Rock core pore and crack identification and network model construction method and device

Abstract

The invention relates to the technical field of oil and gas field development, in particular to a core pore and crack recognition and network model construction method and device, and the method comprises the steps: obtaining a target core three-dimensional structure image; determining a target rock core three-dimensional structure image training sample, and training the rock core three-dimensional structure image by using the training sample to obtain a rock core pore and crack identification model; segmenting the rock core three-dimensional structure image based on a core pore and crack identification model, and extracting segmented cracks and pores; dividing pores and throats in the rock core three-dimensional structure image according to the segmented cracks and pores; and establishing a core pore and crack network model according to the pores and the throats. The problems that an existing traditional processing method for identifying the cracks and the pores of the digital rock core is simple in algorithm, the identification and segmentation effect on the cracks and the pores of the complex rock core is extremely poor, and consequently the experiment result is inaccurate in the subsequent digital rock core experiment are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas field development, and in particular to a method and a device for identifying core pores and fractures and building a network model. Background Art

[0002] Comprehensive and accurate experimental analysis of reservoir cores is the key to increasing the production of oil and gas wells. The experimental analysis includes core sensitivity experiments, conductivity analysis, and acid and fracturing fluid damage experiments. These experiments require a large number of cores, but due to the limited number of sampled cores, it is difficult to complete too many necessary experiments, which brings great difficulties to the increase of oil and gas well production. It is urgent to solve the problem of insufficient core quantity by establishing digital cores.

[0003] Digital core technology is a technology that reconstructs cores based on two-dimensional scanning electron microscope images or three-dimensional CT scanning images, using computer image processing technology and certain algorithms. It can achieve three-dimensional high-precision visual representation and visual evaluation of reservoir properties. To a certain extent, it can replace traditional cores and alleviate the problem of difficulty in conducting core experiments due to the small number of cores.

[0004] However, the establishment of digital cores is still in the core modeling stage. One of the main problems is that the traditional processing methods such as threshold segmentation methods, cluster analysis methods and boundary detection methods are simple, and the identification and segmentation technology of core fractures and pores is extremely ineffective, especially for cores with strong heterogeneity. Summary of the invention

[0005] The present invention proposes a method and device for identifying core pores and fractures and building a network model, so as to solve the problem that the algorithm of the existing traditional processing method for identifying fractures and pores in digital cores is simple, and the identification and segmentation effect of fractures and pores in complex cores is extremely poor, resulting in inaccurate experimental results in subsequent digital core experiments.

[0006] According to one aspect of the present invention, a method for identifying core pores and fractures and constructing a network model is provided, comprising:

[0007] Obtain the three-dimensional structural image of the target core;

[0008] Determine a target core three-dimensional structure image training sample, and use the training sample to train the core three-dimensional structure image to obtain a core pore and fracture recognition model;

[0009] Based on the core pore and fracture identification model, the core three-dimensional structure image is segmented to extract the segmented fractures and pores;

[0010] According to the segmented fractures and pores, pores and throats in the three-dimensional structure image of the core are divided;

[0011] Based on the pores and throats, a core pore and fracture network model is established.

[0012] Preferably, before acquiring the target core three-dimensional structure image, the target core three-dimensional structure image is determined, and the method includes:

[0013] Acquire target core;

[0014] Performing tomography scanning on the target core to obtain spatial voxels of the target core;

[0015] Determine a two-dimensional distribution matrix of CT values ​​of the target core according to the target core space voxels;

[0016] A three-dimensional structural image of the target core is constructed according to the two-dimensional distribution matrix of CT values.

[0017] Preferably, the method for determining a target core three-dimensional structure image training sample comprises:

[0018] Acquire existing core pore and fracture data of the same rock type as the target core, or identify pore and fracture samples in the three-dimensional structure image of the core;

[0019] The identified pore and fracture samples or the existing core pore and fracture data are the training samples.

[0020] Preferably, the method for dividing the pores and throats in the three-dimensional structure image of the core according to the segmented fractures and pores comprises:

[0021] Determine the number N of pores in the 26 neighborhoods of any point in the three-dimensional structure image of the core;

[0022] If the number N of the pores is less than the predetermined number Ns, the corresponding fractures and / or pores are throats, and the others are pores.

[0023] Preferably, the method for establishing a core pore and fracture network model based on pores and throats comprises:

[0024] Considering the pores as spheres of equal volume, an equivalent pore radius is determined;

[0025] Considering the throat as a cylindrical pipe of the same volume, determining the equivalent throat radius;

[0026] According to the equivalent pore radius and the equivalent throat radius, the core pore and fracture network structure is established, thereby completing the construction of the core pore and fracture network model.

[0027] Preferably, the method for determining the equivalent pore radius comprises:

[0028] Determine the equivalent pore radius using a pore equivalent radius calculation formula;

[0029] The calculation formula of the pore equivalent radius is:

[0030]

[0031] Where n is the number of pores that are continuously determined from any direction in the 26 neighborhoods of a point, V voxel_k is the volume of a voxel of the digital core, and r is the equivalent radius of the pore.

[0032] Preferably, the method for determining the equivalent throat radius comprises:

[0033] Determine the equivalent throat radius using a throat equivalent radius calculation formula;

[0034] The calculation formula of the pore equivalent radius is:

[0035]

[0036] Where, L = m × L voxel

[0037] Where: m is the number of throats that are continuously determined to be from any direction in the 26-neighborhood of a point, V voxel_k is the volume of a voxel of the digital core, R is the equivalent radius of the throat, L voxel is the length of one voxel.

[0038] Preferably, it also includes:

[0039] Determining the porosity of the core pore and fracture network model, comparing it with the original target core porosity, and determining the relative error between the two;

[0040] It is determined whether the relative error is within a predetermined error range. If not, the core pore and fracture network model is retrained until the relative error is within the predetermined error range.

[0041] Preferably, the method for determining the porosity of the core pores and fracture network model comprises:

[0042] The porosity of the core pore and fracture network model is determined using formula (3);

[0043]

[0044] In the formula, is the porosity, V voxelf is the volume of the voxels that constitute the crack (corresponding to the crack part in the network model), V voxelpis the volume of the voxels constituting the pores, n1 is the number of voxels constituting the cracks, and n2 is the number of voxels constituting the pores.

[0045] According to one aspect of the present invention, a device for identifying core pores and fractures and building a network model is provided, comprising:

[0046] An acquisition unit, used for acquiring a three-dimensional structural image of a target core;

[0047] The recognition model determination unit is used to determine a target core three-dimensional structure image training sample, and use the training sample to train the core three-dimensional structure image to obtain a core pore and fracture recognition model;

[0048] A pore and fracture extraction unit, used to segment the core three-dimensional structure image based on the core pore and fracture identification model, and extract the segmented fractures and pores;

[0049] A pore and throat determination unit, configured to divide the pores and throats in the core three-dimensional structure image according to the segmented fractures and pores;

[0050] The network model determination unit is used to establish a core pore and fracture network model according to the pores and throats.

[0051] The present invention has at least the following beneficial effects:

[0052] The present invention proposes a method and device for identifying and building a network model of core pores and fractures. A pore and fracture identification model is established through a three-dimensional structural image of the core. The model is used to accurately divide the pores and fractures in the three-dimensional image of the core, thereby realizing the identification and segmentation of pores and fractures of different types of cores, as well as the construction of a fracture and pore network model, solving the problem of segmentation of core fractures, pores, and matrix, and laying a solid foundation for the construction and development of high-quality digital cores. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present invention and, together with the specification, are used to explain the technical solutions of the present invention.

[0054] Figure 1 A flow chart of a method for identifying core pores and fractures and building a network model according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0055] Various exemplary embodiments, features and aspects of the present invention will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0056] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0057] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0058] In addition, in order to better illustrate the present invention, numerous specific details are provided in the following specific embodiments. It should be understood by those skilled in the art that the present invention can be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present invention.

[0059] Figure 1 A flow chart of a method for identifying core pores and fractures and building a network model according to an embodiment of the present invention is shown. Figure 1 As shown, a method for identifying and building a core pore and fracture network model is characterized in that it includes: step S01: acquiring a target core three-dimensional structure image; step S02: determining a target core three-dimensional structure image training sample, and using the training sample to train the core three-dimensional structure image to obtain a core pore and fracture identification model; step S03: based on the core pore and fracture identification model, segmenting the core three-dimensional structure image, and extracting the segmented fractures and pores; step S04: dividing the pores and throats in the core three-dimensional structure image according to the segmented fractures and pores; step S05: establishing a core pore and fracture network model according to the pores and throats.

[0060] The core pore and fracture identification and network model construction method provided by the embodiment of the present invention specifically includes the following steps:

[0061] Step S01: Acquire a three-dimensional structural image of a target core.

[0062] In the present invention, before obtaining the three-dimensional structural image of the target core, the three-dimensional structural image of the target core is determined, and the method includes: obtaining the target core; performing a tomographic scan on the target core to obtain the target core spatial voxels; determining the two-dimensional distribution matrix of the CT value of the target core according to the target core spatial voxels; and constructing the three-dimensional structural image of the target core according to the two-dimensional distribution matrix of the CT value.

[0063] In an embodiment of the present invention, the method for obtaining the target core is: a target core with strong heterogeneity and natural fractures is drilled from the core column obtained by drilling along the bedding direction at a certain distance, and the drilled core is prepared according to the national standard "GB / T29172-2012".

[0064] The prepared target core is placed on a rotating stage, and the X-ray source emits rays to penetrate the target core on the rotating stage for tomography. The spatial voxels of the scanned core are obtained by filtered back projection to obtain a two-dimensional distribution matrix of μ values ​​(attenuation coefficients), and the two-dimensional distribution matrix of CT values ​​is obtained by converting the two-dimensional distribution matrix of μ values.

[0065] The CT value of each core voxel is converted into the grayscale corresponding to each pixel to obtain the core grayscale distribution matrix. The grayscale values ​​in the core grayscale distribution matrix are used as an image to represent and construct a virtual target core three-dimensional structure image.

[0066] The noise of the constructed three-dimensional structure image of the core is eliminated by the non-local mean filtering method.

[0067] Step S02: determining a target core three-dimensional structure image training sample, and using the training sample to train the core three-dimensional structure image to obtain a core pore and fracture recognition model.

[0068] In the present invention, the method for determining the training samples of the target core three-dimensional structural image includes: obtaining existing core pore and fracture data belonging to the same rock and mineral type as the target core, or identifying pore and fracture samples in the core three-dimensional structural image; the identified pore and fracture samples or the existing core pore and fracture data are the training samples.

[0069] In an embodiment of the present invention, training samples can be obtained in two ways. One is to manually identify and mark some pores and cracks in the three-dimensional structure image of the core as training samples; the other is to refer to the pore and crack data of the core belonging to the same rock and mineral type as the target core in an existing established database, and the required data can be quickly extracted as training samples by automatically matching the database data with the three-dimensional structure image of the target core.

[0070] The training samples and the core three-dimensional structure images are input into the deep-learning software for training to obtain the core pore and fracture recognition model after the first training.

[0071] Step S03: Based on the core pore and fracture identification model, the core three-dimensional structure image is segmented to extract the segmented fractures and pores.

[0072] In the embodiment of the present invention, the core three-dimensional structure image is trained by using partial fracture and pore data existing in the training sample, so that all fractures and pores existing in the target core can be simulated to generate a core pore and fracture identification model.

[0073] Among them, the core pore and fracture recognition model segments the core three-dimensional image by using the U-Net architecture to perform valid convolution, structural feature and detail feature fusion, maximum pooling, deconvolution, and loss function calculation.

[0074] The corresponding pores and fractures in the core 3D structure image are segmented and extracted for subsequent ball-and-stick model establishment.

[0075] Step S04: dividing the pores and throats in the core three-dimensional structure image according to the segmented fractures and pores.

[0076] In the present invention, the method for dividing the pores and throats in the three-dimensional structure image of the core according to the segmented fractures and pores includes: determining the number N of pores judged to be in the 26 neighborhood of any point in the three-dimensional structure image of the core; if the number N of pores is less than the predetermined number Ns, the corresponding fractures and / or pores are throats, and the others are pores.

[0077] In the embodiment of the present invention, the fracture is composed of pores and throats. The division of pores and throats is based on the number N of pores in the 26 neighborhoods of any point. When the number of pores is lower than the predetermined number Ns, it will be judged as a throat, and the rest will be judged as pores. The value of Ns is also different according to the different physical properties of the core. The better the connectivity of the core, the smaller the value of Ns. The value of Ns can be: 10.

[0078] That is, it is determined whether the number of pores N in the 26 neighborhoods of each point in the three-dimensional structure image of the core is less than 10. If so, the pores and / or cracks in the 26 neighborhoods of the point are determined to be throats. Otherwise, the pores and / or cracks in the 26 neighborhoods are determined to be pores.

[0079] Step S05: establishing a core pore and fracture network model based on the pores and throats.

[0080] In the present invention, the method for establishing a core pore and fracture network model based on pores and throats includes: considering the pores as spheres of the same volume to determine the equivalent pore radius; considering the throats as cylindrical pipes of the same volume to determine the equivalent throat radius; establishing a core pore and fracture network structure based on the equivalent pore radius and the equivalent throat radius, thereby completing the construction of the core pore and fracture network model.

[0081] In the present invention, the method for determining the equivalent pore radius comprises: using a pore equivalent radius calculation formula to determine the equivalent pore radius;

[0082] The calculation formula of the pore equivalent radius is:

[0083]

[0084] Where n is the number of pores that are continuously determined to be pores in any direction of the 26 neighborhood of a point in the core 3D structure image, V voxel_k is the volume of a voxel of the digital core (three-dimensional structure image of the core), and r is the equivalent radius of the pore.

[0085] In the present invention, the method for determining the equivalent throat radius includes: using a throat equivalent radius calculation formula to determine the equivalent throat radius;

[0086] The calculation formula of the pore equivalent radius is:

[0087]

[0088] Where, L = m × L voxel ;

[0089] Where: m is the number of throats that are continuously identified from any direction in the 26-neighborhood of a point in the core 3D structure image, V voxel_k is the volume of a voxel of the digital core (core 3D structure image), R is the equivalent radius of the throat, L is voxel is the length of one voxel.

[0090] In this embodiment of the present invention, the pores are regarded as spheres of the same volume, and the throats are regarded as cylindrical pipes of the same volume. The equivalent pore radius and the equivalent throat radius of the fractures and pores segmented by step S03 are calculated respectively, and then the core pore and fracture network structure is established according to the calculation results to complete the construction of the core pore and fracture network model (ball-and-stick model).

[0091] In the present invention, it also includes: step S06: determining the porosity of the core pores and fracture network model, comparing it with the original target core porosity, and determining the relative error between the two;

[0092] It is determined whether the relative error is within a predetermined error range. If not, the core pore and fracture network model is retrained until the relative error is within the predetermined error range.

[0093] In the present invention, the method for determining the porosity of the core pore and fracture network model comprises:

[0094] The porosity of the core pore and fracture network model is determined using formula (3);

[0095]

[0096] In the formula, is the porosity, V voxelf is the volume of the voxels that constitute the crack (corresponding to the crack part in the network model), V voxelp is the volume of the voxels constituting the pores, n1 is the number of voxels constituting the cracks, and n2 is the number of voxels constituting the pores.

[0097] In an embodiment of the present invention, the porosity of the original target core can be calculated according to different core physical properties, including nuclear magnetic resonance method, nitrogen adsorption method, alcohol method, gas drive method and overburden method; the porosity of the core pore and fracture network model is compared with the porosity of the original target core, and the relative error between the two is calculated. If the relative error is within the allowable range (predetermined error range), the core pore and fracture identification model can be used for subsequent core pore and fracture identification or other simulation experiments such as core sensitivity experiments, conductivity analysis, and acid and fracturing fluid damage experiments; if the relative error is large, the cause of the error is analyzed to determine whether the core pore and fracture network model needs to be re-learned and trained. If re-training is required, the above steps S02-S06 are repeated until the relative error between the porosity of the core pore and fracture network model and the porosity of the original target core is within the allowable range.

[0098] It can be understood that the above-mentioned various method embodiments mentioned in the present invention can be combined with each other to form a combined embodiment without violating the principle logic. Due to space limitations, the present invention will not go into details.

[0099] The execution subject of the core pore and fracture identification and network model construction method may be a core pore and fracture identification and network model construction device, for example, the core pore and fracture identification and network model construction method may be executed by a terminal device or a server or other processing device, wherein the terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the core pore and fracture identification and network model construction method may be implemented by a processor calling a computer-readable instruction stored in a memory.

[0100] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0101] In an embodiment of the present invention, a core pore and fracture identification and network model construction device is also provided, characterized in that it includes: an acquisition unit, used to acquire a target core three-dimensional structure image; an identification model determination unit, used to determine a target core three-dimensional structure image training sample, and use the training sample to train the core three-dimensional structure image to obtain a core pore and fracture identification model; a pore and fracture extraction unit, used to segment the core three-dimensional structure image based on the core pore and fracture identification model, and extract the segmented fractures and pores; a pore and throat determination unit, used to divide the pores and throats in the core three-dimensional structure image according to the segmented fractures and pores; a network model determination unit, used to establish a core pore and fracture network model according to the pores and throats.

[0102] In some embodiments, the functions or modules and units included in the device provided by the embodiment of the present invention can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

[0103] At present, the identification of digital core fractures and pores is mainly based on traditional processing methods such as threshold segmentation method, cluster analysis method and boundary detection method, which generally separates fractures and pores through threshold segmentation. Its algorithm is simple, and the identification and segmentation effect of fractures and pores in complex cores is extremely poor. In order to solve the problem of poor identification and segmentation effect of fractures and pores in digital cores, the present invention proposes a method for realizing pore and fracture identification and segmentation of different types of cores, as well as the construction of fracture-pore network model. This method is suitable for the construction and evaluation of digital cores of various reservoirs, and is the basis for three-dimensional visualization research on reservoir properties, fluid seepage, transformation evaluation, etc. It solves the segmentation problem of core fractures, pores and matrix, and lays a solid foundation for the construction and development of high-quality digital cores. It is of great significance for the promotion and application of digital cores in oil fields.

[0104] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for identifying core pores and fractures and building a network model, characterized in that: include: Obtain the three-dimensional structural image of the target core; Determine a target core three-dimensional structure image training sample, and use the training sample to train the core three-dimensional structure image to obtain a core pore and fracture recognition model; Based on the core pore and fracture identification model, the core three-dimensional structure image is segmented to extract the segmented fractures and pores; According to the segmented fractures and pores, pores and throats in the three-dimensional structure image of the core are divided; Based on the pores and throats, a core pore and fracture network model is established.

2. The method for core pore and fracture identification and network model construction according to claim 1, characterized in that: Before acquiring the target core three-dimensional structure image, determining the target core three-dimensional structure image, the method includes: Acquire target core; Performing tomography scanning on the target core to obtain spatial voxels of the target core; Determine a two-dimensional distribution matrix of CT values ​​of the target core according to the target core space voxels; A three-dimensional structural image of the target core is constructed according to the two-dimensional distribution matrix of CT values.

3. The method for core pore and fracture identification and network model construction according to claim 1, characterized in that: The method for determining a target core three-dimensional structure image training sample comprises: Acquire existing core pore and fracture data of the same rock type as the target core, or identify pore and fracture samples in the three-dimensional structure image of the core; The identified pore and fracture samples or the existing core pore and fracture data are the training samples.

4. The method for core pore and fracture identification and network model construction according to claim 1, characterized in that: The method for dividing the pores and throats in the core three-dimensional structure image according to the segmented fractures and pores includes: Determine the number N of pores in the 26 neighborhoods of any point in the three-dimensional structure image of the core; If the number N of the pores is less than the predetermined number Ns, the corresponding fractures and / or pores are throats, and the others are pores.

5. The method for core pore and fracture identification and network model construction according to claim 1, characterized in that: The method for establishing a core pore and fracture network model based on pores and throats includes: Considering the pores as spheres of equal volume, an equivalent pore radius is determined; Considering the throat as a cylindrical pipe of the same volume, determining the equivalent throat radius; According to the equivalent pore radius and the equivalent throat radius, the core pore and fracture network structure is established, thereby completing the construction of the core pore and fracture network model.

6. The method for core pore and fracture identification and network model construction according to claim 5, characterized in that: The method for determining the equivalent pore radius comprises: Determine the equivalent pore radius using a pore equivalent radius calculation formula; The calculation formula of the pore equivalent radius is: Where n is the number of pores that are continuously determined to be pores in any direction from the 26 neighborhoods of a point, V voxel_k is the volume of a voxel of the digital core, and r is the equivalent radius of the pore.

7. The method for core pore and fracture identification and network model construction according to claim 5, characterized in that: The method for determining the equivalent throat radius comprises: Determine the equivalent throat radius using a throat equivalent radius calculation formula; The calculation formula of the pore equivalent radius is: Where, L = m × L voxel Where: m is the number of throats that are continuously determined from any direction in the 26 neighborhoods of a point, V voxel_k is the volume of a voxel of the digital core, R is the equivalent radius of the throat, L voxel is the length of one voxel.

8. The method for core pore and fracture identification and network model construction according to claim 1, characterized in that: Also includes: Determining the porosity of the core pore and fracture network model, comparing it with the original target core porosity, and determining the relative error between the two; It is determined whether the relative error is within a predetermined error range. If not, the core pore and fracture network model is retrained until the relative error is within the predetermined error range.

9. The method for core pore and fracture identification and network model construction according to claim 8, characterized in that: The method for determining the porosity of the core pore and fracture network model comprises: The porosity of the core pore and fracture network model is determined using formula (3); In the formula, is the porosity, V voxelf is the volume of the voxels that constitute the crack (corresponding to the crack part in the network model), V voxelp is the volume of the voxels constituting the pores, n1 is the number of voxels constituting the cracks, and n2 is the number of voxels constituting the pores.

10. A core pore and fracture identification and network model building device, characterized in that: include: An acquisition unit, used for acquiring a three-dimensional structural image of a target core; The recognition model determination unit is used to determine a target core three-dimensional structure image training sample, and use the training sample to train the core three-dimensional structure image to obtain a core pore and fracture recognition model; A pore and fracture extraction unit, used to segment the core three-dimensional structure image based on the core pore and fracture identification model, and extract the segmented fractures and pores; A pore and throat determination unit, configured to divide the pores and throats in the core three-dimensional structure image according to the segmented fractures and pores; The network model determination unit is used to establish a core pore and fracture network model according to the pores and throats.

Images