bubble flow, slug flow, annular flow, etc.), and the phase content is calculated and output to the image display module for visualization.

[0030] Example: The device of this invention was used to measure the gas-liquid two-phase flow in an annular pipe. Eight coils of the electromagnetic sensor array were used sequentially as excitation coils, and the remaining seven coils as detection coils. A composite excitation signal of 10MHz and 100kHz was applied, and 28 sets of response voltage data were collected. After extracting amplitude and phase information through digital orthogonal demodulation, preliminary low-frequency and high-frequency images were reconstructed using a linear back-projection algorithm, and then the final gas-liquid two-phase flow distribution image was generated through adaptive weighted fusion. The imaging results clearly showed the position and morphology of the bubbles, verifying the effectiveness of the method of this invention.

[0031] 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A synchronous dual-frequency excited annular air-liquid two-phase flow electromagnetic tomography imaging device, characterized in that, It includes a synchronous dual-frequency excitation module, an electromagnetic sensor array, a signal acquisition and demodulation module, an image reconstruction module, and an image display module; The synchronous dual-frequency excitation module consists of an FPGA main controller and a dual DDS signal synthesis unit; The excitation signal generator and power amplifier circuit form a synchronous dual-frequency excitation module; The signal acquisition and demodulation module consists of a data acquisition unit, a digital mixer, a digital low-pass filter, and a CORDIC computing unit. The image reconstruction module consists of a sensitivity matrix memory, an image reconstruction unit, a weighted fusion processor, and an image denoising processing unit. The FPGA main controller controls the excitation signal generator to generate an excitation signal, which is then amplified by a power amplifier circuit and distributed to the electromagnetic sensor array by a multiplexer. The detection signal is pre-amplified and anti-aliasing filtered before being sampled and converted into a digital signal by an 8-channel synchronous high-speed ADC. The digital signal is then processed by a digital mixer and a digital low-pass filter to extract the in-phase and quadrature components of the low and high frequencies. The CORDIC calculation unit processes the data to calculate the amplitude and phase of each detection channel corresponding to the low and high frequencies. The amplitude and phase data are then transmitted to the image reconstruction module via the FPGA processor.

2. The synchronous dual-frequency excited annular air-liquid two-phase flow electromagnetic tomography imaging device according to claim 1, characterized in that, The electromagnetic sensor array uses a ring-shaped fixed bracket on which eight electromagnetic coils are evenly distributed. The central angle between adjacent coils is 45°, and the center of the coil faces the axis of the annular measuring pipe. The electromagnetic sensor array is arranged close to the inner ring sidewall of the annular space to generate a uniform magnetic field in the detection area.

3. A synchronous dual-frequency excited electromagnetic tomography method for annular air-liquid two-phase flow, characterized in that, include: Step 1, synchronous dual-frequency excitation application: Two direct digital frequency synthesis cores are synchronously activated to generate high-frequency 10MHz and low-frequency 100kHz sine signals respectively. These signals are superimposed in the digital domain by an adder to form a composite synchronous dual-frequency digital signal. The signal is then converted into an analog signal by a high-speed digital-to-analog converter, amplified by a broadband power amplifier, and then selected by a multiplexer to apply synchronous dual-frequency composite electromagnetic excitation to the annular measurement area. Step 2, Response Signal Acquisition and Digital Quadrature Demodulation: The synchronously sensed response signals from the remaining 7 detection coils in the array are pre-amplified and anti-aliasing filtered before being synchronously sampled and converted into digital signals by an 8-channel high-speed ADC. The digitized mixed signals are then digitally multiplied with the four quadrature reference signals. After filtering out the sum-frequency components by a digital low-pass filter, the in-phase and quadrature components corresponding to the low and high frequencies are extracted. The amplitudes are then synchronously extracted from the I / Q components through parallel processing by the CORDIC computing unit using a 12-stage pipelined CORDIC algorithm. With phase In the formula, For in-phase components, For orthogonal components, For amplitude, Using phase as the basis, the amplitude and phase information corresponding to low and high frequencies for each detection channel are calculated; Step 3, Image Reconstruction and Fusion Processing: Using the demodulated low-frequency and high-frequency amplitude data, combined with the sensitivity matrix pre-calculated and stored through finite element simulation, a linear back-projection algorithm is used to reconstruct a low-frequency preliminary image reflecting the conductivity distribution and a high-frequency preliminary image reflecting the electromagnetic property distribution under high-frequency excitation. An adaptive weighted fusion algorithm based on region features is then used to fuse these two preliminary images: the local variance of each pixel region in the two preliminary images is calculated, and fusion weights are dynamically allocated according to the variance magnitude. The fusion formula is as follows: In the formula, , The fusion weights are dynamically determined from the local variance. , Preliminary images of low and high frequencies. The fused image is then subjected to median filtering for noise reduction to generate the final fluid distribution image. Step 4, Flow pattern identification and parameter calculation: The reconstructed final image is matched with the preset flow pattern database to identify the flow pattern category of the gas-liquid two-phase flow, and the phase content is calculated and output to the image display module for visualization.

4. The synchronous dual-frequency excited electromagnetic tomography method for annular air-liquid two-phase flow according to claim 4, characterized in that, In step 1, the adjacent excitation mode is adopted. After one excitation and measurement is completed, the multiplexer automatically switches to the next coil as the excitation coil until all 8 coils are excited once, completing a complete measurement cycle.

5. The synchronous dual-frequency excited electromagnetic tomography method for annular air-liquid two-phase flow according to claim 4, characterized in that, In step 2, the four quadrature reference signals are synchronously generated by the excitation DDS core to ensure that they are in phase and frequency with the excitation signal.

6. The synchronous dual-frequency excited electromagnetic tomography method for annular air-liquid two-phase flow according to claim 4, characterized in that, In step 3, the adaptive weighted fusion algorithm based on regional features calculates the local variance for each pixel region. The larger the local variance, the greater the fusion weight of the high-frequency image, so as to preserve the edge details of the gas-liquid interface; the smaller the local variance, the greater the fusion weight of the low-frequency image, so as to suppress noise in the uniform region.

Images