A method for identifying and analyzing the genetic mechanism of weak structures in deep thermal reservoirs

Abstract

The present invention provides a method for identifying and analyzing the causal mechanism of weak structures in deep heat reservoirs, which relates to the technical field of deep high-temperature geothermal reservoir exploration and development. The present invention selects multiple exploration wells in the deep heat reservoir exploration and development area, uses drilling steel pipes to collect samples, and during the drilling process, rock samples are transported to the surface sample collection point through a spiral conveying device in the steel pipe. In addition, the sample length is measured in real time, and the sample depth is recorded in combination with the order in which the steel pipe is driven, so that the sample depth interval can be accurately divided, the sample surface characteristics and composition information can be analyzed in detail, and the location, scale and nature of the weak structure in the deep heat reservoir can be accurately identified. In addition, the database generated by combining these data can accurately grasp the causal mechanism of the weak structure in the deep heat reservoir, which helps to more efficiently develop deep heat reservoir resources and increase the supply of clean energy.

Description

Technical Field

[0001] The present invention relates to the technical field of deep high-temperature geothermal reservoir exploration and development, and in particular to a method for identifying and analyzing the genetic mechanism of weak structures in deep geothermal reservoirs. Background Art

[0002] As the global energy mix accelerates toward clean energy, deep thermal reservoirs, a highly promising clean energy source, have become a key focus of energy research and development due to their abundant reserves, high stability, and environmental friendliness. The key to the efficient development and utilization of deep thermal reservoir resources lies in the accurate identification and in-depth study of their weak structures, which is directly related to drilling safety, thermal reservoir extraction efficiency, and energy output quality. During actual exploration and development, it is necessary to collect and analyze samples from deep thermal reservoir areas to identify the location, scale, and nature of these weak structures, thereby providing a scientific basis for the development of extraction plans.

[0003] Currently, traditional sample collection and analysis methods for studying weak structures in deep thermal reservoirs have numerous limitations. For one thing, when determining sample depth, most methods rely solely on the approximate depth of the drill pipe, failing to fully consider the variations in the actual source depths of rock samples within the pipe. This results in inaccurate sample depth positioning and an inability to accurately reflect deep geological structural characteristics. Furthermore, when analyzing sample characteristics, the methods used are limited in their analysis of surface features and composition, making it difficult to fully capture the characteristic information associated with weak structures and accurately identify their location and properties.

[0004] Therefore, how to accurately determine the spatial distribution and formation mechanism of weak structures has become an urgent problem that needs to be solved. Summary of the Invention

[0005] The present invention provides a method for identifying and analyzing the formation mechanism of weak structures in deep thermal reservoirs, which can accurately determine the spatial distribution and formation mechanism of weak structures.

[0006] A first aspect of the present invention provides a method for identifying and analyzing the formation mechanism of weak structures in deep thermal reservoirs, comprising:

[0007] Acquire a sample set collected from a weak structural area of a deep heat reservoir, and divide the depth interval of the sample set into subintervals of each subsample according to the collection order and length measurement results of the sample set;

[0008] Recognizing the image data to obtain surface features of each sub-sample, and marking the sub-sample corresponding to the surface features that meet the sampling conditions;

[0009] The deep geological parameters are generated by combining the composition information and subintervals corresponding to the subsamples of fixed nodes and marked nodes;

[0010] The deep geological parameters of the subsamples are correlated and integrated to construct a comprehensive database.

[0011] Optionally, in a possible implementation of the first aspect, dividing the depth interval of the sample set to obtain subintervals of each subsample according to the acquisition order and length measurement result of the sample set includes:

[0012] Based on the acquisition sequence and the length measurement result, cumulatively recording the depth of each of the subsamples based on the depth interval, and determining an initial depth interval when there is no gap between each of the subsamples;

[0013] When the total measured length of each subsample is less than the total length of the depth interval, comparing surface features corresponding to the image data of adjacent subsamples and marking the subsample groups with gaps, the surface features including edge features and color features;

[0014] Determining the predicted spacing of the subsample groups according to the feature difference in the image data;

[0015] The predicted interval is inserted into the initial depth interval corresponding to the subsample group, and the cumulative depth of each subsample is recalculated to obtain a subinterval of each subsample.

[0016] Optionally, in a possible implementation manner of the first aspect, comparing surface features corresponding to image data of adjacent subsamples and marking subsample groups with gaps therebetween includes:

[0017] When comparing edge features, edge contours are extracted from image data of adjacent subsamples;

[0018] Obtaining concave points and convex points in each of the edge contours, calculating a concave point difference value and a convex point difference value in each of the edge contours, and when the concave point difference value and / or the convex point difference value is greater than a quantity threshold, determining that there is a gap in the subsample group and marking the gap;

[0019] When the concave point difference and the convex point difference are both less than the quantity threshold, obtaining a matching set of concave points and convex points with the smallest distance in each of the edge contours;

[0020] The total distance of each matching set is counted, and when the total distance is greater than a fit threshold, it is determined that there is a gap between the subsample groups and the gap is marked.

[0021] Optionally, in a possible implementation of the first aspect, obtaining a matching set of concave points and convex points with the smallest distance in each edge contour includes:

[0022] For each concave point of each edge contour, calculate the distance between it and each convex point in another edge contour, and select the convex point with the smallest distance to generate a matching set;

[0023] For each convex point of each edge contour, the distance between the convex point and each concave point in another edge contour is calculated, and the concave point with the smallest distance is selected to generate a matching set.

[0024] Optionally, in a possible implementation manner of the first aspect, comparing surface features corresponding to image data of adjacent subsamples and marking subsample groups with gaps therebetween includes:

[0025] When comparing color features, extracting sample contours from image data of adjacent subsamples;

[0026] Acquire a target area in which pixel values of each sample contour are located in the same pixel interval and at the edge;

[0027] The target area of each sample outline is compared with the target area of another sample outline. If there is no target area with the same pixel interval, it is determined that there is a gap between the subsample groups and the gap is marked.

[0028] Optionally, in a possible implementation manner of the first aspect, determining the predicted spacing of the subsample group according to the feature difference in the image data includes:

[0029] For edge features, the preset difference corresponding to the largest concave point difference or convex point difference is determined as the feature difference, or the preset difference corresponding to the total distance of the matching set is determined as the feature difference;

[0030] For the color feature, obtaining the minimum color difference corresponding to a plurality of target areas adjacent to the subsamples, and determining a preset difference corresponding to the minimum color difference as the feature difference;

[0031] The length difference between the total length of the depth interval and the total measured length of each subsample is obtained, and the length difference is proportionally divided according to the ratio between the characteristic differences of the subsample groups to obtain the predicted spacing of each subsample group.

[0032] Optionally, in a possible implementation of the first aspect, identifying the image data to obtain surface features of each subsample, and marking the subsamples corresponding to the surface features that meet the sampling conditions includes:

[0033] Selecting each of the subsamples in sequence according to the acquisition order, and identifying color features included in surface features in the image data of the subsamples;

[0034] determining a pixel mean of the subsample based on the color feature, and calculating a pixel difference between the pixel mean of a current subsample and the pixel mean of a previous subsample;

[0035] When the pixel difference is within the abnormal difference interval, it is determined that the current subsample meets the sampling condition and is marked.

[0036] Optionally, in a possible implementation of the first aspect, combining the composition information and subintervals corresponding to the subsamples of the fixed nodes and the marked nodes to generate deep geological parameters includes:

[0037] According to the corresponding relationship between the composition information and the temperature, the predicted temperature corresponding to the corresponding sub-interval is determined, and according to the mapping relationship between the sub-interval and the predicted temperature, the deep geological parameters are generated.

[0038] Optionally, in a possible implementation of the first aspect, the method further includes:

[0039] Obtaining scanning information of sample data of each well location in the deep heat storage weak structure area by inspection equipment;

[0040] A blank depth interval where sample data is missing is determined based on the scanning information, and sub-regions adjacent to the blank depth interval are selected. The filling data of the blank depth interval is obtained by combining the content difference of the component elements.

[0041] Optionally, in a possible implementation of the first aspect, selecting adjacent sub-regions of the blank depth interval and combining content differences of their component elements to obtain filling data for the blank depth interval includes:

[0042] Selecting a subsample corresponding to a subinterval adjacent to the blank depth interval as a target sample;

[0043] Calculating the content difference between the same component elements in each of the target samples, and obtaining the content mean corresponding to the component elements whose content difference is less than the content difference threshold;

[0044] The content mean value is determined as the filling value of the corresponding component element in the blank depth interval, and the filling data of the blank depth interval is obtained according to the filling value.

[0045] A second aspect of the present invention provides a system for identifying and analyzing the formation mechanism of weak structures in deep thermal reservoirs, comprising:

[0046] A sample module is used to obtain a sample set collected from a weak structure area of a deep heat reservoir, and divide the depth interval of the sample set into sub-intervals of each subsample according to the collection order and length measurement results of the sample set;

[0047] An identification module, configured to identify the image data to obtain surface features of each subsample, and mark the subsample corresponding to the surface features that meet the sampling conditions;

[0048] The parameter module is used to generate deep geological parameters by combining the composition information and subintervals corresponding to the subsamples of fixed nodes and marked nodes;

[0049] The integration module is used to associate and integrate the deep geological parameters of the sub-samples to construct a comprehensive database.

[0050] The beneficial effects of the present invention are as follows:

[0051] 1. This method uses the acquisition sequence and sample length measurement results to finely divide the sample depth interval. It not only determines the initial depth interval when there is no gap between subsamples, but also identifies the interval between samples by comparing the edge and color features of adjacent subsamples. It also determines the predicted spacing based on the difference in features, ultimately obtaining accurate subsample subintervals. This method can more realistically reflect the actual situation of deep strata, providing a reliable data foundation for in-depth analysis of deep geological structures, and helping researchers more accurately understand the distribution characteristics of weak structures in deep heat reservoirs at different depths.

[0052] 2. Utilizing image recognition technology, this invention conducts in-depth analysis of sample surface features. By setting appropriate sampling conditions, such as color signature changes, it can accurately screen subsamples potentially associated with weak structures. Simultaneously, combined with component analysis equipment, it acquires sample composition information and generates deep geological parameters. This multi-dimensional, in-depth exploration of sample characteristics, from surface features to compositional information, significantly enhances the ability to identify weak structures in deep thermal reservoirs, thereby more accurately determining their location, scale, and nature.

[0053] 3. This invention systematically integrates deep geological parameters such as subsample composition information, precise depth, and predicted temperature to construct a comprehensive database. This database allows for rapid acquisition of required information. By comparing parameters across subsamples, potential connections between data can be explored, enabling in-depth research into the formation mechanisms and distribution patterns of weak structures in deep thermal reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of the present invention;

[0055] Figure 2 This is a flow chart of a method for identifying and analyzing the formation mechanism of weak structures in deep thermal reservoirs provided by an embodiment of the present invention;

[0056] Figure 3 It is a structural diagram of a system for identifying and analyzing the formation mechanism of weak structures in deep thermal reservoirs provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0058] See also Figure 1 , is a schematic diagram of an application scenario provided by an embodiment of the present invention. The present invention selects multiple exploration well locations in the deep heat storage exploration and development area, uses drilling steel pipes to collect samples, and during the drilling process, the rock samples are transported to the surface sample collection point through the spiral conveying device in the steel pipe. In addition, the sample length is measured in real time, and the sample depth is recorded in combination with the steel pipe driving sequence, so that the sample depth interval can be accurately divided, the sample surface characteristics and composition information can be analyzed in detail, and the location, scale and nature of the weak structure of the deep heat storage can be accurately identified. And combined with the database generated by these data, the causal mechanism of the weak structure of the deep heat storage can be accurately grasped, which is conducive to more efficient development of deep heat storage resources and increase the supply of clean energy.

[0059] See also Figure 2 , is a flow chart of a method for identifying and analyzing the formation mechanism of weak structures in deep thermal reservoirs provided by an embodiment of the present invention, Figure 2 The execution subject of the method shown may be a software and / or hardware device. The execution subject of the present application may include but is not limited to at least one of the following: user equipment, network equipment, etc. Among them, user equipment may include but is not limited to computers, smart phones, personal digital assistants (PDAs) and the electronic devices mentioned above. Network equipment may include but is not limited to a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. It includes steps S1 to S4, as follows:

[0060] S1, obtaining a sample set collected from a weak structure area of a deep heat reservoir, and dividing the depth interval of the sample set into subintervals of each subsample according to the collection sequence and length measurement results of the sample set.

[0061] Precisely determining the depth of each rock sample underground lays the foundation for subsequent analysis of the relationship between geological conditions at different depths and sample properties, helping to accurately identify the distribution of weak structures in deep thermal reservoirs. While each section of steel pipe has a rough drilling range, the depth of the rock source within it varies. By refining the depth range, the corresponding stratigraphic location of the sample can be precisely located, allowing for the study of the relationship between different stratigraphic properties and weak structures.

[0062] The sample set refers to the sum of all rock samples collected by each section of spliced steel pipe in the weak structure area of deep heat storage; the collection order refers to the order in which the samples are collected on the ground when the steel pipe is drilled section by section to collect samples; the length measurement result is the data obtained by measuring the length of each rock sample obtained from the steel pipe itself; the depth interval is the approximate depth range covered by each section of steel pipe during the drilling process. For example, each section of 10-meter-long steel pipe corresponds to a specific depth range after being driven into the ground. For example, the fifth section of steel pipe may correspond to a sample interval of 40-50 meters; a subsample refers to a single rock sample in the sample set; the subinterval is the precise depth interval corresponding to each subsample determined based on the collection order and sample length.

[0063] Based on the above embodiment, the specific implementation of "dividing the depth interval of the sample set to obtain subintervals of each subsample according to the acquisition order and length measurement results of the sample set" in step S1 can be:

[0064] S11 , based on the acquisition sequence and the length measurement result, cumulatively recording the depth of each of the subsamples based on the depth interval, and determining an initial depth interval when there is no gap between each of the subsamples.

[0065] In order to provide basic depth data, the underground location range of each subsample can be preliminarily determined without considering the possible intervals between samples, which facilitates subsequent more precise depth adjustment and analysis and understands the distribution of samples underground.

[0066] Cumulative depth records are calculated by adding up the sample depths according to the collection sequence, the length of each section of steel pipe, and the position of the sample within the pipe. The initial depth interval is the subsample depth range preliminarily determined based on the collection sequence and sample length, without considering the interval between samples.

[0067] For example, during the data preparation phase, if each section of steel pipe is known to be 10 meters long, sampling begins at a cumulative depth of 300 meters after 30 sections have been driven. Samples of 2 meters, 3 meters, and 1 meter in length are then taken. Based on the sampling sequence and length measurement results, the first sample's cumulative depth begins at 300 meters and its length is 2 meters, so its initial depth range is 300-302 meters. The second sample follows the first, with an initial depth range of 302-305 meters; and the third sample's initial depth range is 305-306 meters.

[0068] S12, when the total measured length of each subsample is less than the total length of the depth interval, comparing surface features corresponding to the image data of adjacent subsamples, and marking the subsample groups with gaps, wherein the surface features include edge features and color features.

[0069] In actual exploration, gaps may exist between samples. For example, if the total depth interval is assumed to be 10 meters, and the total measured length of the three subsamples is 6 meters, which is less than the total depth interval, this indicates a possible gap. By comparing the surface characteristics of adjacent subsamples to identify these gaps and marking the subsample groups with gaps, we can accurately adjust the sample depth intervals to ensure that the sample depth information better matches the actual geological conditions.

[0070] The total measured length of the subsample is the sum of the lengths of all collected subsamples; the total length of the depth interval is the length of the continuous depth range involved in this exploration; the surface features include the edge features and color features of the subsample; and the subsample group refers to a group of adjacent subsamples that are judged to have gaps.

[0071] Edge and color characteristics are used to determine whether gaps exist between rocks because these features can directly reflect the differences in geological processes during rock formation and transport, providing important clues for identifying gaps. Under normal circumstances, rocks deposited continuously or in close contact in underground geological environments should have edges that are relatively coherent and consistent. However, when gaps occur due to tectonic movement, erosion, or other factors, these edge characteristics can change significantly. For example, if one sample has a smooth edge while another has a jagged, non-natural connection, this is likely because the two rocks were not originally closely connected and were separated by other geological processes, creating a gap. Rock color is primarily determined by its mineral composition and formation environment. Rocks formed at different depths and under different geological conditions often have different colors. If adjacent rocks differ significantly in color, and the transition is not gradual, it is likely that a gap exists between them. For example, if the color of the first sample suddenly changes from gray to brown, this may indicate a significant change in the geological environment between the two rocks, such as a shift in redox conditions or a different source of sedimentary material. Such changes are often accompanied by a sedimentary hiatus or other geological event, suggesting the presence of a gap between the rocks.

[0072] In some embodiments, the following steps may be performed to compare surface features corresponding to image data of adjacent subsamples and mark subsample groups with gaps, including:

[0073] When comparing edge features, edge contours are extracted from the image data of adjacent subsamples; concave points and convex points in each edge contour are obtained, and the concave point difference and convex point difference in each edge contour are calculated. When the concave point difference and / or convex point difference are greater than a quantity threshold, it is determined that there is a gap in the subsample group and the subsamples are marked. When the concave point difference and the convex point difference are both less than the quantity threshold, a matching set of concave points and convex points with the smallest distance in each edge contour is obtained; the total distance of each matching set is counted. When the total distance is greater than a fit threshold, it is determined that there is a gap in the subsample group and the subsamples are marked.

[0074] The edge profile contains the shape and structure of the subsample edge and is the basis for analyzing edge feature differences. Concave and convex points are important feature points of the edge profile, and their number and distribution can reflect the degree of edge irregularity and differences. By calculating the difference between concave and convex points and comparing them with the number threshold, the degree of difference in the edge features of adjacent subsamples can be quickly determined. If the difference is too large, it is likely that there is a gap between the subsample groups, and these subsample groups can be marked.

[0075] When the difference between concave and convex points is small, it is difficult to determine whether there are gaps between subsample groups based solely on the difference in quantity. In this case, further analysis of the positional relationship between concave and convex points in the edge contour is necessary. Obtaining a matching set of concave and convex points with the smallest distance can further determine the degree of edge fit from the perspective of position matching, and more accurately determine whether there are gaps between subsample groups. The total distance of the matching set reflects the overall fit between concave and convex points on the edge contours of adjacent subsamples. By comparing the total distance with the fit threshold, the overall edge alignment can be determined. When the total distance is too large, it indicates low edge fit and a high probability of gaps between subsample groups. Therefore, these subsample groups can be marked.

[0076] Edge features refer to the shape, smoothness and other characteristics of the sample edge; the edge contour is the shape contour of the sample edge in the sample image; a concave point is a point where the edge contour is concave inward, and its curvature is negative; a convex point is a point where the edge contour protrudes outward, and its curvature is positive; the concave point difference is a quantitative value of the difference in the number of concave points in the edge contours of adjacent subsamples; the convex point difference is a quantitative value of the difference in the number of convex points in the edge contours of adjacent subsamples; the quantity threshold is a pre-set critical value for judging whether the concave point difference or the convex point difference is significant; the matching set is a set consisting of the concave points and convex points with the smallest distance in each edge contour, which is used to measure the degree of fit between adjacent sample edges; the fit threshold is a pre-set critical value for judging whether the total distance is too large.

[0077] Specifically, when analyzing adjacent subsamples, image processing techniques can be used to extract edge contours from the acquired image data. After obtaining the edge contours, the curvature of each point is calculated based on the series of points obtained by discretizing the contours. Concave and convex points are determined based on the sign of the curvature. For example, points with negative curvature are concave, and points with positive curvature are convex. The number of concave and convex points in the edge contours of adjacent subsamples is then counted, and their difference is calculated. Suppose the edge contour of subsample A has 8 concave points, and the edge contour of subsample B has 3 concave points, with a concave point difference of 5. If the pre-set threshold is 3 and 5 is greater than 3, then a gap is determined between the subsample groups containing subsamples A and B, and the gap is marked.

[0078] When both the concave point difference and the convex point difference are less than the quantity threshold, the matching set can be determined by the following steps:

[0079] For each concave point of each edge contour, the distance between it and each convex point in the other edge contour is calculated, and the convex point with the smallest distance is selected to generate a matching set; for each convex point of each edge contour, the distance between it and each concave point in the other edge contour is calculated, and the concave point with the smallest distance is selected to generate a matching set.

[0080] For each salient point in one contour, calculate its distance to all concave points in the other contour. Select the concave point with the smallest distance to form salient-concave matching set 1. For each concave point in one contour, calculate its distance to all salient points in the other contour. Select the salient point with the smallest distance to form concave-convex matching set 2. For example, if the edge contour of subsample C has a salient point P, calculate the distance between P and all concave points in the edge contour of subsample D. Find the concave point Q with the smallest distance and add (P, Q) to salient-concave matching set 1.

[0081] By obtaining matching sets, we supplement the judgment basis from the perspective of position matching, improving the accuracy of determining whether there are gaps in the subsample groups. Even if the difference in the number of concave and convex points is not obvious, the position relationship can still reveal potential gaps.

[0082] After calculating the matching sets, the sum of the distances of all point pairs is calculated using a predefined distance calculation method, such as Manhattan distance. For example, suppose there are five point pairs in both convex-concave matching set 1 and concave-convex matching set 2. The calculated Manhattan distances for these five pairs are 2, 3, 4, 3, and 2, respectively, for a total distance of 14. If 14 is greater than the predefined fit threshold of 10, then a gap is determined in the subsample group and marked as such.

[0083] By calculating the total distance and comparing it with the fit threshold, we can accurately determine whether there are gaps in the subsample groups. Marking subsample groups with gaps helps improve the accuracy of deep heat reservoir research.

[0084] In some other embodiments, the following steps may be performed to compare surface features corresponding to the image data of adjacent subsamples and mark the subsample groups with gaps, including:

[0085] When comparing color features, sample contours are extracted from the image data of adjacent subsamples; target areas whose pixel values in each sample contour are located in the same pixel interval and are located at the edge are obtained; the target area of each sample contour is compared with the target area of another sample contour. If there is no target area with the same pixel interval, it is determined that there is a gap in the subsample group and the gap is marked.

[0086] Color characteristics can reflect information such as the formation environment and composition changes of rocks. Under normal continuous sedimentation or close contact geological conditions, the color characteristics of adjacent subsamples should have a certain degree of consistency. If the color characteristics of adjacent subsamples are significantly different and there is no target area with the same pixel interval at the edge, this likely means that there are sedimentary discontinuities, geological structural changes, etc. between them, resulting in gaps between subsamples. This method can assist in determining whether there are gaps between subsample groups, provide a basis for determining the true distribution of rock samples in deep thermal reservoirs, and thus more accurately study the causal mechanism of weak structures in deep thermal reservoirs.

[0087] Color features refer to the color characteristics of the subsample in the image; sample contour refers to the boundary contour of the sample in the sample image; target area refers to the area in the sample contour where the pixel values are in the same pixel interval and are located at the edge. It is a key area for comparing color features because the edge part can better reflect the relationship between adjacent samples.

[0088] Suppose adjacent subsamples A and B are captured. First, extract the sample outlines of subsamples A and B from the captured image data. Next, identify regions with identical pixel values. For example, pixel values between 100 and 120 represent a specific color (assuming light gray). Within the edge of subsample A's sample outline, search for regions with pixel values between 100 and 120 to obtain target region A1. Similarly, search for regions with the same pixel range within the edge of subsample B's sample outline. If no region with pixel values between 100 and 120 is found within the edge of subsample B, meaning that no target region with the same pixel range exists within the edges of subsamples A and B, then a gap is determined between the subsample groups containing subsamples A and B, and this subsample group is marked. For example, if the edge of subsample A appears light gray, while the edge of subsample B appears dark gray, with completely different corresponding pixel values, this indicates a possible gap between the two.

[0089] S13: Determine the predicted distance of the subsample group according to the feature difference in the image data.

[0090] In order to accurately adjust the depth interval of the samples, it is necessary to determine the distance between the sample groups. By analyzing the feature differences in the image data to quantify the interval size, it can provide a basis for inserting the appropriate interval distance in the initial depth interval.

[0091] Feature difference is a quantitative value of the degree of difference obtained by comparing the surface features of adjacent subsamples, which is used to measure the size of the interval between subsample groups; the predicted interval is the predicted value of the interval distance between subsample groups calculated based on the feature difference.

[0092] In some embodiments, the predicted distance of the subsample group can be obtained by the following steps:

[0093] For edge features, the preset difference corresponding to the largest concave point difference or convex point difference is determined as the feature difference, or the preset difference corresponding to the total distance of the matching set is determined as the feature difference; for color features, the minimum color difference corresponding to multiple target areas adjacent to the subsamples is obtained, and the preset difference corresponding to the minimum color difference is determined as the feature difference; the length difference between the total length of the depth interval and the total measured length of each subsample is obtained, and the length difference is divided proportionally according to the ratio between the feature differences of the subsample groups to obtain the predicted spacing of each subsample group.

[0094] Accurately determining the spacing between subsample groups helps precisely delineate sample depth intervals, leading to a more accurate understanding of the distribution of rock samples within deep geothermal reservoirs and providing a reliable basis for studying the mechanisms underlying the formation of weak structures within these reservoirs. By analyzing the differences in edge and color features, combined with depth-related data, the spacing between sample groups can be quantified, improving the accuracy and scientific nature of the research.

[0095] The preset difference degree corresponding to the concave point difference value or convex point difference value is a numerical value set in advance for different concave point difference values and convex point difference values, representing the corresponding degree of difference. The larger the difference value, the higher the preset difference degree, which means that the difference in edge features is greater, the possibility of a gap between sample groups is greater, and the gap is likely to be larger. The preset difference degree corresponding to the total distance of the matching set refers to the preset difference degree set for different total distance values after calculating the total distance of the matching set composed of the concave points and convex points with the smallest distance in the edge contour. The larger the total distance, the higher the preset difference degree, indicating that the fit of the edge contour is lower and the possibility of a gap between sample groups is greater. The preset difference degree corresponding to the minimum color difference is the minimum color difference corresponding to multiple target areas when comparing the color features of adjacent sub-samples, and the preset difference degree is set for different minimum color differences. The larger the minimum color difference, the higher the preset difference degree, indicating that the color difference is greater and the possibility of a gap between sample groups is greater.

[0096] When calculating the predicted spacing, assume that the total depth interval length for each subsample group is 20 meters, the total measured length for each subsample is 16 meters, and the length difference is 4 meters. The feature difference between the subsample groups based on edge features is 0.7, and the feature difference based on color features is 0.3, with a ratio of 7:3. Using this ratio, the predicted spacing for the 4-meter length difference is 2.8 meters for the edge features and 1.2 meters for the color features. The sum of the feature difference ratios for each subsample group is 1.

[0097] S14: inserting the predicted interval into the initial depth interval corresponding to the subsample group, recalculating the cumulative depth of each subsample to obtain a subinterval of each subsample.

[0098] After determining the predicted spacing between sample groups, we insert it into the initial depth interval and recalculate the cumulative depth of the subsamples to obtain a more accurate subsample depth interval.

[0099] For example, if the first sample depth was originally 300-302 meters, and the second sample depth was 302-305 meters, and a 0.15-meter interval was determined between the two samples, this interval was inserted between the initial depths of the first and second samples. After this adjustment, the first sample depth remained 300-302 meters, the interval depth became 302-302.15 meters, and the second sample depth became 302.15-305 meters. The cumulative depth was then recalculated to ensure that the sum of the depths of all samples and intervals matched the actual depth. This process resulted in more accurate subintervals for each subsample.

[0100] By inserting the predicted intervals and recalculating the cumulative depth, the obtained subintervals are more consistent with the actual geological conditions, which improves the accuracy of the sample depth information.

[0101] S2, identifying the image data to obtain surface features of each sub-sample, and marking the sub-samples corresponding to the surface features that meet the sampling conditions.

[0102] The formation of weak structures in deep thermal reservoirs often creates distinctive surface features in the surrounding rocks. Identifying these surface features in subsamples and labeling those that meet specific criteria allows us to identify samples potentially associated with weak structures, avoiding the need for extensive analysis of common samples. This improves research efficiency and allows us to more quickly identify clues related to weak structures.

[0103] Specifically, after a sample is collected and dropped onto the ground and into a sample collection box, a high-definition image acquisition device automatically captures each subsample from multiple angles, acquires image data, and analyzes the image data. Suppose that a sampling condition is satisfied when a subsample's color changes, for example, from the light gray of the surrounding samples to a dark gray. If a subsample collected from Section 9 exhibits this characteristic change, the system automatically marks it as a special sample and records its subinterval.

[0104] Image data is digitally captured from multiple angles of surface-collected subsamples using high-definition image acquisition equipment mounted above the sample collection box. Surface features include external characteristics such as subsample color and edge morphology. Sampling conditions are pre-defined criteria used to determine whether subsample surface features are unique or potentially related to weak structures in deep thermal reservoirs. Characteristics such as sudden color changes may be included as sampling conditions.

[0105] Based on the above embodiment, the specific implementation of step S2 may be:

[0106] Each of the subsamples is selected in sequence according to the acquisition order, and the color features contained in the surface features in the image data of the subsample are identified; the pixel mean of the subsample is determined based on the color features, and the pixel difference between the pixel mean of the current subsample and the previous subsample is calculated; when the pixel difference is within the abnormal difference interval, it is determined that the current subsample meets the sampling conditions and is marked.

[0107] By analyzing the color characteristics of subsamples in the order they were collected, calculating the pixel mean difference, and comparing it with the abnormal difference interval, we can screen out subsamples with abnormal color changes. These subsamples may represent sudden changes in geological conditions and are associated with the formation or distribution of weak structures in deep thermal reservoirs. Marking these subsamples will help focus subsequent research on them, improving research efficiency and accuracy.

[0108] The abnormal difference interval refers to a pre-set range used to determine whether the pixel difference is abnormal. If the pixel difference is within this interval, it is considered that the subsample color change is significant, which may be related to the special geological conditions of the deep heat reservoir.

[0109] Suppose that in a series of subsamples collected in the order in which they were collected, the first subsample is selected. Image recognition technology is used to analyze its image data and extract color features. Assume that the calculated pixel mean for this subsample is 120. Next, the second subsample is selected, and color features are similarly extracted and the pixel mean is calculated, assuming it is 150. The pixel difference between these two subsamples is calculated to be 30. Since 30 falls within the pre-set abnormal difference interval of [20, 40], the second subsample is determined to meet the sampling criteria and is automatically marked by the system. Subsequent researchers can further analyze the composition, structure, and other characteristics of this marked subsample to explore its relationship to the weak structure of deep thermal reservoirs.

[0110] Through the screening method based on color features, sub-samples with abnormal color feature changes can be accurately found, avoiding unnecessary in-depth analysis of a large number of ordinary samples and saving time.

[0111] S3, combining the composition information and subintervals corresponding to the subsamples of the fixed nodes and the marked nodes to generate deep geological parameters.

[0112] Integrating the composition information, precise depth and other information of sub-samples to generate parameters that can intuitively reflect the deep geological conditions can provide key data support for the construction of a deep thermal reservoir geological database.

[0113] Fixed nodes are subsample locations corresponding to specific depths of interest, designated by researchers based on their needs. These locations are used to focus on the geological conditions at that depth. Marked nodes are subsample locations that have been marked because their surface features meet the sampling criteria. Compositional information is obtained by analyzing subsamples at fixed and marked nodes using compositional analysis equipment, such as spectrometers, to obtain data on the chemical composition and content of each component. Deep geological parameters are generated by integrating various information, including compositional information of subsamples and subintervals, to describe deep geological characteristics. Common parameters include predicted temperature.

[0114] The generated deep geological parameters organically combine the various attributes of the subsamples, providing intuitive and quantitative data for studying the geological structure, reservoir characteristics, and the causes of weak structures in deep thermal reservoirs. Using these parameters, researchers can analyze the intrinsic relationship between samples of varying depths and compositions and the weak structures in deep thermal reservoirs. For example, they can determine the temperature conditions under which rocks with specific compositional combinations are more likely to form weak structures, providing a scientific basis for the exploration and development of deep thermal reservoirs.

[0115] Based on the above embodiment, the specific implementation of step S3 may be:

[0116] According to the corresponding relationship between the composition information and the temperature, the predicted temperature corresponding to the corresponding sub-interval is determined, and according to the mapping relationship between the sub-interval and the predicted temperature, the deep geological parameters are generated.

[0117] The geological characteristics of deep thermal reservoirs are complex and diverse, with composition and temperature being two key elements. By establishing a correspondence between composition and temperature, determining predicted temperatures for different sub-intervals, and generating deep geological parameters, we can integrate information such as the composition, depth, and temperature of rock samples. This helps researchers gain a more comprehensive and in-depth understanding of the geological conditions of deep thermal reservoirs and provides key data support for exploring the mechanisms underlying the formation of weak structures within these reservoirs.

[0118] Compositional information refers to the chemical composition and content of each component obtained by testing rock samples using compositional analysis equipment (such as spectrometers). This information reflects the material composition of the rock. Predicted temperature refers to the temperature value of a specific subrange, inferred from the composition of the rock sample within that range, based on the corresponding relationship between compositional information and temperature. Because measuring temperatures at different locations in deep heat reservoirs is difficult, predicted temperatures can provide an important reference for research.

[0119] Integrating compositional information, depth, and temperature information into deep geological parameters allows researchers to gain a more intuitive and comprehensive understanding of the geological characteristics of deep heat reservoirs, avoiding fragmented and isolated information and helping to discover potential connections between different geological elements. The generated deep geological parameters not only provide an important basis for the exploration and development of deep heat reservoirs, but also provide key data for exploring the causal mechanisms of weak structures in deep heat reservoirs. Different temperature and composition conditions affect the physical and mechanical properties of rocks. By analyzing deep geological parameters, we can study how these factors interact with each other, thereby affecting the formation and distribution of weak structures, and promote a deeper understanding of the geological processes of deep heat reservoirs.

[0120] S4, correlating and integrating the deep geological parameters of the subsamples to construct a comprehensive database.

[0121] The systematic integration and storage of dispersed deep geological parameters in a single database facilitates centralized data management, efficient querying, and in-depth analysis. Researchers can quickly access the required information from the comprehensive database and, by comparing parameters across different subsamples, uncover potential connections between data, enabling in-depth research into the formation mechanisms and distribution patterns of weak structures in deep thermal reservoirs.

[0122] The comprehensive database is a collection specifically used to store deep geological parameters of all sub-samples. It has functions such as data storage, classification management, and fast query, and can meet the researchers' needs for processing and analyzing large amounts of geological data.

[0123] For example, the deep geological parameters of all previously generated subsamples can be entered into a database, using the corresponding pipe section number and subinterval as indexes. For example, the database records information such as subsample 1 (from pipe section 6, subinterval [59.2, 59.5] meters, primarily composed of quartz and mica, with a predicted temperature of 48°C) and subsample 2 (from pipe section 13, subinterval [129.1, 129.4] meters, primarily composed of feldspar, with a predicted temperature of 62°C). Researchers can quickly query other parameter information for related subsamples by entering various conditions, such as pipe section number, subinterval range, and composition type.

[0124] In addition, based on the above embodiments, this solution also includes the following embodiments:

[0125] Obtain scanning information of sample data of each well location in the deep heat storage weak structure area by inspection equipment; determine the blank depth interval where sample data is missing based on the scanning information, select the sub-areas adjacent to the blank depth interval, and combine the content difference of their component elements to obtain the filling data of the blank depth interval.

[0126] Understandably, sample data is crucial for analyzing geological structures and their mechanisms. However, due to various reasons, such as equipment failure and complex geological conditions, sample data may be missing. Obtaining scan information from inspection equipment and processing missing data can supplement sample data, ensuring the accuracy and continuity of research. This allows researchers to explore the characteristics and patterns of weak structures in deep thermal reservoirs based on more comprehensive data.

[0127] Inspection equipment refers to equipment used to scan and collect sample data from each well location in the weak structural area of deep thermal reservoirs, such as an inspection robot, which can obtain a variety of sample information, such as composition, images, etc. Scanning information refers to the digital information obtained after the inspection equipment scans the sample data from each well location, which contains various attribute data of the sample. The blank depth interval refers to the depth range corresponding to the missing sample data. The content difference of the component element refers to the difference in the content of the same component element in the samples of adjacent sub-areas, which is used to evaluate the composition changes in adjacent areas and provide a basis for filling the blank depth interval data. Filling data refers to the data estimated and generated for the blank depth interval based on information such as the content difference of the component element in the samples of adjacent sub-areas, in order to supplement the missing sample data.

[0128] In some embodiments, the following steps may be performed to select adjacent sub-regions of the blank depth interval and combine the content differences of their component elements to obtain filling data for the blank depth interval, including:

[0129] Select subsamples corresponding to subintervals adjacent to the blank depth interval as target samples; calculate the content difference between the same component elements in each of the target samples, and obtain the content mean corresponding to the component elements whose content difference is less than the content difference threshold; determine the content mean as the filling value of the corresponding component element in the blank depth interval, and obtain the filling data of the blank depth interval according to the filling value.

[0130] Suppose there is a blank depth interval of 200-210 meters. Adjacent subintervals to this blank depth interval are 190-200 meters and 210-220 meters, respectively. Subsample A (from the 190-200 meter interval) and subsample B (from the 210-220 meter interval) are selected as target samples. Compositional analysis of these two target samples reveals the presence of silicon, aluminum, and iron. The calculated silicon content in subsample A is 30% and in subsample B is 32%, a difference of 2%. The aluminum content in subsample A is 20% and in subsample B is 25%, a difference of 5%. The iron content in subsample A is 10% and in subsample B is 11%, a difference of 1%. The pre-set content difference threshold is 3%. The difference between the silicon and iron content is less than the content difference threshold. The calculated average silicon content is 31%, and the average iron content is 10.5%. 31% is determined as the fill value for silicon in the blank depth interval, and 10.5% is determined as the fill value for iron. Similarly, fill values for other elements that meet the conditions are determined, resulting in fill data for the blank depth interval of 200-210 meters.

[0131] Component elements refer to the various chemical elements that make up a rock sample, such as silicon, aluminum, and iron. The content of different components reflects the characteristics of the rock. The content difference refers to the difference in the content of the same component element between target samples and is used to measure the degree of variation in the content of the component element between adjacent subsamples. The content difference threshold is a pre-set value used to determine whether the content difference is within an acceptable range. When the content difference is less than the threshold, the content variation of the component element between adjacent subsamples is considered small and can be used to calculate the content mean.

[0132] By obtaining scanning information to determine the blank depth interval and filling it with adjacent sub-area data, the missing sample data is effectively supplemented and the integrity of the data is ensured.

[0133] See also Figure 3 , is a schematic structural diagram of a system for identifying and analyzing the genetic mechanism of weak structures in deep thermal reservoirs provided by an embodiment of the present invention. The system for identifying and analyzing the genetic mechanism of weak structures in deep thermal reservoirs includes:

[0134] A sample module is used to obtain a sample set collected from a weak structure area of a deep heat reservoir, and divide the depth interval of the sample set into sub-intervals of each subsample according to the collection order and length measurement results of the sample set;

[0135] An identification module, configured to identify the image data to obtain surface features of each subsample, and mark the subsample corresponding to the surface features that meet the sampling conditions;

[0136] The parameter module is used to generate deep geological parameters by combining the composition information and subintervals corresponding to the subsamples of fixed nodes and marked nodes;

[0137] The integration module is used to associate and integrate the deep geological parameters of the sub-samples to construct a comprehensive database.

[0138] Figure 3 The apparatus of the embodiment shown can be used to perform Figure 2 The implementation principles and technical effects of the steps in the method embodiment shown are similar and will not be repeated here.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying and analyzing the formation mechanism of weak structures in deep thermal reservoirs, characterized by: include: Obtain a sample set collected from a weak structure area of a deep heat reservoir, and divide the depth interval of the sample set into subintervals of each subsample according to the collection order and length measurement results of the sample set, including: Based on the acquisition sequence and the length measurement result, cumulatively recording the depth of each of the subsamples based on the depth interval, and determining an initial depth interval when there is no gap between each of the subsamples; When the total measured length of each subsample is less than the total length of the depth interval, comparing surface features corresponding to the image data of adjacent subsamples and marking the subsample groups with gaps, the surface features including edge features and color features; Determining the predicted spacing of the subsample groups according to the feature difference in the image data; Inserting the predicted interval into the initial depth interval corresponding to the subsample group, recalculating the cumulative depth of each subsample to obtain a subinterval of each subsample; Recognizing the image data to obtain surface features of each sub-sample, and marking the sub-samples corresponding to the surface features that meet the sampling conditions, including: Selecting each of the subsamples in sequence according to the acquisition order, and identifying color features included in surface features in the image data of the subsamples; determining a pixel mean of the subsample based on the color feature, and calculating a pixel difference between the pixel mean of a current subsample and the pixel mean of a previous subsample; When the pixel difference is within the abnormal difference interval, determining that the current subsample meets the sampling condition and marking it; Combining the composition information and subintervals corresponding to the subsamples of fixed nodes and marked nodes, deep geological parameters are generated, including: Determine the predicted temperature corresponding to the corresponding subinterval based on the corresponding relationship between the composition information and the temperature, and generate deep geological parameters based on the mapping relationship between the subinterval and the predicted temperature; The deep geological parameters of the subsamples are correlated and integrated to construct a comprehensive database.

2. The method according to claim 1, characterized in that Comparing surface features corresponding to image data of adjacent subsamples and marking subsample groups with gaps therebetween comprises: When comparing edge features, edge contours are extracted from image data of adjacent subsamples; Obtaining concave points and convex points in each of the edge contours, calculating a concave point difference value and a convex point difference value in each of the edge contours, and when the concave point difference value and / or the convex point difference value is greater than a quantity threshold, determining that there is a gap in the subsample group and marking the gap; When the concave point difference and the convex point difference are both less than the quantity threshold, obtaining a matching set of concave points and convex points with the smallest distance in each of the edge contours; The total distance of each matching set is counted, and when the total distance is greater than a fit threshold, it is determined that there is a gap between the subsample groups and the gap is marked.

3. The method according to claim 2, characterized in that Obtaining a matching set of concave points and convex points with the smallest distance in each edge contour, including: For each concave point of each edge contour, calculate the distance between it and each convex point in another edge contour, and select the convex point with the smallest distance to generate a matching set; For each convex point of each edge contour, the distance between the convex point and each concave point in another edge contour is calculated, and the concave point with the smallest distance is selected to generate a matching set.

4. The method according to claim 1, wherein Comparing surface features corresponding to image data of adjacent subsamples and marking subsample groups with gaps therebetween comprises: When comparing color features, extracting sample contours from image data of adjacent subsamples; Acquire a target area in which pixel values of each sample contour are located in the same pixel interval and at the edge; The target area of each sample outline is compared with the target area of another sample outline. If there is no target area with the same pixel interval, it is determined that there is a gap between the subsample groups and the gap is marked.

5. The method according to claim 1, wherein Determining the predicted spacing of the subsample groups according to the feature difference in the image data includes: For edge features, the preset difference corresponding to the largest concave point difference or convex point difference is determined as the feature difference, or the preset difference corresponding to the total distance of the matching set is determined as the feature difference; For the color feature, obtaining the minimum color difference corresponding to a plurality of target areas adjacent to the subsamples, and determining a preset difference corresponding to the minimum color difference as the feature difference; The length difference between the total length of the depth interval and the total measured length of each subsample is obtained, and the length difference is proportionally divided according to the ratio between the characteristic differences of the subsample groups to obtain the predicted spacing of each subsample group.

6. The method according to claim 1, characterized in that Also includes: Obtaining scanning information of sample data of each well location in the deep heat storage weak structure area by inspection equipment; A blank depth interval where sample data is missing is determined based on the scanning information, and sub-regions adjacent to the blank depth interval are selected. The filling data of the blank depth interval is obtained by combining the content difference of the component elements.

7. The method according to claim 6, characterized in that Selecting adjacent sub-regions of the blank depth interval and combining the content differences of their component elements to obtain filling data for the blank depth interval includes: Selecting a subsample corresponding to a subinterval adjacent to the blank depth interval as a target sample; Calculating the content difference between the same component elements in each of the target samples, and obtaining the content mean corresponding to the component elements whose content difference is less than the content difference threshold; The content mean value is determined as the filling value of the corresponding component element in the blank depth interval, and the filling data of the blank depth interval is obtained according to the filling value.

Images