Identification method and application of proppant in rock debris and mud

Abstract

The invention belongs to the technical field of oil-gas field development fracturing, and provides a method for identifying propping agents in rock debris and mud and application. According to the identification method, sample image particles are analyzed by using sphericity, transparency TR and red-blue difference YCf to identify the proppant; firstly, sphericity analysis is carried out on a sample particle image through Hough circle transform CHT, and solid particles with substandard sphericity are eliminated; secondly, excluding solid particles with substandard transparency through the image transparency TR; then, mineral particles are further eliminated through the red and blue color difference YCf of the image, and finally particle size measurement is conducted to obtain a proppant distribution result. According to the method, proppant particles are identified from a mixture of rock debris and mud, the spatial position and distribution of the proppant in a whole well section are obtained, the migration and paving effects of the proppant are further evaluated, and a visual fact basis is provided for the paving rule of the proppant in a hydraulic fracture under the real stratum condition.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas field development fracturing, and relates to a method for identifying proppants in rock cuttings and mud and an application thereof. Background Art

[0002] Hydraulic fracturing is the most effective method to improve oil reservoir recovery. During the hydraulic fracturing process, a large amount of fracturing fluid is injected into the well at a sufficiently high rate to break the target formation to form artificial fractures. When fractures of a certain size are formed, solid proppants (such as sand or ceramic particles) are injected into the fractures together with the fluid. Once the pumping stops, the fluid pressure decreases under the action of geostress, the fractures close, and the proppants are retained between the fractures to prevent the fractures from closing completely and provide a high conductivity flow channel from the formation to the wellbore. The final placement of the proppant in the fracture plays a key role in the conductivity of the fracture and the production increase effect. Therefore, it is of great significance to study the placement of proppants in fractures.

[0003] In the process of hydraulic fracturing, commonly used proppant particle sizes include 20-40 mesh, 30-50 mesh, 40-70 mesh, 70-100 mesh, etc. The migration and placement of proppants will be affected by many factors such as the properties of the fracturing fluid, the properties of the proppant, the injection displacement, the proppant concentration, temperature, pressure, etc.

[0004] In order to understand the laying state of proppant in the crack, the existing technology mostly uses visualization devices to identify the distribution of proppant, such as China Utility Model Patent CN206129257U discloses a visualization device for the laying law of proppant in hydraulic fracturing network under simulated formation temperature, which realizes the implantation of temperature conditions by setting heat-resistant silica gel between two glass plates simulating the crack wall; China Utility Model Patent CN206903650U discloses a visualization dynamic variable crack width proppant sand laying simulation device, which realizes the proppant delivery simulation under the condition of crack width change under single crack. However, the above devices can only realize the distribution mode of proppant particles, and cannot obtain the distribution of proppant in the actual crack by obtaining cores.

[0005] In addition, Zhang Guodong studied the sedimentation and horizontal migration of proppants in fractures in "Study on the Migration Law of Proppants in Horizontal Well Fracturing under Complex Action Mechanisms", comprehensively considered the sedimentation of solid phase particles, simulated the migration of sand-carrying fluid in fractures, explored the proppant migration stimulus, clarified its laying state in fractures, and analyzed the influence of proppant particles and density, fluid viscosity, and injection parameters on the proppant laying state in fractures; this study provided theoretical guidance for the proppant laying state and gave the influencing factors affecting the proppant laying state, but it was still impossible to obtain the laying situation of the proppant in actual fractures.

[0006] The coring experiment can intuitively grasp the distribution of proppant with the depth of the formation, so as to accurately judge the initiation and extension of hydraulic fracturing cracks, the distribution of proppant, etc. These parameter data can be used to establish a more reliable reservoir model that is close to the actual situation, so as to achieve the purpose of optimizing volume fracturing parameters and improving fracturing effects. However, proppant particles are mixed with the returned rock cuttings and mud particles, which are difficult to identify with the naked eye and ordinary machines. A set of theories and methods are needed to accurately and reliably identify the proppant entering the formation. Summary of the invention

[0007] In view of the problem in the prior art that proppant particles are mixed with returned rock cuttings and mud particles and cannot be identified by the naked eye or ordinary machines, the present invention provides a method and application for identifying proppant in rock cuttings and mud, which can identify proppant particles from the mixture of rock cuttings and mud, obtain the spatial position and distribution of the proppant in the entire well section, and then evaluate the migration and laying effect of the proppant, providing an intuitive factual basis for the laying law of the proppant in hydraulic fractures under real formation conditions.

[0008] The technical solution of the present invention is as follows:

[0009] A method for screening and identifying proppants in rock cuttings and mud, the identification method sequentially uses sphericity, transparency TR and red-blue color difference YC f Three parameters are used to analyze the sample image particles to identify the proppant.

[0010] Furthermore, the method specifically comprises the following steps:

[0011] (1) Preliminary screening: The sphericity of the sample particle image is analyzed by Hough circle transform (CHT) to exclude solid particles that do not meet the sphericity standards;

[0012] (2) Eliminate solid particles that do not meet the transparency standards through image transparency TR;

[0013] (3) Through the red and blue difference YC f Mineral particles are removed to obtain proppant particles.

[0014] First, compared with formation mineral particles, proppant particles have better sphericity. The sphericity of sample particles is analyzed based on the mature Hough circle transform (CHT), so that particles that may be proppants can be preliminarily screened out based on their appearance.

[0015] The Hough circle transform CHT is written using Python software and OpenCV database.

[0016] The Hough circle transform CHT maps non-zero pixel points in the Cartesian coordinate system to the abr coordinate system. When the number of repeated points in the abr coordinate system is ≤ 50% of the total number of non-zero pixel points in the Cartesian coordinate system, the particles are solid particles whose sphericity does not meet the standard.

[0017] Secondly, in view of the fact that some formation mineral particles have better sphericity, in order to exclude some formation mineral particles with better sphericity, after analyzing a large number of samples and standard proppants, it was found that another characteristic of proppant particles is that they have a certain degree of transparency under transparent imaging conditions, while the transparency of other mineral particles is very low or even non-transparency; therefore, transparency is an important indicator for further identifying whether a particle is a proppant.

[0018] The transparency TR is calculated by processing the image with machine vision processing software, using a color range selector, taking the color sample as the background light color, setting the tolerance to 130, and the pixels with a tolerance of less than 130 are used as transparent pixels.

[0019] Furthermore, the transparency TR is the percentage of the number of pixels of the transparent part of the particle to the number of pixels of the entire particle.

[0020] Furthermore, particles with a transparency TR ≤ 20% are particles that do not meet the transparency standard.

[0021] Finally, another feature of the proppant particles in the image is that the proppant particles have a higher blue tone and a lower red tone; while other mineral particles such as feldspar, shale and other particles have a lower blue tone and a higher red tone; therefore, using the red and blue shades of the particles in the image can further exclude some of the mineral particles, making the recognition result more accurate.

[0022] The red and blue color difference YC f The determination process is as follows: first remove the background interference part other than the proppant particles determined in step (1) and step (2), and use the red color level median value R obtained by the color level histogram f and the blue middle value B f The red and blue difference YC f The red color level median value R of the target particle f The blue color scale middle value B f The difference is calculated as follows:

[0023] YC f =R f -B f .

[0024] Furthermore, in step (4), the red and blue color difference YC f Particles with a particle size of ≥30 are mineral particles.

[0025] The present invention also provides application of the above method in identifying proppant in a mixture of rock cuttings and mud.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] (1) The present invention comprehensively considers the three main aspects of sphericity, transparency and color to establish the proppant particle identification standard, effectively excluding formation mineral particles, and the results are accurate and reliable;

[0028] (2) The present invention verifies the sphericity, transparency TR and red-blue color YC of standard proppant and natural quartz sand particles through experiments. f Determine the proppant sphericity, transparency TR and red-blue color YC f This threshold range is more accurate for proppant identification;

[0029] (3) The present invention uses the Hough circle transform (CHT) program to process the sample particle image. Once this method is established, it can be used at almost zero cost. The transparency is then obtained by instruments such as low-magnification stereo microscopes and electronic eyepieces, and the red-blue color difference is obtained by image processing software, and the use and maintenance costs are extremely low. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flow chart of the proppant identification method of the present invention;

[0031] Figure 2 The sphericity recognition result of the embodiment of the present invention;

[0032] Figure 3 The transparency imaging result diagrams of A and particle A, B and particle B, and C and particle C in Example 1 of the present invention are shown in FIG.

[0033] Figure 4 This is a transparency visualization picture of Example A of the present invention, particle A; B, particle B; C, particle C;

[0034] Figure 5 The transparency analysis result of the embodiment of the present invention;

[0035] Figure 6 The color and color scale histogram of the proppant particle 1 according to the embodiment of the present invention;

[0036] Figure 7 The color and color level histogram of the mineral particles 2 according to the embodiment of the present invention;

[0037] Figure 8 This is the red and blue color difference analysis result of the embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following non-limiting examples can enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. The following content is merely an exemplary description of the scope of the present invention, and those skilled in the art can make various changes and modifications to the present invention based on the disclosed content, which should also fall within the scope of the present invention.

[0039] Take 2 samples to be tested, one for theoretical calculation and the other for identification of proppant using the method of the present invention;

[0040] The samples to be tested were processed as follows: 300g of rock cuttings were taken from each section, washed, dried and weighed. The contents of 20-40 mesh and 40-70 mesh particles in the mud were about 0.33‰ and 1.6‰, respectively. The nine-point method was used to sample and the final observation sample was controlled to be about 0.1g. The average volume density of 20-40 mesh and 40-70 mesh quartz sand proppants was 1.53 and 1.49g / cm, respectively. 3 , let the average volume density of the mixture take the middle value of 1.51g / cm 3 , take the middle value particle size of 40 mesh (i.e. 0.45mm), the proportion of 40 mesh proppant content is: 0.33‰ / 2+1.6‰ / 3=0.725‰;

[0041] Example 1: Proppant identification results by theoretical calculation

[0042] If 0.1g of sample is all 40 mesh quartz sand, the weight of proppant particles contained is: 0.1×0.725‰=7.25×10 -5 g;

[0043] Volume: 7.25×10 -5 / 1.51=4.80×10 -5 cm 3 ;

[0044] The number of 40 mesh quartz sand particles contained is: 4.80×10 -5 ×10 4 / (4 / 3×3.14×(0.45 / 2)^3)=10.07 pieces.

[0045] Example 2: Combination Figure 1 The proppant identification method flow chart shown in the figure identifies the proppant particles by the method of the present invention:

[0046] (1) Preliminary screening: The sphericity of the sample particle image is analyzed by Hough circle transform (CHT) to exclude solid particles that do not meet the sphericity standards;

[0047] The accumulation surface of the standard Hough circle transform is a three-dimensional space, which requires more calculations than the Hough line transform. The Opencv Hough circle transform has made an operational optimization of the standard Hough circle transform, namely the "Hough gradient method". Its detection idea is to traverse and accumulate the centers of circles corresponding to all non-zero points and consider the centers of circles. The center of the circle must be on the modulus vector of each point on the circle, that is, on the vertical line perpendicular to the point and passing through the tangent of the point, the intersection of the modulus vectors on these circles is the center of the circle. The Hough gradient method is to find these centers of circles, and make a final judgment based on the number of intersections of the modulus vectors on the "center of the circle" and the threshold.

[0048] The steps of Hough gradient method for image processing are as follows:

[0049] ① Perform a Canny edge detection on the image to obtain the edge detection map of the target image;

[0050] ② Calculate the local gradient value of each non-zero point in the Canny edge image. The Sobel function can be used to calculate the Sobel first-order derivative in the x and y directions to obtain the local gradient value of the point;

[0051] ③ Initialize the circle center space N(a,b), and set all N(a,b)=0;

[0052] ④ Traverse all non-zero pixels in the Canny edge map, draw a line along the local gradient direction of the point (that is, the perpendicular direction of the tangent of the point), and then let N(a,b)=N(a,b)+1 for all points (a,b) through which the line segment passes;

[0053] ⑤ Sort N(a,b) in order of size, where the larger the N(a,b), the more likely it is to be the center of the circle, thus obtaining the most likely center point;

[0054] ⑥ Calculate the distance from the center of the circle to all non-zero pixels in the Canny edge map;

[0055] ⑦ Sort the distances from small to large, and select the appropriate possible radius based on the threshold;

[0056] ⑧ Initialize the radius space N(r), and set all N(r)=0;

[0057] ⑨ Traverse all non-zero pixel points in the Canny edge map, and draw circles with different possible radii r with this point as the center. If the circle passes through the center of the circle in step ⑤, then set N(r) = N(r) + 1;

[0058] ⑩ Count the possible radius values ​​and sort the size of N(r). The larger N(r), the more times this distance value appears, and the more likely the corresponding r is the radius value.

[0059] In this embodiment, a Hough circle transform program suitable for proppant particle identification is written using Python software and OpenCV database; OpenCV is used to perform Hough circle transform on the sample image by calling the HoughCircles function, thereby identifying samples with repeated points in the abr coordinate system ≤ 50% of the total number of non-zero pixels taken in the Cartesian coordinate system and excluding them. The results are as follows: Figure 2 As shown, the particles pointed by the arrows have better sphericity;

[0060] (2) eliminating solid particles whose transparency does not meet the standard in step (1) through the image transparency TR;

[0061] (2-1) Determination of transparency TR threshold:

[0062] 400 proppant samples were selected and divided into 20 groups, each with 20 samples. Taking one group of 20 quartz proppant particle sample images as an example, the number of pixels of each particle was counted, and the color sample was taken as the background illumination color. The image obtained after the tolerance was set to 130 was calculated, and the pixels with tolerance <130 were taken as transparent pixels. The statistical results are shown in Table 1. According to the results in Table 1, the larger the particle size of the quartz proppant, the lower its transparency is generally, but it still maintains a certain degree of transparency. The judgment threshold of transparency TR is set to 20%, that is, when the transparency TR of a particle is greater than 20%, it can be considered as a proppant.

[0063] Table 1 Transparency and related data of proppant

[0064]

[0065] (2-2) Perform transparency analysis on the proppant particles identified in step (1):

[0066] exist Figure 2 Three particles are randomly selected from the above, taking particles A, B, and C as an example. Figure 3 As shown in the figure, it can be seen that the TR of the proppant particles in the figure is higher, so the particles are brighter; at the same time, there are some proppant particles with lower TR, which appear darker; and the TR of other mineral particles is very low, so they appear very dark; the relevant data of particles A, B, and C obtained are shown in Table 2; the transparency visualization picture obtained after Photoshop processing is shown in Figure 4 As shown, the white part is the transparent part of the particle, and the TR value can be obtained by calculating the percentage of the number of pixels of the white part to the number of pixels of the entire particle;

[0067] Table 2 Transparency and related data of particles A, B, and C After the initial identification in step (1), the sample Figure 2After continuing the transparency analysis, the results are as follows Figure 5 As shown, the arrows point to proppant particles identified by transparency TR;

[0068] (3) Use Photoshop to process the target particles and obtain the red and blue difference YC of the image f ; Through the red and blue difference YC f Removes mineral particles:

[0069] (3-1) Red and blue difference YC f Threshold determination:

[0070] The iron content in the stratum of this example is relatively high, which makes many natural quartz particles in the stratum have a high transparency, but contain a lot of iron elements, and appear obvious red as a whole. When switching to the upper light source for observation, the difference between its color and that of the standard proppant is particularly obvious; the red-blue color difference is calculated for the particles identified in step (2);

[0071] 200 proppant particles and 200 natural quartz sand particles were selected and divided into 20 groups, each group containing 10 proppant particles and 10 natural quartz sand particles. f Value comparison determines the red and blue difference YC f Threshold, red and blue difference YC f The value calculation method is as follows: Taking proppant particle 1 and natural quartz sand particle 2 as examples for testing, the obtained color and color level histogram is as follows Figure 6 , Figure 7 As shown, according to the data in the color level histogram, we can get:

[0072] YC f1 =R f1 -B f1 =137-129=8

[0073] YC f2 =R f2 -B f2 =173-78=95

[0074] From the calculation results, it can be seen that the YC of particle 1 f The value is less than 8, which is a proppant particle; while the YC f The value is greater than 95, which is other mineral particles; by taking 10 quartz sand proppants and 10 local natural quartz sand in Table 3 as an example, the final YC f The threshold of YC is set to 30, that is, f When <30, it is judged to be a proppant particle.

[0075] Table 3 Statistics of red and blue median values ​​and red and blue difference of proppant

[0076]

[0077] (3-2) The result of step (2) is further processed by red and blue difference YC f analyze:

[0078] The results are as follows Figure 8 As shown, the arrow points to the red and blue color difference YC f The obtained particles are further identified as proppant particles.

[0079] In summary, when identifying proppant particles, the sample particle image is first processed using the Hough circle transform program to find particles with better roundness, and then the particle transparency TR and red-blue color difference YC are observed. f , comprehensively considering these three factors to judge whether the particle is a proppant particle. As shown by the arrow, the final determination is that the number of proppant particles in the 0.1g observation sample is 9, which is equivalent to the theoretical calculation value.

[0080] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. A method for identifying proppants in rock cuttings and mud, characterized in that: The identification method uses sphericity, transparency TR and red-blue color difference YC in sequence. f Three parameters are used to analyze the sample image particles to identify the proppant.

2. The identification method according to claim 1, characterized in that: The specific steps include: (1) Preliminary screening: The sphericity of the sample particle image is analyzed by Hough circle transform (CHT) to exclude solid particles that do not meet the sphericity standards; (2) Eliminate solid particles that do not meet the transparency standards through image transparency TR; (3) Through the red and blue difference YC f Mineral particles are removed to obtain proppant particles.

3. The method according to claim 2, characterized in that The Hough circle transform CHT is written using Python software and OpenCV database.

4. The method according to claim 3, characterized in that: The Hough circle transform program CHT maps non-zero pixel points in the Cartesian coordinate system to the abr coordinate system. When the number of repeated points in the abr coordinate system is ≤ 50% of the total number of non-zero pixel points in the Cartesian coordinate system, the particles are solid particles whose sphericity does not meet the standard.

5. The method according to claim 2, characterized in that: The transparency TR is calculated by using a color range selector in machine vision processing software, taking a color sample as the background light color, setting the tolerance to 130, and the pixels with a tolerance of less than 130 are regarded as transparent pixels.

6. The method according to claim 2, characterized in that The transparency TR is the percentage of the number of pixels of the transparent part of the particle to the number of pixels of the entire particle.

7. The method according to claim 2, characterized in that In step (2), particles with a transparency TR ≤ 20% are particles that do not meet the transparency standard.

8. The method according to claim 2, characterized in that: The red and blue color difference YC f The determination process is as follows: first remove the background interference part other than the proppant particles determined in step (1) and step (2), and use the color level histogram to obtain the red color level median value R of the target particle. f and the blue middle value B f The red and blue difference YC f The red color level peak R of the target particle f With blue color level peak B f The difference is calculated as follows: YC f =R f -B f 。 9. The method according to claim 2, characterized in that: In step (3), the red and blue difference YC f Particles with a particle size of ≥30 are mineral particles.

10. Use of the method according to any one of claims 1 to 9 in identifying proppants in a mixture of rock cuttings and mud.

Images