Complex fracture fluid-proppant diversion flow diversion three-dimensional velocity method

Abstract

The application discloses a kind of complex fracture fracturing fluid-proppant diversion three-dimensional velocity measurement methods, method includes: (1) install complex fracture simulation experiment device;(2) set laser transmitter position, adjust the angle and focal length of first camera and second camera, carry out flow field calibration;(3) image acquisition, synchronous trigger laser transmitter, first camera and second camera image acquisition, image enhancement is carried out to the image acquisition and separates;(4) proppant particle and tracer particle are positioned;(5) adjacent two frames of images are matched, and the missing particle is completed, and the three-dimensional velocity measurement value of particle field is obtained.The application provides a kind of complex fracture fracturing fluid-proppant diversion velocity measurement method based on PIV / PTV technology, with the advantages of strong anti-interference, high accuracy.

Description

Technical Field

[0001] This invention relates to the field of hydraulic fracturing technology, specifically to a three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid and proppant in complex fractures. Background Technology

[0002] Unconventional oil and gas resources are crucial for alleviating my country's energy supply and demand imbalance and achieving strategic energy succession. Their complex geological characteristics, such as low porosity and low permeability, necessitate hydraulic fracturing technology for large-scale, efficient extraction. Hydraulic fracturing involves pumping high-pressure proppant-carrying fluid from the surface into the formation, creating multiple artificial high-speed oil and gas channels within the reservoir. The fracturing fluid flows through the fractures, carrying proppant to deeper layers. Due to its own gravity and the resistance of the fracture walls, the proppant settles and accumulates within the fractures, forming sand dikes to open the artificial fractures. After pumping stops, the proppant remains within the fractures to maintain their openness. However, unconventional oil and gas reservoirs, after hydraulic fracturing, form fracture networks. The complex fracture structure causes the proppant to flow with the fracturing fluid to different secondary fractures. Since the productivity of oil and gas wells is largely related to the migration and placement of proppant within these secondary fractures, studying the movement of proppant at complex fracture nodes is crucial.

[0003] Currently, existing experimental setups simulating proppant transport in complex fractures extract the characteristics of real underground fracture morphology. These setups are designed using a visualization slab and different parameters, such as the angle and width ratio of primary and secondary fractures. Most researchers studying proppant transport and placement within complex fractures employ qualitative analysis of sandbank placement morphology. By studying the evolution and final placement of sandbanks within primary and secondary fractures under different conditions, qualitative patterns of proppant transport and settlement in complex fractures are derived. In 2009, Adam Dayan investigated proppant transport within multi-stage fractures by adding a branch fracture parallel to the primary fracture. (Adam Dayan, Shaun M. Stracener, and Peter E. Clark. 2009. “Proppant Transport in Slick-Water Fracturing of Shale-GasFormations|SPE Annual Technical Conference and Exhibition|OnePetro.” 2009.)

[0004] (https: / / onepetro.org / SPEATCE / proceedings / 09ATCE / All-09ATCE / SPE-125068-MS / 147307). Sahai then increased the number of secondary fractures in the experimental setup to replicate the complexity of real fractures, and investigated the proppant placement process in orthogonal multi-level fractures (Sahai, Raki. 2014).

[0005] "Laboratory Results of Proppant Transport in Complex Fracture Systems | SPE Hydraulic Fracturing Technology Conference and Exhibition | OnePetro." 2014. https: / / onepetro.org / SPEHFTC / proceedings / 14HFTC / All-14HFTC / SPE-168579-MS / 211036?searchresult=1). The movement state of proppant carried by fracturing fluid when entering secondary fractures determines its subsequent migration and placement within those fractures. Proppant that does not flow into the secondary fracture will change its original movement state and continue to move in the primary fracture. Therefore, the migration and placement process of proppant in complex fractures can be decomposed into movement within the primary fracture directly connected to the wellbore, turning movement at the primary-secondary fracture junction, and movement within another single fracture after entering the secondary fracture. The process of fracturing fluid and proppant turning into the fracture lacks precise quantitative characterization. Most scholars study the behavior of proppant at the junction of primary and secondary fractures through numerical simulation. However, in the current technology, there is no physical experiment to recreate the splitting of fracturing fluid and the microscopic turning motion of proppant at the junction of primary and secondary fractures. If the proppant velocity field can be accurately measured and characterized, we can gain a deeper understanding of the splitting and turning process of fracturing fluid and proppant in complex fractures.

[0006] Patent CN109779593A discloses for the first time a flat-plate fracture device that simplifies the measurement and characterization of fracturing fluid flow field and proppant particle velocity field within fractures. This device incorporates a particle image velocimeter (PIV), using laser illumination of the flow region within the fracture and a CCD camera to capture the transient motion of proppant and tracer particles, thereby measuring the fracturing fluid flow field and proppant particle field. However, this device is a single-slit flat plate device and cannot capture and characterize the fracturing fluid diversion and proppant diversion behavior at complex fracture nodes. Patent CN111119848A incorporates branch fractures to achieve velocity measurement at intersections within complex fractures. However, due to the use of fully transparent acrylic glass panels, multiple reflections occur at the joints after laser illumination, hindering PIV measurement. Furthermore, these patents do not disclose the specific method for three-dimensional PIV testing at branch fracture nodes or how to separate and characterize the fracturing fluid and proppant after image acquisition. Based on this, this invention patent provides a three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid and proppant in complex fractures, based on three-dimensional PIV (particle imaging velocimetry) and three-dimensional PTV (particle tracking velocimetry) technology. Summary of the Invention

[0007] To address the shortcomings of existing methods, the present invention aims to provide a three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppant in complex fractures. This method aims to solve the technical problem of the lack of quantitative understanding of the migration law of proppant at complex fracture nodes during hydraulic fracturing in the prior art, and to provide a new approach and method for the quantitative characterization of fracturing fluid-proppant velocity in complex underground fractures.

[0008] To achieve the above technical objectives, the present invention adopts the following technical solution.

[0009] On one hand, the present invention provides a three-dimensional velocity measurement experimental device for the diversion and splitting of fracturing fluid and proppant in complex fractures. The experimental device includes a main fracture metal frame, a vertical plate, a rotatable channel, a main fracture plate, a top and bottom width strip of the main fracture, a sealing strip, and fasteners. The main fracture metal frame is tightly attached to the left side wall of the main fracture plate, and grooves are formed on the upper and lower sides of the right side wall of the main fracture plate. The sealing strip is placed in the grooves, and the top and bottom width strips of the main fracture are tightly attached to the right side of the grooves to form a sealing effect. The vertical plate is tightly attached to the right side wall of the top and bottom width strips of the main fracture. Symmetrical slots are formed on the main fracture metal frame and the vertical plate, and they are connected and fixed by fasteners to compress and press the various parts between the main fracture metal frame and the vertical plate.

[0010] The main seam crack plate, vertical plate, top seam width strip of the main seam, and bottom seam width strip of the main seam are combined left, right, up and down to form the main seam flow channel.

[0011] A rotatable channel is provided on the right side wall of the upright plate. A slit is opened in the rotatable channel to form a secondary seam flow channel. The rightmost end of the rotatable channel can be connected to the secondary seam crack plate.

[0012] A circular through-hole is located in the center of the upright plate, and a rubber sealing ring is installed over the through-hole. Multiple screw holes are located around the circular through-hole on the upright plate. The left end of the rotatable channel is a cylinder with the same thickness as the upright plate and the same diameter as the circular through-hole. To its right is a circular plate with holes corresponding to the multiple screw holes on the upright plate. Bolts are used to connect and fasten the rotatable channel and the upright plate through the holes and the screw holes on the upright plate. To the right of the circular plate is a component with a certain height that is T-shaped when viewed from above. A longitudinal slit runs through the rotatable channel from left to right, and liquid flows within this slit.

[0013] Furthermore, the angle between the main seam and the branch seam can be changed by loosening and removing the bolts connecting the rotatable channel and the upright plate, then rotating the rotatable channel to a certain angle and tightening the bolts again.

[0014] Furthermore, small holes are opened on the end faces of the main seam outlet P1, main seam outlet P2 and branch seam outlet B1, which can be connected to other crack plates by bolts.

[0015] Furthermore, to reduce interference such as internal space reflections during optical testing, a matte black coating is applied to the left side wall of the upright panel to minimize optical interference.

[0016] Furthermore, both the main seam visualization crack panel and the branch seam crack panel are made of acrylic mirror glass to reduce optical interference, thereby ensuring the accuracy of the high-speed camera's data acquisition.

[0017] Furthermore, a laser can be projected into the crack through the top of the main seam, and then a high-speed camera can be used to capture and collect data to achieve the function of speed measurement.

[0018] Another method, the present invention provides a three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppant in complex fractures, characterized by comprising the following steps:

[0019] (1) Install complex crack simulation experimental device;

[0020] (2) Set the position of the laser emitter, adjust the angle and focal length of the first and second cameras, and perform flow field calibration;

[0021] (3) Acquire images, synchronously trigger the laser emitter, the first camera and the second camera to acquire images, perform image enhancement and separation on the acquired images;

[0022] (4) Locate the proppant particles and tracer particles;

[0023] (5) Perform particle matching on two adjacent frames and complete the actual particles to obtain the three-dimensional velocity measurement value of the particle field.

[0024] Furthermore, step (1) also includes:

[0025] The experimental device is made of transparent material and allows light to pass through from the top. The inner wall of the experimental device is coated with matte black paint. The experimental device contains fracturing fluid, which includes proppant and tracer particles.

[0026] A laser emitter is mounted on top of the experimental device, and a first camera and a second camera are mounted on the front of the experimental device.

[0027] Furthermore, step (2) also includes:

[0028] Set the location of the laser emitter;

[0029] Place the fracturing fluid container made of acrylic glass on top of the experimental apparatus. Fill the fracturing fluid container with fracturing fluid and place the calibration plate into the fracturing fluid container.

[0030] Move the fracturing fluid container until the convex surface of the calibration plate inside the fracturing fluid container coincides with the laser surface;

[0031] Adjust the angles and focal lengths of the first and second cameras until the calibration plates on the acquisition interfaces of the first and second cameras are symmetrically distributed and their calibration centers coincide, thus completing the flow field calibration.

[0032] Furthermore, step (3) also includes:

[0033] The laser emitter, the first camera, and the second camera are simultaneously triggered to acquire images. The acquired images include the first frame, the second frame, the first frame, the second frame, and the second frame.

[0034] The acquired images are enhanced and separated to obtain a first separated image and a second separated image. The first separated image contains only tracer particles, and the second separated image contains only proppant particles.

[0035] Furthermore, step (3) also includes:

[0036] Perform image enhancement;

[0037] The images acquired by the first camera and the images acquired by the second camera are divided into vertical region grids, and the vertical region grids equally divide the images.

[0038] Statistical image exposure total intensity P all With the total number of particles N all Calculate the average exposure intensity P all(average) :

[0039]

[0040] Calculate the maximum exposure threshold P of the image by comparing the total number of particles N1 and the average exposure intensity P of the particles in the top first region of the image. max ,

[0041]

[0042] Image with exposure intensity greater than P max The particles were identified as proppant particles;

[0043] Record enhanced exposure intensity P 增 With reduced exposure intensity P 减 The increase and decrease are recorded as correction ratio coefficients, and the enhanced particle pixels P in the image are... 增 P 减 Perform statistical analysis and store the results in a matrix; then, select pixel values ​​greater than P. max The particles identified are proppant particles, and the remaining particles are tracer particles.

[0044] Furthermore, step (4) also includes locating the proppant particles and tracer particles, including:

[0045] (4.1) Preliminary particle localization

[0046] Traverse the particle image exposure matrix to find the coordinates (x, y) of the first luminous pixel of the nth particle. n,i y n,i )

[0047] (4.2) Obtain the center coordinates of the unsaturated particles using the geometric centroid method

[0048] The formula for calculating the center coordinates is:

[0049]

[0050]

[0051] x center,i —The x-center coordinate of the i-th unsaturated exposed particle, dimensionless;

[0052] y center,i —The y-center coordinate of the i-th unsaturated exposed particle, dimensionless;

[0053] x j —The x-coordinate of the j-th pixel of the i-th unsaturated exposed particle, dimensionless;

[0054] y j—The ordinate of the j-th pixel of the i-th unsaturated exposed particle, dimensionless.

[0055] N i —The total number of pixels in the i-th unsaturated exposure particle;

[0056] (4.3) Obtain the center coordinates of the oversaturated exposure particles by weighted average.

[0057]

[0058]

[0059]

[0060] In the formula X cenite —The x-center coordinate of the i-th oversaturated exposed particle, dimensionless;

[0061] Y center,i —The y-center coordinate of the i-th oversaturated exposure particle, dimensionless;

[0062] P t,i —Total exposure value of all pixels of the i-th oversaturated exposure particle, counts;

[0063] P i,j —The exposure value of the j-th pixel of the i-th oversaturated exposure particle, counts;

[0064] x i,j —The x-coordinate of the j-th pixel of the i-th oversaturated exposure particle, dimensionless;

[0065] y i,j —The ordinate of the j-th pixel of the i-th oversaturated exposure particle, dimensionless.

[0066] Obtain the three-dimensional world coordinates of the particles through three-dimensional particle reconstruction;

[0067]

[0068] Z c f, R 旋转 T 平移 These represent the scaling factor, camera focal length, dual-camera rotation matrix, and dual-camera translation matrix, respectively. (u, v) is the pixel coordinate system, (x, y) is the coordinate system with physical dimensions as coordinates, and (u0, v0) are the origins of the (x, y) coordinate system. 世界 Y 世界 Z 世界 () represents the actual coordinates.

[0069] Furthermore, step (5) also includes:

[0070] The ant colony algorithm uses the particle coordinates in the first frame as the starting coordinates and the particle images in the second frame as the process of ants finding the optimal path, thus completing particle matching between adjacent frames.

[0071] Furthermore, step (5) also includes:

[0072] Missing particles are filled in;

[0073] When adjacent particles are successfully paired, the particle motion angles φ1 and φ2 and displacement distances S1 and S2 in the first and second frame images are calculated. The completed particles should then meet the following two conditions:

[0074] φ1 (x,y,z) ≤φ 补(x,y,z) ≤φ2 (x,y,z) (9)

[0075]

[0076] Calculate the average motion angle and displacement distance of neighboring particles to obtain the motion angle and displacement distance of the complete particle.

[0077] On the other hand, the present invention provides a three-dimensional velocity measurement system for the diversion and splitting of fracturing fluid-proppant in complex fractures, characterized in that it includes:

[0078] An experimental setup for simulating complex cracks;

[0079] A flow field calibration module is used to set the position of the laser emitter, adjust the angle and focal length of the first camera and the second camera, and perform flow field calibration.

[0080] An image acquisition module is used to synchronously trigger the laser emitter, the first camera, and the second camera to acquire images, and to enhance and separate the acquired images.

[0081] A positioning module, which is used to position the proppant particles and tracer particles;

[0082] The velocity measurement module is used to perform particle matching on two adjacent frames of images and to complete the actual particles, thereby obtaining the three-dimensional velocity measurement value of the particle field.

[0083] Furthermore, the experimental device is made of transparent material and allows light to pass through from the top. The inner wall of the experimental device is coated with matte black paint. The experimental device contains fracturing fluid, which includes proppant and tracer particles.

[0084] A laser emitter is mounted on top of the experimental device, and a first camera and a second camera are mounted on the front of the experimental device.

[0085] Furthermore, the similarities between the prediction system of this invention and the prediction method provided by this invention will not be repeated here.

[0086] The beneficial technical effects of this invention are as follows: The experimental device in this invention is made of transparent material and allows light to pass through from the top. The inner wall is coated with matte black paint to reduce optical interference, thereby ensuring the accuracy of high-speed camera acquisition. In addition, the three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppant in complex fractures provided by this invention solves the technical problem of the lack of quantitative understanding of the migration law of proppant at complex fracture nodes during hydraulic fracturing in the prior art, and provides a new idea and method for the quantitative characterization of fracturing fluid-proppant velocity in complex underground fractures. Attached Figure Description

[0087] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0088] Figure 1 This is a side view of the experimental apparatus of the present invention.

[0089] Figure 2 This is a schematic diagram of the main slit flow channel structure of the experimental apparatus of the present invention.

[0090] Figure 3 This is a schematic diagram of the overall connection of the experimental apparatus of the present invention.

[0091] Figure 4 This is a side view of the upright plate of the experimental apparatus of the present invention.

[0092] Figure 5 This is a side view of the rotatable channel of the experimental apparatus of the present invention.

[0093] Figure 6 This is a test schematic diagram of the present invention.

[0094] Figure 7 This is a schematic diagram of the flow field calibration of the present invention.

[0095] Figure 8 This is the original particle image obtained in this invention.

[0096] Figure 9 This is the enhanced particle diagram of the present invention.

[0097] Figure 10 This is an image of the proppant after separation according to the present invention.

[0098] Figure 11 This is an image of the tracer particles after separation according to the present invention.

[0099] Figure 12 This is a three-dimensional particle reconstruction velocity map obtained by the present invention.

[0100] Numbering on the map:

[0101] 1—Main seam metal frame; 2—Upright plate; 3—Main seam crack plate; 4—Main seam top seam width strip; 5—Main seam bottom seam width strip; 6—Sealing strip; 7—Fastener; 8—Symmetrical slots; 9—Rotating channel; 10—Sealing ring; P1, P2—Main seam outlet; B1—Support seam outlet; Detailed Implementation

[0102] like Figures 1-2 As shown, this invention provides a three-dimensional velocity measurement experimental device for the diversion and splitting of fracturing fluid and proppant in complex fractures. The experimental device includes: a main fracture metal frame 1, a vertical plate 2, a main fracture plate 3, a top main fracture width strip 4, a bottom main fracture width strip 5, a sealing strip 6, fasteners 7, symmetrical slots 8, and a rotatable channel 9. The main fracture metal frame 1 is tightly attached to the left side wall of the main fracture plate 3. The right side wall of the main fracture plate 3 has horizontal rectangular grooves at the top and bottom, and the sealing strip 6 is placed in the grooves. The top main fracture width strip 4 and the bottom main fracture width strip 5 are respectively tightly attached to the sealing strip. The right side wall of 6 and the thickness of both cover the thickness of the sealing strip 6. The material of the seam width strip is acrylic glass, thus forming a good sealing effect. The upright plate 2 is closely attached to the right side wall of the top seam width strip 4 and the bottom seam width strip 5 of the main seam. The cavity formed by the left and right combination of the main seam crack plate 3, the upright plate 2, the top seam width strip 4 and the bottom seam width strip 5 of the main seam is the main seam flow channel. A rotatable channel 9 is provided on the right side wall of the upright plate 2. The rotatable channel 9 has a slit inside to form a secondary seam flow channel. The rightmost end of the rotatable channel 9 can be connected to the secondary seam crack plate.

[0103] like Figure 3 As shown, the main seam outlet P1, main seam outlet P2, and branch seam outlet B1 of the device can all be connected to other crack plates to realize the proppant delivery experimental plate device for PIV / PTV measurement, thereby forming a complex crack structure.

[0104] like Figures 4-5As shown, the vertical plate 2 has a circular through hole in the center, and a rubber sealing ring 10 is laid on the through hole. Multiple screw holes are opened around the circular through hole on the vertical plate 2. The left end of the rotatable channel 9 is a cylinder with the same thickness as the vertical plate and the same diameter as the circular through hole. The right side of the cylinder is a circular piece with holes corresponding to the multiple screw holes on the vertical plate 2. Bolts are used to connect and fasten the rotatable channel 9 and the vertical plate 2 through the holes and the screw holes on the vertical plate. The right side of the circular piece is a component with a certain height that is T-shaped when viewed from above. A longitudinal slit runs through the rotatable channel 9 from left to right, and the liquid flows in this slit.

[0105] To simulate different main joint widths, the three-dimensional steering behavior of the proppant at complex crack nodes under different main and secondary joint combinations can be studied by replacing the top and bottom main joint width strips (4 and 5) with different widths. The branch joint width can be varied by replacing different branch joint width strips and branch joint crack plates, greatly reducing the number of replacement components and the cost.

[0106] To simulate the angle change between the main seam and the branch seam, the bolts connecting the rotatable channel and the upright plate can be loosened and removed, then the rotatable channel can be rotated to a certain angle and then screwed back in and tightened.

[0107] When conducting a three-dimensional proppant redirection behavior experiment using the aforementioned proppant delivery and branch joint connection device within a complex fracture, the specific experimental steps are as follows:

[0108] (1) Install complex crack simulation experimental device

[0109] This invention studies the fracturing fluid-proppant diversion behavior at complex fracture nodes based on three-dimensional PIV / PTV technology. First, the required proppant and fluid are prepared, a complex fracture simulation experimental device is assembled, and the support fracture connection position is designed. Then, a fracture plate is connected to the main fracture outlet P1 end of the device, another fracture plate is connected to the main fracture outlet P2 end, and a support fracture plate is connected to the support fracture outlet B1 end. Fasteners 7 achieve overall fastening of the connection device through symmetrical slots 8 opened on the upright plate 2 and the main fracture metal frame 1.

[0110] Existing technologies use fully transparent acrylic glass panels, which can cause multiple reflections at the joints after laser irradiation, thus affecting the accuracy of experimental testing. The experimental device provided by this invention meets the requirements of full visibility of the flow field and full light transmission from the top. The front is fully transparent and polished, while the inner wall is coated with matte black paint to minimize optical interference.

[0111] After the experimental setup is complete, liquid is injected into the apparatus via a pump. Local leaks are addressed by tightening fastener 7. The pumped liquid includes fracturing fluid, proppant, and tracer ions. The tracer particles have a density close to that of the fluid, and their movement represents the flow field. Figure 6 As shown: Pixel-level processing of tracer particle and proppant particle images at fracture intersections was performed using 3D PIV / PTV imaging. The two types of particles were separated by differences in exposure characteristics. Interpolation was used to complete the images of optically obstructed particles. 3D reconstruction of multi-media penetrating particles was achieved by simulating binocular vision using dual cameras. Particle stereo matching was performed using a nearest neighbor optimization algorithm to reconstruct the 3D velocity of tracer particles and proppant particles, thereby obtaining the 3D velocity of the fracturing fluid-proppant split in complex fractures.

[0112] (2) Set the position of the laser emitter, adjust the angle and focal length of the first and second cameras, and perform flow field calibration.

[0113] It should be noted that, since the flow field inside the crack is closed, the calibration plate cannot be placed inside the flow field for camera calibration. Therefore, it is necessary to recreate the actual calibration process at the top of the crack. The specific method is as follows: Figure 7 As shown: (2.1) Set the position of the laser emitter and keep it in adjustment mode; (2.2) Use acrylic glass to splice an inner rectangular groove container, fill the inner rectangular groove container with fracturing fluid, and put the calibration plate into the inner rectangular groove container; (2.3) Slowly move the inner rectangular groove container until the convex surface of the calibration plate in the groove just coincides with the laser surface; (2.4) Adjust the camera angle and focal length until the calibration plates in the acquisition interfaces of the two cameras are symmetrically distributed and the calibration centers coincide, thus completing the calibration.

[0114] (3) Acquire images, synchronously trigger the laser emitter, the first camera and the second camera to acquire images, and perform image enhancement and separation on the acquired images.

[0115] The laser and camera are synchronously triggered, and a two-frame shooting interval is set to enable the dual cameras to continuously and simultaneously capture instantaneous particle images.

[0116] like Figure 8 As shown, the images captured by the dual cameras include the first frame from the first camera, the second frame from the first camera, the first frame from the second camera, and the second frame from the second camera, arranged vertically. By dividing the images into equal-sized segments, four particle images of the same size are obtained (e.g., ...). Figure 9 (As shown).

[0117] Because the camera is not perpendicular to the shooting surface, the angle between the camera and the shooting surface results in only a portion of the viewport being in focus, while another portion is in front of the focal point. This leads to cluttered and unclear particle images in the exposed areas, easily causing multiple particles to clump together after exposure, requiring sharpening processing. In one embodiment, smoothing is performed based on the particle's grayscale to ensure its edge features remain unchanged.

[0118] Since there is a significant difference in the exposure of tracer particles and proppant particles in two frames (a total of four images) from a dual-camera setup, it is necessary to perform particle image enhancement and separation separately for tracer particles and proppant particles to obtain particle images containing only tracer particles and particle images containing only proppant particles. The specific steps are as follows:

[0119] Because the two laser beams have different intensities and the indoor light source distribution is uneven, the images captured by the two cameras have different exposure intensities. Furthermore, to better separate the tracer particles from the proppant particles, the proppant particles need to be enhanced to achieve greater differences in exposure characteristics (taking the first frame from the first camera and the first frame from the second camera as an example). The enhanced image is shown below. Figure 9 As shown. In one embodiment, the enhancement step includes: first, dividing the two images into vertically gridded regions; since the top is more strongly irradiated by the laser, the exposed area and intensity of the proppant in this region are also the largest. The total exposure intensity P on the entire particle image is then calculated. all Total number of particles N all The average exposure intensity P can be obtained. all(average) Since the top receives the most laser illumination, the particles in the lower part experience insufficient illumination due to laser dissipation. In one embodiment, the image can be divided vertically into 10 parts (it should be noted that those skilled in the art can adjust the vertical image pixel size to achieve equal division). The total number of particles N1 in the top first region and the average particle exposure intensity P are used as the maximum exposure threshold P on the image. max All exposure intensities below this area that are greater than P max These can be directly identified as proppant particles.

[0120]

[0121]

[0122] Since the particle exposure intensity in different regions has been corrected, the increased exposure intensity (underexposure) and decreased exposure intensity (overexposure) are recorded, and the increase and decrease are recorded as correction ratio coefficients. The enhanced particle pixel P in the image is then recorded. 增 P 减 The statistics are performed and stored in a matrix; the corrected pixel values ​​are greater than P. max The remaining particles are still identified as proppant particles, while the rest are tracer particles.

[0123] (4) Locate the proppant particles and tracer particles.

[0124] (4.1) Achieve preliminary positioning of all particles

[0125] Traverse the particle image exposure matrix to find the coordinates (x, y) of the first luminous pixel of the nth particle. n,i y n,i ).

[0126] (4.2) The center coordinates of unsaturated particles are obtained by the geometric centroid method;

[0127] Since the unsaturated particles are not fully exposed, they do not exhibit the halo effect caused by overexposure. Therefore, the geometric centroid of the exposed image is the center point of the proppant. The formula for calculating the center coordinates is:

[0128]

[0129]

[0130] x center,i —The x-center coordinate of the i-th unsaturated exposed particle, dimensionless;

[0131] y center,i —The y-center coordinate of the i-th unsaturated exposed particle, dimensionless;

[0132] x j —The x-coordinate of the j-th pixel of the i-th unsaturated exposed particle, dimensionless;

[0133] y j —The ordinate of the j-th pixel of the i-th unsaturated exposed particle, dimensionless.

[0134] N i —The total number of pixels in the i-th unsaturated exposure particle.

[0135] (4.3) The center coordinates of overexposed particles are obtained by weighted averaging.

[0136] For oversaturated particles, due to their strong reflected light, a noticeable halo effect will appear in the particle image, causing the exposed area to deviate from its actual shape. A weighted average of exposure values ​​is used to calculate the center of the oversaturated particles; the formula is as follows:

[0137]

[0138]

[0139]

[0140] In the formula X cenite —The x-center coordinate of the i-th oversaturated exposed particle, dimensionless;

[0141] Y center,i —The y-center coordinate of the i-th oversaturated exposure particle, dimensionless;

[0142] P t,i —Total exposure value of all pixels of the i-th oversaturated exposure particle, counts;

[0143] P i,j —The exposure value of the j-th pixel of the i-th oversaturated exposure particle, counts;

[0144] x i,j —The x-coordinate of the j-th pixel of the i-th oversaturated exposure particle, dimensionless;

[0145] y i,j —The ordinate of the j-th pixel of the i-th oversaturated exposure particle, dimensionless.

[0146] At this point, by separating the enhanced image, we can obtain the separated proppant particle image (e.g., Figure 10 (as shown) and tracer particle images (such as) Figure 11 (As shown).

[0147] After image processing, 3D particle reconstruction needs to be completed: due to dual-camera illumination and the fact that the images are not perpendicular to the shooting plane, the images on the imaging plane contain depth information, and the same particle has a fixed and unique exposure point on the dual cameras. Using the limiting criterion, the 3D world coordinates of the particles can be obtained based on the depth information and the intrinsic and extrinsic parameter matrices of the dual cameras.

[0148]

[0149] Z c f, R 旋转 T 平移 These represent the camera focal length, dual-camera rotation matrix, and dual-camera translation matrix, respectively. (u, v) are the pixel coordinate system, (x, y) is the coordinate system with physical dimensions as coordinates, and (u0, v0) are the origins of the (x, y) coordinate system. 世界 Y 世界 Z 世界 () represents the actual coordinates.

[0150] (5) Perform particle matching on two adjacent frames and complete the actual particles to obtain the three-dimensional velocity measurement value of the particle field.

[0151] Due to the high concentration of proppant during the experiment, an ant colony algorithm was introduced. Each proppant particle was treated as an ant, and the particle coordinates in the first frame image were used as the starting coordinates. The process of finding the optimal path for the ants was represented by the particle images in the second frame image. This initially completed the matching of particles within the two frames.

[0152] However, due to the mutual occlusion of proppant particles, there are situations where particles are not in contact in the first frame, but become mixed together in the second frame due to mutual occlusion. Therefore, it is necessary to complete the missing particles. In one embodiment of the present invention, the pinch principle can be used for completion. When adjacent particles are successfully paired, the motion angles φ1 and φ2 of the two particles and the displacement distances S1 and S2 are calculated. Then, the completed particles should meet the following two conditions:

[0153] φ1 (x,y,z) ≤φ 补(x,y,z) ≤φ2 (x,y,z) (9)

[0154]

[0155] The motion angle and displacement of the completed particles are then compared with the average value of the neighboring particle values.

[0156] φ 补(x,y,z) =0.5(φ1) (x,y,z) +φ2 (x,y,z) (11)

[0157]

[0158] Its particle pixel matrix adopts the particle image matrix corresponding to the first frame, thus completing the three-dimensional particle matching and particle field completion, and completing the drawing of the three-dimensional velocity vector, thereby realizing the test of the three-dimensional velocity of the fracturing fluid-proppant split in complex fractures.

[0159] Although specific embodiments of the present invention have been described in detail with reference to the accompanying drawings, this should not be construed as limiting the scope of protection of this patent. Various modifications and variations that can be made by those skilled in the art without inventive effort within the scope described in the claims still fall within the scope of protection of this patent.

Claims

1. A three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid and proppant in complex fractures, characterized in that, Includes the following steps: (1) Install a complex crack simulation experimental device; the experimental device includes a main seam metal frame, a vertical plate, a rotatable channel, a main seam crack plate, a main seam top crack width strip, a main seam bottom crack width strip, a sealing strip, and fasteners; the main seam metal frame is tightly attached to the left side wall of the main seam crack plate, and grooves are opened on the right side wall of the main seam crack plate. The sealing strip is placed in the groove, and the main seam top crack width strip and the main seam bottom crack width strip are tightly attached to the right side of the groove to form a sealing effect; the vertical plate is tightly attached to the right side wall of the main seam top crack width strip and the main seam bottom crack width strip; the main seam metal frame and the vertical plate have symmetrical slots and are connected and fixed by fasteners to press and tighten, thereby pressing the various parts between the main seam metal frame and the vertical plate; the main seam crack plate, the vertical plate, the main seam top crack width strip, and the main seam bottom crack width strip are combined left and right and up and down to form the main seam flow channel; A rotatable channel is provided on the right side wall of the upright plate. A slit is opened in the rotatable channel to form a secondary seam flow channel. The rightmost end of the rotatable channel is connected to the secondary seam crack plate. The experimental device is made of transparent material and allows light to pass through from the top. The inner wall of the experimental device is coated with matte black paint. The experimental device contains fracturing fluid, which includes proppant and tracer particles. (2) Set the position of the laser emitter, adjust the angle and focal length of the first and second cameras, and perform flow field calibration; (3) Acquire images, synchronously trigger the laser emitter, the first camera and the second camera to acquire images, and perform image enhancement and separation on the acquired images; (4) Locate the proppant particles and tracer particles; (5) Perform particle matching on two adjacent frames and fill in the missing particles to obtain the three-dimensional velocity measurement value of the particle field; The step of performing particle matching on two adjacent image frames includes: Using the ant colony algorithm, the particle coordinates in the first frame image are used as the starting coordinates, and the particle images in the second frame image are used as the process of ants finding the optimal path, thus completing particle matching in adjacent two frames. The process of completing the missing particles includes: When adjacent particles successfully pair up, calculate the particle motion angles in the first and second frame images. , Displacement distance S 1, S 2. The completed particles should meet the following two conditions: （1） （2） Calculate the average motion angle and displacement distance of neighboring particles to obtain the motion angle and displacement distance of the complete particle.

2. The three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppane in complex fractures as described in claim 1, further comprising step (1): A laser emitter is mounted on top of the experimental device, and a first camera and a second camera are mounted on the front of the experimental device.

3. The three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppane in complex fractures as described in claim 2, further comprising step (2): Set the location of the laser emitter; Place the fracturing fluid container made of acrylic glass on top of the experimental apparatus. Fill the fracturing fluid container with fracturing fluid and place the calibration plate into the fracturing fluid container. Move the fracturing fluid container until the convex surface of the calibration plate inside the fracturing fluid container coincides with the laser surface; Adjust the angles and focal lengths of the first and second cameras until the calibration plates on the acquisition interfaces of the first and second cameras are symmetrically distributed and their calibration centers coincide, thus completing the flow field calibration.

4. The three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppane in complex fractures as described in claim 1, further comprising step (3): Simultaneously trigger the laser emitter, the first camera, and the second camera to acquire images. The acquired images include the first frame, the second frame, the first frame, the second frame, and the second frame. The acquired images are enhanced and separated to obtain a first separated image and a second separated image. The first separated image contains only tracer particles, and the second separated image contains only proppant particles.

5. The three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppane in complex fractures as described in claim 4, further comprising step (3): Perform image enhancement; The images acquired by the first camera and the images acquired by the second camera are divided into vertical region grids, and the vertical region grids divide the images equally. Statistical image exposure total intensity P all With the total number of particles N all Calculate the average exposure intensity P all（average） : （3） Statistical analysis of the total number of particles in the first region at the top of the image N 1 and average intensity of particle exposure P Find the maximum exposure threshold of the image. P max , （4） Image with exposure intensity greater than P max The particles were identified as proppant particles; Record increased exposure intensity P 增 With reduced exposure intensity P 减 The increase and decrease are recorded as correction ratio coefficients, and the enhanced particle pixels within the image are... P 增 , P 减 Perform statistical analysis and store the results in a matrix; the corrected pixel values ​​will be greater than... P max The particles identified are proppant particles, and the remaining particles are tracer particles.

6. The three-dimensional velocity measurement method for the diversion and splitting of fracturing fluid-proppant in complex fractures as described in claim 4, wherein step (4) further includes locating the proppant particles and tracer particles, including: (4.1) Preliminary particle localization Traverse the particle image exposure matrix to find the first... n The coordinates of the first luminous pixel of each particle ( x n,i ，y n,i ) (4.2) Obtain the center coordinates of unsaturated particles using the geometric centroid method The formula for calculating the center coordinates is: （5） （6） —No. i Unsaturated exposure particles x Central coordinates, dimensionless; —No. i Unsaturated exposure particles y Central coordinates, dimensionless; —No. i The first unsaturated exposure particle j The x-coordinate of each pixel is dimensionless. —No. i The first unsaturated exposure particle j The y-coordinate of each pixel is dimensionless. —No. i The total number of pixels in the unsaturated exposure particles; (4.3) Obtain the center coordinates of the oversaturated exposure particles by weighted averaging. （7） （8） （9） In the formula —No. i Oversaturated exposure particles x Central coordinates, dimensionless; —No. i Oversaturated exposure particles y Central coordinates, dimensionless; —No. i Total exposure value of all pixels of an oversaturated exposure particle, counts; —No. i The first oversaturated exposure particle j Exposure values ​​per pixel, counts; —No. i The first oversaturated exposure particle j The x-coordinate of each pixel is dimensionless. —No. i The first oversaturated exposure particle j The vertical coordinate of each pixel is dimensionless. Obtain the three-dimensional world coordinates of the particles through three-dimensional particle reconstruction; (10) in Z c , f , R 旋转 , T 平移 These are the scaling factor, camera focal length, dual-camera rotation matrix, and dual-camera translation matrix, respectively. u , v () is the pixel coordinate system, ( x, y () is a coordinate system with physical dimensions as coordinates. u 0 , v 0 )for( x, y The origin of the coordinate system, X 世界 , Y 世界 , Z 世界 () represents the actual coordinates.

7. A three-dimensional velocity measurement system for diverting and splitting fracturing fluid and proppant in complex fractures, applied to the method described in any one of claims 1-6, characterized in that, include: An experimental setup for simulating complex cracks; A flow field calibration module is used to set the position of the laser emitter, adjust the angle and focal length of the first camera and the second camera, and perform flow field calibration. An image acquisition module is used to synchronously trigger the laser emitter, the first camera, and the second camera to acquire images, and to enhance and separate the acquired images. A positioning module, which is used to position the proppant particles and tracer particles; The velocity measurement module is used to perform particle matching on two adjacent frames of images and to complete the actual particles, thereby obtaining the three-dimensional velocity measurement value of the particle field.

8. The complex fracture fracturing fluid-proppant diversion and splitting three-dimensional velocity measurement system as described in claim 7, wherein the experimental device is made of transparent material and the top is light-transmitting, the inner wall of the experimental device is sprayed with matte black paint, the experimental device contains fracturing fluid, and the fracturing fluid includes proppant and tracer particles; A laser emitter is mounted on top of the experimental device, and a first camera and a second camera are mounted on the front of the experimental device.

Images