Ultrasonic imaging method for detecting liquid droplets and gas bubbles in oil-gas-water three-phase medium

Abstract

The present application relates to a kind of ultrasonic A scanning signal assisted B mode imaging methods for detecting liquid drop and bubble in oil-gas-water three-phase medium, the ultrasonic phased array UPA sensor for detecting use is shared M array elements, for the transmitting array element of synthesizing ultrasonic plane wave is Tx, and the array element for receiving ultrasonic wave is Rx;Realize the ultrasonic imaging of liquid drop and bubble in oil-gas-water three-phase medium in two cases, first, in the need of low consumption time detection, and have the condition of preparation for calibrating oil phase sound velocity test model: using BI method can visualize bubble and liquid drop, and AS method measures bubble and liquid drop;Second, in the high requirement of liquid drop measurement robustness, and do not have the condition of preparation for calibrating oil phase sound velocity test model: using BI method can visualize bubble and liquid drop, and measure liquid drop, and AS method measures bubble.

Description

Technical Field

[0001] The present invention belongs to the field of ultrasonic visualization and parameter measurement of industrial process parameters in multiphase media, and relates to an ultrasonic A-scan signal-assisted B-mode imaging (AS-assisted-BI) method for detecting droplets and bubbles in an oil-gas-water three-phase medium. The method involves a B-mode imaging (BI) method using a single plane wave (PW) synthesized by an ultrasonic phased array (UPA) sensor, and an A-scan signal (AS) analysis method to visualize, identify, and numerically measure the size and position information of bubbles and droplets in the oil-gas-water three-phase medium, as well as measure (calibrate) the sound velocity of the liquid in the droplets, thereby synchronously obtaining images and numerical information of the three-phase medium. Background Art

[0002] Multiphase media, consisting of two or more physically immiscible media with distinct boundaries, are widely used in industries such as energy, chemicals, aerospace, healthcare, and pharmaceuticals. Accurately measuring industrial process parameters in multiphase media is crucial for critical decisions regarding production safety and efficiency, economic benefits, and environmental protection. Ultrasonic process parameter measurement offers the advantages of being non-invasive, non-perturbing, unaffected by the salinity of the medium, easy to use, and environmentally friendly.

[0003] Ultrasonic testing methods are divided into two categories based on whether the detection results of industrial process parameters are visualized. One category involves ultrasonic measurement (UM) methods, which quantify process parameters such as phase holdup, flow velocity, and size of discrete phases (such as bubbles, droplets, and particles). The other category involves ultrasonic B-mode imaging (UBI) or tomography (UT) methods, which visualize fluid distribution and flow patterns through opaque pipes. Both UM and UT methods suffer from low detection accuracy, narrow range, limited functionality, and low efficiency. Low detection accuracy is manifested in poor numerical accuracy of particle size and phase holdup for UM methods, while UT methods exhibit distortion and artifacts in the shape, size, and spatial distribution of discrete phases in the image. The narrow detection range of UM and UT methods is reflected in their applicability only to two-phase media where both discrete phases possess strong acoustic impedance and high echogenicity, such as bubble-water and particle-water. However, few studies have examined media containing discrete phases with weak acoustic impedance and low echogenicity, such as oil-water two-phase media containing droplets, and oil-gas-water three-phase media with both high and low acoustic impedance (bubbles and droplets). UM and UT methods have limited detection capabilities, as they have yet to achieve simultaneous visualization and numerical measurement of discrete phases such as bubbles, particles, and droplets in multiphase media. UM and UT methods also suffer from low detection efficiency, manifested in slow signal acquisition, numerical solution, and image reconstruction. One reason for the narrow detection range and low accuracy is the low spatial sampling accuracy of the acquired ultrasonic data. Existing research typically uses independent piezoelectric chip ultrasonic sensors. The large probe and chip size limit the total number of acquired ultrasonic signals and reduce the spatial sampling rate. Consequently, ultrasonic testing instruments struggle to accurately determine the spatial distribution of discrete phases in multiphase media. A second reason for the narrow detection range and low accuracy, as well as the limited detection capabilities, is the incomplete utilization of ultrasonic data due to imperfect detection method design. Both the UM and UT methods use only signal amplitude (Amplitude) or time of flight (TOF) as raw data for inverting discrete phase acoustic impedance differences, size, and spatial distribution, without considering the complete signal waveform—that is, the discrete phase information details contained in the full waveform. This low detection efficiency is due to the fact that the UM method utilizes an iterative approach combined with relevant mathematical models of the measured object to invert the industrial process parameters of the discrete phase, while the UT method uses the Radon transform as its mathematical foundation to invert the spatial distribution image of the multiphase medium. The mathematical foundations used to design these two methods dictate that obtaining high-precision discrete phase values ​​and images requires the acquisition, storage, and processing of a large number of ultrasonic signals, resulting in high computational costs.

[0004] To address these challenges, some researchers have applied the UBI method, a technique used in medical diagnostics and nondestructive testing, to industrial process parameter detection. This method measures or visualizes the size, spatial distribution, and phase fraction of discrete phases in industrial multiphase media. The UBI method utilizes a single UPA sensor, comprised of numerous tiny ultrasonic probes, to acquire A-scan signals by increasing the spatial sampling rate of the detection field, thereby improving ultrasonic signal acquisition accuracy. The mathematical foundation of B-mode imaging is the delay and sum (DAS) method, rather than the Wraiden transform used in UT. The DAS method synthesizes B-mode images using the full waveform of the A-scan signal—that is, the pulse echo amplitude at each time-of-flight (TOF) instant, rather than relying solely on a large number of amplitudes or TOF values ​​as raw data before the inversion step in UM or UT methods. This reduces the number of ultrasonic signals required during the signal acquisition step and has the potential to improve the detection efficiency of the UBI method. On the other hand, the DAS method fully utilizes detailed information in the full ultrasound waveform signal during the B-mode image reconstruction step, which can characterize the size and spatial distribution of the measured object. It also avoids the errors caused by the extensive calculation of ultrasound amplitude attenuation or time-of-flight values ​​required before the inversion step in the UM and UT methods. Therefore, it has the potential to improve detection accuracy and range. However, the detection efficiency of existing UBI methods still needs to be improved. The imbalance between computational time and detection accuracy leads to low detection efficiency of UBI methods. One reason for this is that the total matrix focus (TMF) method used by researchers sacrifices signal acquisition and image reconstruction speed in exchange for the highest B-mode image accuracy. A second reason is that some researchers, when attempting to use TMF-reconstructed B-mode images to obtain numerical information such as the size and position of discrete phases (such as bubbles) in multiphase media, have not reduced the number of key pixels required to be identified in the image. Compared to the TMF method, ultrasound plane wave imaging (PWI), a method that has garnered widespread attention in the medical detection and diagnosis field, enables ultrafast A-scan signal acquisition and B-mode image reconstruction, enabling the detection of high- or low-acoustic-impedance media such as internal organ structures, cysts, and calculi. PWI offers lower computational cost and higher detection efficiency in acquiring ultrasound A-scan signals and reconstructing B-mode images. However, PWI has not yet been applied to the detection of industrial multiphase media. Furthermore, the aforementioned limitations of the UBI method, shared with the UM and UT methods, such as its narrow detection range and limited functionality, still require further resolution. For example, it has not yet addressed the detection of industrial engineering parameters in oil-gas-water three-phase media containing both high- and low-acoustic-impedance media, and it has not yet enabled the joint visualization and numerical measurement of bubbles and droplets in these three-phase media. This is primarily due to imperfect detection method design, which incompletely utilizes the discrete phase information contained in ultrasound B-mode images and A-scan signals.To obtain the numerical information of the size and position of bubbles and droplets in oil-gas-water three-phase medium, it is worth learning the A-scan signal analysis method proposed by scholars in the field of medical detection and diagnosis for calculating the characteristic parameters of human tumors. The method analyzes the amplitude, interval and density of the ultrasonic echoes in the signal that can represent the tumor parameters by manually marking the key points in the A-scan signal, and then calculates the size and position of the tumor in the human tissue. However, the manual marking of the key points limits the detection accuracy and efficiency. SUMMARY

[0005] Based on the ultrasonic phased array sensor and the reflected mode ultrasonic A-scan signal, the present application provides an ultrasonic A-scan signal assisted B-mode imaging method for detecting droplets and bubbles in oil-gas-water three-phase medium, which is called AS-assisted-BI method. The AS-assisted-BI method includes two detection schemes based on BI and AS methods, which are suitable for different detection requirements and conditions in actual industrial scenes. The first scheme is that when low-time-consuming detection is urgently needed and the oil phase sound velocity can be calibrated: the BI method is used to visualize bubbles and droplets, and the AS method is used to measure bubbles. The second scheme is that when the low-time-consuming requirement can be compromised, the droplet measurement robustness requirement is high, and there is no condition to calibrate the sound velocity: the BI method is used to visualize bubbles and droplets and measure droplets; the AS method is used to measure bubbles. The functions of the BI method include: visualizing the oil-gas-water three-phase medium and identifying the bubbles and droplets in the image; measuring the position of the bubbles and the size and position of the droplets. The functions of the AS method include: measuring the position of the bubbles and the size and position of the droplets; calibrating the liquid sound velocity in the droplet; reducing the overall time consumption of the AS-assisted-BI method by reducing the frequency of using the BI method. Finally, the visualization, identification, size and position parameters of bubbles and droplets in the oil-gas-water three-phase medium are realized. The technical scheme is as follows:

[0006] An ultrasonic A-scan signal assisted B-mode imaging method for detecting droplets and bubbles in oil-gas-water three-phase medium, the ultrasonic phased array UPA sensor used for detection has M array elements, the transmitting array element for synthesizing ultrasonic plane wave is Tx, and the array element for receiving ultrasonic wave is Rx; a two-dimensional Cartesian coordinate system x-o-y is established with the center of the measured field as the origin, the bubble and droplet bubble diameter size is D, and the center position coordinates are (x s ,y s ), the ultrasonic imaging of droplets and bubbles in oil-gas-water three-phase medium is realized in two cases,

[0007] The first one is that low-time-consuming detection is needed, and the test model for calibrating the oil phase sound velocity can be prepared: the BI method is used to visualize bubbles and droplets, and the AS method is used to measure bubbles and droplets, including the following steps:

[0008] All Rx array elements of the S11 ultrasonic phased array UPA sensor receive the full-waveform time-domain signal of the ultrasonic plane wave reflected by the oil-gas-water three-phase medium in the measured field, completing the acquisition of the ultrasonic A-scan signal;

[0009] S12 uses ultrasonic A-scan signals and ultrasonic plane wave imaging (PWI) method based on the time domain delay sum (DAS) algorithm to construct B-mode images of the oil-gas-water three-phase medium.

[0010] S13 identifies bubbles and droplets from the B-mode image of the oil-gas-water three-phase medium;

[0011] S14 identifies bubbles and droplets from the A-scan signal of oil-gas-water three-phase medium;

[0012] S15 intercepts the A-scan signal of each bubble in the oil-gas-water three-phase medium one by one. Each bubble signal contains only one "peak". The transit time TOF corresponding to the "peak" represents the spatial position x of the front surface of the bubble in the measured field. fs , locate the unique key point u of the "peak" representing the position of the bubble front surface in each bubble A scan signal fs ;

[0013] S16 The unique key point u in each bubble signal located according to step (5) fs , calculate the horizontal coordinate x of the front surface position fs ;

[0014] S17 preparation is used to calibrate the sound velocity V of the oil phase medium in the droplet liquid The test model is used to collect the A-scan signal of the test model and calculate the sound velocity V of the oil phase medium to be calibrated. liquid ;

[0015] S18 intercepts the A-scan signal of each droplet in the oil-gas-water three-phase medium one by one on the basis of step (4). Each droplet signal contains two "peaks", which represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bs , the amplitude of the “peak” in the bubble signal is larger than the amplitude of any “peak” of the droplet;

[0016] S19 locates the two key points u representing the "peak" of the front surface position in the A-scan signal of each droplet. fs and u bs , according to the two key points u in each droplet signal located fs and u bs , combined with the calibrated sound velocity V of the liquid inside the droplet liquid , calculate the droplet center coordinate x s , bubble diameter D;

[0017] The second method, when robustness of droplet measurement is required and a test model for calibrating the oil-phase acoustic velocity is not available, is to use the BI method to visualize bubbles and droplets and measure droplets, and the AS method to measure bubbles. This method includes the following steps:

[0018] All Rx array elements of the S21 ultrasonic phased array UPA sensor receive the full-waveform time-domain signal of the ultrasonic plane wave reflected by the oil-gas-water three-phase medium in the measured field, completing the acquisition of the ultrasonic A-scan signal;

[0019] S22 uses ultrasonic A-scan signals and ultrasonic plane wave imaging (PWI) method based on the time domain delay sum (DAS) algorithm to construct B-mode images of the oil-gas-water three-phase medium.

[0020] S23 identifies bubbles and droplets from the B-mode image of the oil-gas-water three-phase medium;

[0021] S24 intercepts the B-mode image of the droplet in the oil-gas-water three-phase medium one by one, traverses all pixels of the droplet B-mode image, and locates the horizontal coordinates of the four key points of the two pixel strips representing the front and back surfaces of each droplet: the key point at the top of the front surface x fst , key point x at the bottom of the front surface fse , the key point x at the top of the back surface bst , key point x at the bottom of the rear surface bse ;

[0022] S25 calculates the droplet center coordinate x based on the four key points in the droplet B-mode image. s , bubble diameter D;

[0023] S26 identifies bubbles from the A-scan signal of the oil-gas-water three-phase medium;

[0024] S27, based on step S23, intercepts the A-scan signal of each bubble in the oil-gas-water three-phase medium one by one;

[0025] S28 locates the unique key point u in the A-scan signal of each bubble, representing the peak position of the front surface of the bubble fs ;

[0026] S29 is based on the unique key point u in each bubble signal located in S28 fs , calculate the position coordinate x of the bubble front surface fs .

[0027] Furthermore, the identification methods of bubbles and droplets from the B-mode images of the oil-gas-water three-phase medium in S13 and S23 are both: each bubble image contains only one pixel band, which represents the spatial position x of the front surface of the bubble in the measured field. fsEach droplet image contains two pixel bands, which represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bs .

[0028] Furthermore, the method of S14 for identifying bubbles and droplets from the A-scan signal of the oil-gas-water three-phase medium is as follows:

[0029] Each bubble signal contains only one "peak", and the transit time TOF corresponding to the "peak" represents the spatial position x of the front surface of the bubble in the measured field. fs ;

[0030] Each droplet signal contains two "peaks", which represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bs , the amplitude of the “peak” in the bubble signal is larger than the amplitude of any “peak” of the droplet.

[0031] Furthermore, in S15, the unique key point u representing the peak position of the front surface of the bubble is located in the A-scan signal of each bubble. fs The method is: the unique key point u fs The position of the unique “peak” of the A-scan signal corresponding to the bubble; the A-scan signal τ of a bubble of any Rx array element i (u)i∈[1,M], find G bubble ={u|A ext [τ i (u)]},H bubble ={u|u=mT}(m∈N + ), {u fs}=U bubble =G bubble ∩H bubble , where G bubble is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H bubble is the set of ultrasonic pulse intervals, m is the number of pulse intervals, T = 1 / f is the pulse interval period of the excitation, U bubble is the intersection of the two sets above.

[0032] Furthermore, in S16, the horizontal coordinate x of the front surface position of a bubble is determined. fs The method is:

[0033] calculate where x fsj is the front surface position of the bubble calculated based on the A-scan signal of the bubble of the jth Rx array element, V water is the speed of sound in water, u is the A-scan signal fs TOF value corresponding to the moment;

[0034] The horizontal coordinate x of the front surface position of this bubble fs By Collection All x between the upper quartile Q1 and the lower quartile Q3 fsj The average value is calculated using the formula:

[0035] Furthermore, the method of S17 is:

[0036] 1) Preparation for calibration of the sound velocity V of the oil phase medium in the droplet liquid The test model is as follows: an agar block containing a rectangular cavity is prepared. The density and sound velocity of the agar block are close to those of water. The center of the rectangular cavity coincides with the center of the agar block, and the length of the rectangular cavity is greater than the total length of all elements of the UPA sensor. The UPA sensor is installed directly opposite the rectangular cavity, and the cavity is filled with the oil phase medium whose sound velocity is to be calibrated.

[0037] 2) The A-scan signals from the (1+M / 4)th to (MM / 4)th Rx elements are used to calibrate the sound velocity V of the oil phase medium in the droplet liquid Each A-scan signal contains two “peaks”, which respectively represent the spatial positions of the front and rear surfaces of the rectangular cavity filled with the oil phase medium to be calibrated in the measured field. The two key points u of these two “peaks” are fs and u bs The positioning method is: for any of the above Rx array elements, the A scan signal τ i (u)i∈[1,M], find G liquid ={u|A ext [τ i (u)]},H liquid ={u|u=mT}(m∈N + ), {u fs ,u bs}=U liquid =G liquid ∩H liquid , where G liquid is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H liquid is the set of ultrasonic pulse intervals, T = 1 / f is the pulse interval period of the excitation, m is the number of pulse intervals, U liquid is the intersection of the above two sets;

[0038] 3) Calculate the sound velocity V of the oil phase medium to be calibrated liquid The calculation method is in and are u in the i-th A-scan signal fs and u bs The TOF value corresponding to the moment, k is the total number of A-scan signals, L liquid is the length of the cuboid cavity;

[0039] Furthermore, the method of step S19 is:

[0040] 1) Locate the two key points u representing the "peak" of the front surface position in the A-scan signal of each droplet fs and u bs : Key point u fs and u bs They correspond to the positions of the two “peaks” of the A-scan signal of the droplet; the A-scan signal τ of a droplet of any Rx array element i (u)i∈[1,M], find G droplet ={u|A ext [τ i (u)]},H droplet ={u|u=mT}(m∈N + ), {u fs ,u bs}=U droplet =G droplet ∩H droplet , where G droplet is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H droplet is the set of ultrasonic pulse intervals, T = 1 / f is the pulse interval period of the excitation, m is the number of pulse intervals, U droplet is the intersection of the above two sets;

[0041] 2) Combined with the calibrated sound velocity V of the liquid in the droplet liquid , calculate the droplet center coordinate x s , bubble diameter D:

[0042] in and are u in the i-th A-scan signal fs and u bs The TOF value corresponding to the moment, k is the total number of A-scan signals, L liquid is the length of the rectangular cavity; D j is the droplet size calculated based on the A-scan signal of a droplet of a certain Rx array element, V oil is the sound velocity of the oil phase medium in the droplet; finally, the size D of any droplet is determined by the set {D j All D between the upper quartile Q1 and the lower quartile Q3 jCalculate the average value of

[0043] 3) Calculate the droplet center coordinate x s : where x sj is the droplet center position calculated based on the A-scan signal of a droplet of a certain Rx array element, V water is the speed of sound of water; finally, the center position of the droplet x s By the set {x s All x between the upper quartile Q1 and the lower quartile Q3 sj Calculate the average value of .

[0044] Furthermore, in step S26, the method for identifying bubbles from the A-scan signal of the oil-gas-water three-phase medium is as follows: each bubble signal contains only one "peak", and the transit time TOF corresponding to the "peak" represents the spatial position x of the front surface of the bubble in the measured field. fs .

[0045] Furthermore, in S28, the unique key point u in the A-scan signal of any bubble fs The positioning method is: the unique key point u fs The position of the unique “peak” of the A-scan signal corresponding to the bubble; the A-scan signal τ of a bubble of any Rx array element i (u)i∈[1,M], find G bubble ={u|A ext [τ i (u)]},H bubble ={u|u=mT}(m∈N + ), {u fs}=U bubble =G bubble ∩H bubble , where G bubble is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H bubble is the set of ultrasonic pulse intervals, m is the number of pulse intervals, T = 1 / f is the pulse interval period of the excitation, U bubble is the intersection of the two sets above.

[0046] Furthermore, in S29, the horizontal coordinate x of the front surface position of a bubble is determined. fs The method is:

[0047] calculate where x fsj is the front surface position of the bubble calculated based on the A-scan signal of the bubble of the jth Rx array element, V water is the speed of sound in water, u is the A-scan signalfs TOF value corresponding to the moment;

[0048] The horizontal coordinate x of the front surface position of this bubble fs By Collection All x between the upper quartile Q1 and the lower quartile Q3 fsj The average value is calculated using the formula:

[0049] The beneficial effect of the present invention is that, compared to traditional ultrasonic visualization and measurement methods in the field of industrial process parameter detection of multiphase media, the present invention realizes the joint visualization of the size and spatial position distribution of bubbles and droplets in the oil-gas-water three-phase medium through reflection mode ultrasonic information, identifies bubbles and droplets, measures the position of bubbles, measures the size and position of droplets, and calibrates the sound velocity of the liquid in the droplets, thereby jointly obtaining images and numerical information of the three-phase medium. The present invention designs an ultrasonic A-scan signal assisted B-mode imaging (AS-assisted-BI) method, which includes two sub-methods: BI method and AS method. In the BI method, the oil-gas-water three-phase medium visualization method based on the PWI method of ultrasonic single plane wave B-mode imaging: First, the multiphase medium detection range of the traditional UBI method in the field of industrial process parameter measurement is expanded from two-phase media to oil-gas-water three-phase media. Second, compared with the TMF method in the UBI method, the PWI method improves the detection efficiency of three-phase media by reducing the required number of A-scan signal acquisitions and B-mode image reconstruction calculation steps. Third, the PWI method used in the BI approach can accurately identify bubbles and droplets from lower-precision B-mode images of oil-gas-water three-phase media, achieving higher efficiency in visual detection of industrial process parameters compared to the TMF method. In the BI approach, the acoustic identification method for bubbles and droplets based on B-mode images and the bubble and droplet parameter measurement method based on pixel key point extraction from B-mode images reduce the number of key pixels required to identify from B-mode images for bubble and droplet measurement. This improves detection efficiency compared to traditional industrial process parameter measurement methods based on B-mode images. In the AS approach, the acoustic identification method for bubbles and droplets using A-scan signals provides a previously unseen strategy for acoustic identification of bubbles and droplets using A-scan signals in the field of industrial process parameter detection in multiphase media. In the AS method, a method for measuring bubble and droplet parameters by automatically extracting key points of A-scan signals: Compared with the manual marking of key points in A-scan signals in the field of medical detection and diagnosis in order to measure the size and position of tumors in human tissue, the method for automatically extracting key points of A-scan signals in the present invention can simultaneously improve the detection accuracy and efficiency of the size and position of bubbles and droplets. In the AS method, a method for calibrating the sound velocity of liquid in droplets by automatically extracting key points of A-scan signals: provides a strategy for calibrating the sound velocity of liquid in droplets using A-scan signals that has not yet been proposed in the field of industrial process parameter detection of multiphase media. This method can measure the sound velocity of liquids other than pure water with high precision, provides convenience for understanding the acoustic properties of liquids in droplets as the detected object, and can improve the detection accuracy of droplet size and position measurement using the A-scan signal method. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1This is a flow chart of the "ultrasonic A-scan signal-assisted B-mode imaging method for detecting droplets and bubbles in an oil-gas-water three-phase medium" of the present invention.

[0051] Figure 2 The UPA sensor installation structure, detected object and its field of the UPA industrial process parameter detection simulation system of the present invention are shown. The UPA sensor has a total of M = 64 array elements and a field width W = 50 mm.

[0052] Figure 3 Schematic diagram of the simulation and experimental test models selected for this example. To test the effectiveness of the present method for detecting bubbles and droplets in a region of interest (ROI) within an oil-gas-water three-phase medium within the test field, a UPA sensor with a total array element count of M = 16 and a field width of W = 50 mm was selected: (a) Bubble model, (b) Droplet model.

[0053] Figure 4 This is the result of visualizing and identifying bubbles and droplets in the oil-gas-water three-phase medium using the BI method of the present invention.

[0054] Figure 5 In the embodiment of the present invention, the size D=6mm, the center position (x s ,y s )=(0,0) as an example, a schematic diagram of the result of locating the four key points in its B-mode image.

[0055] Figure 6 BI method to measure the droplet size D and the center longitudinal position x s Box plot of the absolute error AE value: Group ⅰ-a: measurement position x s Statistical results of AE values ​​varying with size D (based on experimental data); Group i-b: measurement position x s The AE value varies with the distance parameter r TX Statistical results of the second group: the statistical results of the AE value of the measured size D with the change of the diameter D (based on simulation data); the second group: the statistical results of the AE value of the measured size D with the change of the center position x s The statistical results of the changes (based on simulation data). Among them, the distance parameter r TX Indicates Figure 2 The x-axis is shown from the center position of the bubble or droplet s Distance to the UPA sensor.

[0056] Figure 7 In the embodiment of the present invention, the size D=6mm, the center position (x s ,y s)=(0,0) as an example, the result diagram of the automatic identification of the unique “peak” of its A-scan signal and the unique key point located.

[0057] Figure 8 The longitudinal position x of the front surface of the bubble is measured by AS method fs Box plot of the absolute error AE value: Group ⅰ-a: measurement position x fs Statistical results of AE values ​​varying with size D (based on experimental data); Group i-b: measurement position x fs The AE value varies with the distance parameter r TX The statistical results of the changes (based on simulation data). Among them, the distance parameter r TX Indicates Figure 2 The x-axis is shown from the center position of the bubble or droplet s Distance to the UPA sensor.

[0058] Figure 9 This is the test model required to prepare for calibrating the sound velocity of oil in a droplet according to the present invention.

[0059] Figure 10 For size L oil =20mm liquid sound velocity calibration model is an embodiment, and the two "peaks" of the A-scan signal are automatically identified, and the result schematic diagram of the two key points located.

[0060] Figure 11 In the embodiment of the present invention, the size D=6mm, the center position (x s ,y s )=(0,0) as an example, the two “peaks” of its A-scan signal are automatically identified, and the two key points are located.

[0061] Figure 12 is the droplet size D and the central longitudinal position x measured by the AS method s Box plot of the absolute error AE value: Group i-a: measurement position x s Statistical results of AE values ​​varying with size D (based on experimental data); Group i-b: about measurement position x s The AE value varies with the distance parameter r TX Ⅱ-a group: statistical results of the change of AE value of the measured size D with the change of diameter D (based on simulation data); Ⅱ-b group: statistical results of the change of AE value of the measured size D with the change of center position x s The statistical results of the changes (based on simulation data). Among them, the distance parameter r TX Indicates Figure 2 The x-axis is shown from the center position of the bubble or droplet s Distance to the UPA sensor. DETAILED DESCRIPTION

[0062] The present invention relates to the field of ultrasonic visualization and measurement of industrial process parameters of multiphase media, and relates to an ultrasonic A-scan signal-assisted B-mode imaging (AS-assisted-BI) method, which includes two sub-methods: the BI method and the AS method. The BI method includes an oil-gas-water three-phase medium visualization method based on the PWI method of ultrasonic single-plane wave B-mode imaging, a bubble and droplet acoustic identification method based on B-mode images, and a bubble and droplet parameter measurement method based on pixel key point extraction from B-mode images. The AS method includes: an A-scan signal bubble and droplet acoustic identification method, a bubble and droplet parameter measurement method that automatically extracts A-scan signal key points, and a liquid sound velocity calibration method within a droplet that automatically extracts A-scan signal key points. By utilizing reflection mode ultrasonic information, the method achieves visualization of the size and spatial position distribution of bubbles and droplets in an oil-gas-water three-phase medium in an opaque pipeline, identification of bubbles and droplets, measurement of bubble positions, measurement of droplet size and position, and calibration of the liquid sound velocity within a droplet.

[0063] First, the basic aspects of the present invention will be described.

[0064] The present invention provides an ultrasonic A-scan signal assisted B-mode imaging (AS-assisted-BI) method for detecting droplets and bubbles in an oil-gas-water three-phase medium. The ultrasonic phased array UPA probe used for detection comprises M array elements, wherein the transmitting array element for synthesizing ultrasonic plane waves is Tx, and the array element for receiving ultrasonic waves is Rx. A two-dimensional Cartesian coordinate system xoy is established with the center of the measured field as the origin. The diameter of the bubble and droplet is D, and the coordinates of the center position are (x s ,y s ). The basic scheme of the present invention is as follows:

[0065] (1) According to the different detection requirements and conditions for bubbles and droplets in the oil-gas-water three-phase medium in actual industrial scenarios, one of the following two detection schemes of the AS-assisted BI method is selected for implementation. Scheme 1: When low-time detection is urgently needed and the conditions for preparing a test model for calibrating the oil phase sound velocity are available: the BI method is used to visualize bubbles and droplets, and the AS method is used to measure bubbles and droplets. Scheme 2: When the low-time detection requirement can be compromised, the robustness of droplet measurement is required, and the conditions for preparing a test model for calibrating the oil phase sound velocity are not available: the BI method is used to visualize bubbles and droplets and measure droplets, and the AS method is used to measure bubbles.

[0066] (2) All Rx array elements of the ultrasonic phased array UPA sensor receive the full waveform signal of the ultrasonic plane wave reflected by the oil-gas-water three-phase medium in the measured field, and complete the acquisition of the ultrasonic A-scan signal.

[0067] (3) Using the ultrasonic A-scan signal, the B-mode image of the oil-gas-water three-phase medium is constructed according to the ultrasonic plane wave imaging (PWI) method based on the time domain delay summation (DAS) algorithm.

[0068] The method for constructing the B-mode image of the oil-gas-water three-phase medium in step (3) is: Among them, P (x,y) is the pixel value of the image, function τ i (u) is the full waveform A-scan signal of the i-th Rx array element, u is the time sampling point sequence, and Δ is the time delay factor related to the pixel position and the distance between Rx array elements.

[0069] (4) Identify bubbles and droplets from the B-mode image of the oil-gas-water three-phase medium.

[0070] (5) Based on the detection scheme selected in step (1), decide whether to execute steps (5) to (7). Based on step (4), capture B-mode images of droplets in the oil-gas-water three-phase medium one by one.

[0071] (6) Positioning the horizontal coordinates of the four key points representing the two pixel strips on the front and back surfaces in the B-mode image of the droplet: the top key point on the front surface x fst , key point x at the bottom of the front surface fse , the key point x at the top of the back surface bst , key point x at the bottom of the rear surface bse .

[0072] (7) Calculate the droplet center coordinate x according to the four key points in the droplet B-mode image s , bubble diameter D.

[0073] (8) Identify bubbles and droplets from the A-scan signal of the oil-gas-water three-phase medium.

[0074] (9) Based on the solution selected in step (1), decide whether to execute steps (9) to (11) as appropriate. On the basis of step (8), intercept the A-scan signal of each bubble in the oil-gas-water three-phase medium one by one.

[0075] (10) Locate the unique key point u in the A-scan signal of each bubble, representing the peak position of the front surface of the bubble fs .

[0076] (11) The unique key point u in each bubble signal located according to step (10) fs, calculate the bubble front surface position coordinate x fs .

[0077] (12) Depending on the selected scheme according to step (1), decide whether to perform steps (13) to (18) or not. Prepare a test model for calibrating the sound speed V liquid of the oil phase medium in the liquid droplet.

[0078] (13) Collect the A-scan signal of the test model for calibrating the sound speed V liquid of the oil phase medium in the liquid droplet.

[0079] (14) Identify and locate the 2 key points of the A-scan signal of the calibration model: u fs and u bs representing the front and back surface positions of the liquid cavity respectively.

[0080] (15) Calculate the sound speed V liquid of the liquid to be calibrated.

[0081] (16) On the basis of step (8), intercept the A-scan signal of each liquid droplet in the oil-gas-water three-phase medium one by one.

[0082] (17) Locate the 2 key points u fs and u bs of the "peak" representing the front surface position in the A-scan signal of each liquid droplet.

[0083] (18) According to the 2 key points u fs and u bs located in the A-scan signal of each liquid droplet according to step (17), and in combination with the calibrated sound speed V liquid of the liquid in the liquid droplet, calculate the liquid droplet center coordinate x s and bubble diameter D.

[0084] The ultrasonic A-scan signal assisted B-mode imaging method for detecting liquid droplets and gas bubbles in an oil-gas-water three-phase medium according to the present application will be described below in combination with the accompanying drawings and examples.

[0085] Figure 1 The flow chart of the implementation of the present application is shown in the figure, and the example includes the following specific steps:

[0086] (1) According to the different detection requirements and conditions of bubbles and droplets in oil-gas-water three-phase medium in actual industrial scenes, one of the following two detection schemes of AS-assisted-BI method is selected for implementation. Scheme one: when low-time consumption detection is urgently needed, and the conditions for preparing a test model for calibrating the sound speed of the oil phase are met: bubbles and droplets are visualized by BI method, and bubbles and droplets are measured by AS method. Scheme two: low-time consumption detection requirement can be compromised, droplet measurement robustness requirement is high, and there is no condition to calibrate the sound speed: bubbles and droplets are visualized by BI method, and droplets are measured, and bubbles are measured by AS method.

[0087] (2) Ultrasonic A-scan signal acquisition: The UPA sensor installation structure, the detected object and its field of the UPA industrial process parameter detection simulation system used in the present application are as shown in Figure 2 . Among them, the UPA sensor has M=64 array elements, and the field width W=50 mm. In order to verify that the BI method can visualize the oil-gas-water three-phase medium and identify the bubbles and droplets therein, the system is used to collect A-scan signals of the embodiment in which the bubbles and droplets are located in the same measured field. In order to compare and verify the accuracy of the AS-assisted-BI method for measuring bubbles and droplets, the UPA industrial process parameter detection hardware system with M=16 array elements of the UPA sensor is used to collect A-scan signals of a bubble or a droplet located in the measured field as an ROI (region of interest) in the oil-gas-water three-phase medium, as shown in Figure 2 . Among them, the field width W=50 mm. The A-scan signal collection method is: a UPA probe with M array elements is installed on the left side of the detected field, the transmitting array elements for synthesizing plane waves are Tx, and the array elements for receiving ultrasonic waves are Rx. The detected medium (including an oil-gas-water three-phase medium model, or a bubble or droplet model in water for simulating a certain ROI) is placed in the corresponding UPA simulation system or hardware system. The excitation-reception strategy of the system is set as: all Tx array elements are excited to synthesize ultrasonic single plane waves, and all Rx array elements collect ultrasonic reflection mode time-domain full waveform A-scan signals τ i (u), where i is the serial number of the Rx array element, and u is the time sampling point sequence. Among them, a two-dimensional Cartesian coordinate system is established with the center of the measured field as the origin O, and the center position of the bubble and the droplet is (x s , y s ). When the oil-gas-water three-phase medium model is used as an embodiment, the bubble D=5 mm, (x s , y s )=(5, 5), the droplet D=8 mm, (x s , y s)=(-5,10). When a bubble or droplet model is used to simulate a certain ROI in water, the bubble and droplet diameter D=3, 6, 7, 8, 9, 10 mm, and the center position x s =-12,-8,-4,0,4,8,12,16mm,y s =0mm.

[0088] (3) Use the BI method to construct the B-mode image of the oil-gas-water three-phase medium and identify bubbles and droplets: the A-scan signal τ of all M Rx array elements of the oil-gas-water three-phase medium i (u) Implementation: Among them, P (x,y) is the pixel value of the image, function τ i (u) is the full waveform A-scan signal of the i-th Rx array element, u is the time sampling point sequence, and Δ is the time delay factor related to the distance between the pixel position and the Rx array element. Figure 4 For D=5mm, (x s ,y s )=(5,5) bubble and D=8mm, (x s ,y s )=(-5,10) The B-mode imaging results of the oil-gas-water three-phase medium by the BI method. In this B-mode image, the image containing only one pixel band with a maximum pixel value close to 1 is identified as a bubble. This pixel band represents the spatial position x of the bubble front surface in the measured field. fs The image containing two pixel bands, where the maximum pixel value is much less than 1, is identified as a droplet. These two pixel bands represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bs .

[0089] (4) First, referring to the method of step (3), on the basis of step (2), the B-mode image of each droplet is constructed using the A-scan signals of all the embodiments of the "water droplet model for simulating a certain ROI" mentioned in step (2). Then, the pixels of the image are searched one by one in the order of position from top to bottom and then from left to right. For each droplet, the horizontal coordinates of the four key points are located in sequence, i.e., the key point x at the top of the front surface and the key point x at the bottom of the front surface. fst , key point x at the bottom of the front surface fse , the key point x at the top of the back surface bst , key point x at the bottom of the rear surface bse In the embodiment, the dimension D=6mm, the center position (x s ,y s )=(0,0) as an example, the result diagram of the four key points located in the B-mode image is as follows Figure 5 shown.

[0090] (5) Based on step (4), calculate the bubble diameter D and center position x of the droplet s , the method is: D=|x fs -x bs |, where x fs and x bs are the horizontal coordinates of the front and back surfaces of the droplet, respectively. Then, The absolute error AE is used as the metric BI method to detect the bubble diameter D and center position x of the droplet s The accuracy index. The AE algorithm is: AE = |ε-ε * |=nλ, where ε and ε * are theoretical and measured values, and the parameter n is the coefficient of the ultrasonic wavelength λ. For the simulation and experimental measurement results of all droplet embodiments, the AE value is calculated. The BI method measures the droplet size D and the central longitudinal position x s The box plot statistics of the absolute error AE value are as follows: Figure 6 shown.

[0091] (6) Based on the detection scheme selected in step (1), decide whether to execute steps (6) to (7) as appropriate. For the A-scan signals of all the embodiments of the "bubble model in water for simulating a certain ROI" mentioned in step (2), use the AS method to identify the bubble and locate its unique key point in the A-scan signal. On the basis of step (2), use the A-scan signals of all the embodiments of the "bubble model in water for simulating a certain ROI" mentioned in step (2) to identify each bubble in turn, and the method is: each bubble signal contains only one "peak". The transit time TOF corresponding to the "peak" characterizes the spatial position x of the front surface of the bubble in the measured field fs The amplitude of the "peak" in the bubble signal is much larger than any "peak" value of the droplet. Then, the unique key point u in the A-scan signal of each bubble is located in turn. fs , the method is: key point u fs The position of the unique “peak” of the A-scan signal of the bubble. i (u)i∈[1,M], implement G bubble ={u|A ext [τ i (u)]},H bubble ={u|u=mT}(m∈N + ), {u fs}=U bubble =G bubble ∩H bubble Among them, G bubbleis the local maximum value of the A-scan signal amplitude A ext The collection of [·], H bubble is the set of ultrasonic pulse intervals, T = 1 / f is the pulse interval period of the excitation, m is the number of pulse intervals, U bubble is the intersection of the above two sets. In the embodiment, the size D = 6mm, the center position (x s ,y s )=(0,0) as an example, the A-scan signal is automatically identified as the only "peak" and the result diagram of the unique key point is as follows: Figure 7 shown.

[0092] (7) Based on step (6), calculate the horizontal coordinate x of the front surface position of the bubble fs , the method is: where x fsj is the front surface position of the bubble calculated based on the A-scan signal of a bubble of a certain Rx array element, V water is the speed of sound in water, is the TOF value corresponding to the ufs moment in the A-scan signal. Finally, the longitudinal position x of the front surface of each bubble fs By Collection All x between the upper quartile Q1 and the lower quartile Q3 fsj Same as step (5), the absolute error AE is used as the horizontal coordinate x of the front surface position of the bubble detected by AS method. fs The same as step (5), the absolute error AE is used as the metric to measure the front surface position x of the bubble by the AS method. fs The AE value is calculated for the simulation and experimental measurement results of all bubble embodiments. The AS method measures the horizontal coordinate x of the front surface position of the bubble. fs The box plot statistics of the absolute error AE value are as follows: Figure 8 shown.

[0093] (8) According to the detection scheme selected in step (1), decide whether to perform steps (8) to (12) as appropriate. oil ,The hardware system is used to collect the A-scan signal of the prepared liquid sound velocity calibration model. The schematic diagram of the sound velocity calibration model of industrial white oil in the prepared droplet is as follows Figure 9 The method for collecting the A-scan signal for calibrating the sound velocity model of industrial white oil in the droplet is consistent with step (2).

[0094] (9) Based on step (8), the A-scan signals from the (1+M / 4)th to (MM / 4)th Rx array elements are used to calibrate the sound velocity V of the industrial white oil in the droplet. oilEach A-scan signal contains two "peaks", which respectively represent the spatial positions of the front and rear surfaces of the rectangular cavity filled with the liquid to be calibrated in the measured field. Then, the two key points u of these two "peaks" are located. fs and u bs , the method is: for any of the above Rx array elements A scan signal τ i (u)i∈[1+4 / M,MM / 4], implement G oil ={u|A ext [τ i (u)]},H oil ={u|u=mT}(m∈N + ), {u fs ,u bs}=U oil =G oil ∩H oil Among them, G oil is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H oil is the set of ultrasonic pulse intervals, T = 1 / f is the pulse interval period of the excitation, m is the number of pulse intervals, U oil is the intersection of the above two sets. oil = 20mm liquid sound velocity calibration model is an embodiment, and the two "peaks" of the A-scan signal are automatically identified, and the result schematic diagram of the two key points is as follows: Figure 10 shown.

[0095] (10) Based on step (9), calculate the sound velocity V of the oil to be calibrated oil , the method is in and are u in the i-th A-scan signal fs and u bs The TOF value corresponding to the moment, k is the total number of A-scan signals, L oil is the length of the rectangular cavity filled with oil. The experimental test results show that at normal temperature and pressure, the sound velocity V of the industrial white oil in the droplet is oil It is 1428m / s, which is almost consistent with the theoretical value of 1421m / s.

[0096] (11) Based on step (2), the A-scan signals of all the embodiments of the “water droplet model for simulating a certain ROI” mentioned in step (2) are used to identify each droplet in turn, in the following way: each droplet signal contains two “peaks”. These two “peaks” respectively represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bsThe amplitude of the "peak" in the droplet signal is much smaller than any "peak" value of the bubble. Then, two key points u fs and u bs in the A-scan signal of each droplet are located in turn fs The key points u bs correspond to the positions of the two "peaks" of the A-scan signal of the droplet. For the A-scan signal τ i of a certain droplet of any Rx element, G droplet = {u | A ext [τ i (u)]} and H droplet = {u | u = mT} (m ∈ N + ), {u fs , u bs} = U droplet = G droplet ∩ H droplet . Where G droplet is the set of local maximum values of the A-scan signal amplitude A ext [·], H droplet is the set of ultrasonic pulse intervals, T = 1 / f is the pulse interval period of excitation, m is the pulse interval number, and U droplet is the intersection of the above two sets. In the embodiment, for a droplet with a size D = 6 mm and a center position (x s , y s ) = (0, 0), the A-scan signal of the droplet is automatically identified as two "peaks", and the results of locating the two key points are shown in Figure 11 .

[0097] (12) Based on step (11), the bubble diameter D of each droplet is calculated as follows: Where TOF and TOF are the TOF values corresponding to u fs and u bs in the ith A-scan signal, and k is the total number of A-scan signals. D j is the bubble diameter of a certain droplet of a certain Rx element calculated from the A-scan signal of the droplet, and V oil is the sound speed of the oil in the droplet. Finally, the bubble diameter D of the droplet is calculated by the average value of all D j between the upper quartile Q1 and the lower quartile Q3 of the set {D j}. Then, the center coordinates x s of the droplet are calculated as follows: Where x sj is the center position of a certain droplet of a certain Rx element calculated from the A-scan signal of the droplet, and V wateris the speed of sound of water. Finally, the center position of the droplet x s By the set {x s All x between the upper quartile Q1 and the lower quartile Q3 sj Same as step (5), the absolute error AE is used as the metric to measure the bubble diameter D and center position x of the droplet using the AS method. s As in step (5), the absolute error AE is used as the metric to measure the droplet size D and the center longitudinal position x. s The AE value is calculated for the simulation and experimental measurement results of all bubble embodiments. The droplet size D and the center longitudinal position x are measured by AS method. s The box plot statistics of the absolute error AE value are as follows: Figure 12 shown.

[0098] 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 it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. An ultrasonic A-scan signal-assisted B-mode imaging method for detecting droplets and bubbles in an oil-gas-water three-phase medium. Assume that the ultrasonic phased array (UPA) sensor used for detection has a total of M array elements, the transmitting array element for synthesizing ultrasonic plane waves is Tx, and the array element for receiving ultrasonic waves is Rx. A two-dimensional Cartesian coordinate system xoy is established with the center of the measured field as the origin. The bubble diameter of the bubble and droplet is D, and the coordinates of the center position are (x s ,y s ), ultrasonic imaging of droplets and bubbles in oil-gas-water three-phase medium is realized in two cases, The first method, which requires a low-time test and is equipped with a test model for calibrating the oil-phase sound velocity, uses the BI method to visualize bubbles and droplets and the AS method to measure them. This method includes the following steps: All Rx array elements of the S11 ultrasonic phased array UPA sensor receive the full-waveform time-domain signal of the ultrasonic plane wave reflected by the oil-gas-water three-phase medium in the measured field, completing the acquisition of the ultrasonic A-scan signal; S12 uses ultrasonic A-scan signals and ultrasonic plane wave imaging (PWI) method based on the time domain delay sum (DAS) algorithm to construct B-mode images of the oil-gas-water three-phase medium. S13 identifies bubbles and droplets from the B-mode image of the oil-gas-water three-phase medium; S14 identifies bubbles and droplets from the A-scan signal of oil-gas-water three-phase medium; S15 intercepts the A-scan signal of each bubble in the oil-gas-water three-phase medium one by one. Each bubble signal contains only one "peak". The transit time TOF corresponding to the "peak" represents the spatial position x of the front surface of the bubble in the measured field. fs , locate the unique key point u of the "peak" representing the position of the bubble front surface in each bubble A scan signal fs ; S16 is the only key point u in each bubble signal located in step S15 fs , calculate the horizontal coordinate x of the front surface position fs ; S17 preparation is used to calibrate the sound velocity V of the oil phase medium in the droplet liquid The test model is used to collect the A-scan signal of the test model and calculate the sound velocity V of the oil phase medium to be calibrated. liquid ; S18 intercepts the A-scan signal of each droplet in the oil-gas-water three-phase medium one by one based on step S14. Each droplet signal contains two "peaks", which represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bs , the amplitude of the "peak" in the bubble signal is greater than the amplitude of any "peak" of the droplet; S19 locates the two key points u representing the "peak" of the front surface position in the A-scan signal of each droplet fs and u bs , according to the two key points u in each droplet signal located fs and u bs , combined with the calibrated sound velocity V of the liquid inside the droplet liquid , calculate the droplet center coordinate x s , bubble diameter D; The second method, when robustness of droplet measurement is required and a test model for calibrating the oil-phase acoustic velocity is not available, is to use the BI method to visualize bubbles and droplets and measure droplets, and the AS method to measure bubbles. This method includes the following steps: All Rx array elements of the S21 ultrasonic phased array UPA sensor receive the full-waveform time-domain signal of the ultrasonic plane wave reflected by the oil-gas-water three-phase medium in the measured field, completing the acquisition of the ultrasonic A-scan signal; S22 uses ultrasonic A-scan signals and ultrasonic plane wave imaging (PWI) method based on the time domain delay sum (DAS) algorithm to construct B-mode images of the oil-gas-water three-phase medium. S23 identifies bubbles and droplets from the B-mode image of the oil-gas-water three-phase medium; S24 intercepts the B-mode image of the droplet in the oil-gas-water three-phase medium one by one, traverses all pixels of the droplet B-mode image, and locates the horizontal coordinates of the four key points of the two pixel strips representing the front and back surfaces of each droplet: the key point at the top of the front surface x fst , key point x at the bottom of the front surface fse , the key point x at the top of the back surface bst , key point x at the bottom of the rear surface bse ; S25 calculates the droplet center coordinate x based on the four key points in the droplet B-mode image. s , bubble diameter D; S26 identifies bubbles from the A-scan signal of the oil-gas-water three-phase medium; S27, based on step S23, intercepts the A-scan signal of each bubble in the oil-gas-water three-phase medium one by one; S28 locates the unique key point u in the A-scan signal of each bubble, representing the peak position of the bubble front surface. fs ; S29 is based on the unique key point u in each bubble signal located in S28 fs , calculate the position coordinate x of the bubble front surface fs .

2. The ultrasound A-scan signal-assisted B-mode imaging method according to claim 1, characterized in that: The identification methods of bubbles and droplets from the B-mode images of oil-gas-water three-phase medium in S13 and S23 are both: each bubble image contains only one pixel band, which represents the spatial position x of the bubble front surface in the measured field. fs Each droplet image contains two pixel bands, which represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bs .

3. The ultrasound A-scan signal assisted B-mode imaging method according to claim 1, characterized in that: The method used by S14 to identify bubbles and droplets from the A-scan signal of the oil-gas-water three-phase medium is as follows: Each bubble signal contains only one "peak", and the transit time TOF corresponding to the "peak" represents the spatial position x of the front surface of the bubble in the measured field. fs ; Each droplet signal contains two "peaks", which represent the spatial position x of the front and rear surfaces of the droplet in the measured field. fs and x bs , the amplitude of the "peak" in the bubble signal is greater than the amplitude of any "peak" of the droplet.

4. The ultrasound A-scan signal assisted B-mode imaging method according to claim 1, characterized in that: In S15, the unique key point u representing the "peak" position of the front surface of the bubble is located in the A-scan signal of each bubble. fs The method is: the unique key point u fs The position of the unique "peak" of the A-scan signal corresponding to the bubble; the A-scan signal τ of a bubble of any Rx array element i (u)i∈[1,M], find G bubble ={u|A ext [τ i (u)]},H bubble ={u|u=mT}(m∈N + ), {u fs }=U bubble =G bubble ∩H bubble , where G bubble is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H bubble is the set of ultrasonic pulse intervals, m is the number of pulse intervals, T = 1 / f is the pulse interval period of the excitation, U bubble is the intersection of the above two sets.

5. The ultrasound A-scan signal assisted B-mode imaging method according to claim 1, characterized in that: In S16, the horizontal coordinate x of the front surface position of a bubble is determined. fs The method is: calculate where x fsj is the front surface position of the bubble calculated based on the A-scan signal of the bubble of the jth Rx array element, V water is the speed of sound in water, u is the A-scan signal fs TOF value corresponding to the moment; The horizontal coordinate x of the front surface position of this bubble fs By Collection All x between the upper quartile Q1 and the lower quartile Q3 fsj The average value is calculated using the formula:

6. The ultrasound A-scan signal assisted B-mode imaging method according to claim 1, characterized in that: The method of S17 is: 1) Preparation for calibration of the sound velocity V of the oil phase medium in the droplet liquid The test model is as follows: an agar block containing a rectangular cavity is prepared. The density and sound velocity of the agar block are close to those of water. The center of the rectangular cavity coincides with the center of the agar block, and the length of the rectangular cavity is greater than the total length of all elements of the UPA sensor. The UPA sensor is installed directly opposite the rectangular cavity, and the cavity is filled with the oil phase medium whose sound velocity is to be calibrated. 2) The A-scan signals from the (1+M / 4)th to (MM / 4)th Rx elements are used to calibrate the sound velocity V of the oil phase medium in the droplet liquid Each A-scan signal contains two "peaks", which respectively represent the spatial positions of the front and rear surfaces of the rectangular cavity filled with the oil phase medium to be calibrated in the measured field. The two key points u of these two "peaks" are fs and u bs The positioning method is: for any of the above Rx array elements, the A scan signal τ i (u)i∈[1,M], find G liquid ={u|A ext [τ i (u)]},H liquid ={u|u=mT}(m∈N + ), {u fs ,u bs }=U liquid =G liquid ∩H liquid , where G liquid is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H liquid is the set of ultrasonic pulse intervals, T = 1 / f is the pulse interval period of the excitation, m is the number of pulse intervals, U liquid is the intersection of the above two sets; 3) Calculate the sound velocity V of the oil phase medium to be calibrated liquid The calculation method is in and are u in the i-th A-scan signal fs and u bs The TOF value corresponding to the moment, k is the total number of A-scan signals, L liquid is the length of the rectangular cavity.

7. The ultrasound A-scan signal assisted B-mode imaging method according to claim 1, characterized in that: The method of step S19 is: 1) Locate the two key points u representing the "peak" of the front surface position in the A-scan signal of each droplet fs and u bs : Key point u fs and u bs They correspond to the positions of the two "peaks" of the A-scan signal of the droplet; the A-scan signal τ of a droplet of any Rx array element i (u)i∈[1,M], find G droplet ={u|A ext [τ i (u)]},H droplet ={u|u=mT}(m∈N + ), {u fs ,u bs }=U droplet =G droplet ∩H droplet , where G droplet is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H droplet is the set of ultrasonic pulse intervals, T = 1 / f is the pulse interval period of the excitation, m is the number of pulse intervals, U droplet is the intersection of the above two sets; 2) Combined with the calibrated sound velocity V of the liquid in the droplet liquid , calculate the droplet center coordinate x s , bubble diameter D: in and are u in the i-th A-scan signal fs and u bs The TOF value corresponding to the moment, k is the total number of A-scan signals, L liquid is the length of the rectangular cavity; D j is the droplet size calculated based on the A-scan signal of a droplet of a certain Rx array element, V oil is the sound velocity of the oil phase medium in the droplet; finally, the size D of any droplet is determined by the set {D j All D between the upper quartile Q1 and the lower quartile Q3 j Calculate the average value of 3) Calculate the droplet center coordinate x s : where x sj is the droplet center position calculated based on the A-scan signal of a droplet of a certain Rx array element, V water is the speed of sound of water; finally, the center position of the droplet x s By the set {x s All x between the upper quartile Q1 and the lower quartile Q3 sj Calculate the average value of .

8. The ultrasound A-scan signal-assisted B-mode imaging method according to claim 1, characterized in that: Step S26: The method for identifying bubbles from the A-scan signal of the oil-gas-water three-phase medium is as follows: each bubble signal contains only one "peak", and the transit time TOF corresponding to the "peak" represents the spatial position x of the front surface of the bubble in the measured field. fs .

9. The ultrasound A-scan signal assisted B-mode imaging method according to claim 1, characterized in that: In S28, the unique key point u in the A-scan signal of any bubble fs The positioning method is: the unique key point u fs The position of the unique "peak" of the A-scan signal corresponding to the bubble; the A-scan signal τ of a bubble of any Rx array element i (u)i∈[1,M], find G bubble ={u|A ext [τ i (u)]},H bubble ={u|u=mT}(m∈N + ), {u fs }=U bubble =G bubble ∩H bubble , where G bubble is the local maximum value of the A-scan signal amplitude A ext The collection of [·], H bubble is the set of ultrasonic pulse intervals, m is the number of pulse intervals, T = 1 / f is the pulse interval period of the excitation, U bubble is the intersection of the two sets above.

10. The ultrasound A-scan signal assisted B-mode imaging method according to claim 1, characterized in that: In S29, the horizontal coordinate x of the front surface position of a bubble is determined. fs The method is: calculate where x fsj is the front surface position of the bubble calculated based on the A-scan signal of the bubble of the jth Rx array element, V water is the speed of sound in water, u is the A-scan signal fs TOF value corresponding to the moment; The horizontal coordinate x of the front surface position of this bubble fs By Collection All x between the upper quartile Q1 and the lower quartile Q3 fsj The average value is calculated using the formula:

Images