Geological space modeling method and device, electronic equipment and storage medium

Abstract

The application provides a geological space modeling method and device, electronic equipment and storage medium. The method comprises: performing a plurality of rounds of geological information screening operations on a predetermined geological space comprising a plurality of voxels; determining a plurality of presumed geological information of each other voxel except all partial voxels in the plurality of voxels; determining a target information confidence degree corresponding to the other voxel, and determining whether to update the plurality of presumed geological information corresponding to the other voxel; in response to determining to update the plurality of presumed geological information corresponding to the other voxel, updating the plurality of presumed geological information corresponding to at least one other voxel to corresponding updated geological information based on the plurality of presumed geological information corresponding to the at least one other voxel and the target information confidence degree, and a plurality of sets of geological information sets; and modeling the predetermined geological space based on all actual geological information, all presumed geological information and all updated geological information, thereby solving the technical problem of poor modeling accuracy of the geological space in the prior art.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a geological spatial modeling method, apparatus, electronic device and storage medium. Background Technology

[0002] As tunnel engineering expands to deeper, longer, larger cross-sections and higher-risk geological areas, construction safety risks have increased significantly. Traditional geological exploration methods (such as single drilling) are limited by exploration depth and construction interference, making it difficult to obtain continuous and detailed geological information. Engineering practice relies heavily on geological modeling technology for advanced prediction.

[0003] The existing geological modeling technology modeling process is "a single geological information set". Manual correction The aforementioned geological modeling techniques have the following defects: (i) Existing techniques are based on a single set of geological information and do not verify the stability of interpolation results through multiple sets of geological information. The absence or deviation of a single borehole location will directly lead to the overall deviation of the model. (ii) Existing techniques handle areas with uncertain attributes (such as sparse borehole areas) through manual correction or static assignment, without combining surrounding high-reliability voxel constraints and statistical optimization methods. This results in highly subjective and biased optimization results, failing to guarantee geological continuity. These defects in geological modeling techniques lead to poor geological modeling accuracy. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a geological spatial modeling method, apparatus, electronic device and storage medium to overcome all or part of the shortcomings of the prior art.

[0005] To achieve the above objectives, this application provides a geological space modeling method, comprising: performing multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space, wherein each set of geological information includes actual geological information corresponding to some voxels; determining multiple inferred geological information for each other voxel other than all some voxels based on the entire set of geological information; for each other voxel, determining a target information confidence level corresponding to the other voxel based on the multiple inferred geological information corresponding to the other voxel, and determining whether to update the multiple inferred geological information corresponding to the other voxel based on the target information confidence level; in response to determining to update the multiple inferred geological information corresponding to at least one other voxel, updating the multiple inferred geological information corresponding to the at least one other voxel to its corresponding updated geological information based on the multiple inferred geological information corresponding to the at least one other voxel, the target information confidence level, and the multiple sets of geological information; and modeling the predetermined geological space based on all actual geological information, all inferred geological information, and all updated geological information.

[0006] Optionally, the updated geological information is first updated geological information or second updated geological information; the step of updating the multiple inferred geological information corresponding to the at least one other voxel to its corresponding updated geological information based on the confidence levels of the multiple inferred geological information and target information corresponding to the at least one other voxel, and the multiple sets of geological information sets, includes: for each of the at least one other voxel, in response to determining that the confidence level of the target information corresponding to the other voxel is less than a first predetermined confidence level and greater than or equal to a second predetermined confidence level, determining the other voxel as a first other voxel, or, in response to determining that the confidence level of the target information is less than a first predetermined confidence level and greater than or equal to a second predetermined confidence level, determining the other voxel as a first other voxel, or, in response to determining the confidence level of the target information... If the confidence level of the information is less than the second predetermined confidence level, the other voxels are determined as second other voxels; each actual geological information and each inferred geological information in the multiple sets of geological information are taken as target geological information; for each first other voxel, at least one target geological information is obtained within a sphere with the first other voxel as the center and a first predetermined distance as the radius; based on the at least one target geological information corresponding to the first other voxel, a first predetermined algorithm is used to determine the first updated geological information corresponding to the first other voxel, and multiple inferred geological information corresponding to the first other voxel are replaced with the first updated geological information; for each second other voxel, at least one target geological information is obtained within a sphere with the second other voxel as the center and a second predetermined distance as the radius; based on the at least one target geological information and multiple inferred geological information corresponding to each second other voxel, multiple rounds of information update operations are performed to obtain the second updated geological information corresponding to each second other voxel, and multiple inferred geological information corresponding to each second other voxel are replaced with its corresponding second updated geological information.

[0007] Optionally, the step of performing multiple rounds of information update operations based on at least one target geological information and multiple inferred geological information corresponding to each second other voxel to obtain second updated geological information corresponding to each second other voxel includes: each round of information update operations is performed as follows: randomly selecting one inferred geological information from the multiple inferred geological information corresponding to each second other voxel to obtain a random combination containing multiple inferred geological information corresponding to the second other voxel; determining an unreasonable score corresponding to the random combination based on the random combination and at least one target geological information corresponding to each second other voxel; updating the unreasonable score to the predetermined optimal benchmark value in the next round of information update operations in response to determining that the unreasonable score is less than the predetermined optimal benchmark value in the current round of information update operations; or, in response to determining that the unreasonable score is greater than or equal to the predetermined optimal benchmark value in the current round of information update operations, using the predetermined optimal benchmark value in the current round of information update operations as the next round of information update. The operation involves determining a predetermined optimal baseline value; in response to determining that the number of completed information update operations is less than or equal to a first predetermined number, executing the next round of information update operations; in response to determining that the number of completed information update operations is greater than the first predetermined number, detecting whether there are any target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal baseline value, wherein the multiple rounds of target information update operations include the current round of information update operations and a predetermined number of consecutive historical information update operations adjacent to it; in response to determining that there are target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal baseline value, executing the next round of information update operations; in response to determining that there are no target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal baseline value, taking the inferred geological information corresponding to each second other voxel contained in the random combination of the current round of information update operations as the second updated geological information corresponding to each second other voxel, and exiting at least one round of information update operations.

[0008] Optionally, the step of performing multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space includes: obtaining a drilling combination containing multiple drilling information; each round of geological information filtering operation is performed as follows: determining the drilling information of each predetermined number in the drilling combination as a set of geological information; in response to determining that the number of all geological information sets is less than or equal to a second predetermined number, updating some predetermined numbers in all predetermined numbers in the current round of geological information filtering operation, and performing the next round of geological information filtering operation; in response to determining that the number of all geological information sets is greater than the second predetermined number, exiting at least one round of geological information filtering operation.

[0009] Optionally, determining multiple inferred geological information for each other voxel (excluding all partial voxels) among the multiple voxels based on the entire geological information set includes: for each geological information set, determining the inferred geological information for other voxels corresponding to the geological information set using a second predetermined algorithm, so as to obtain multiple inferred geological information for each other voxel.

[0010] Optionally, determining the target information confidence level corresponding to the other voxels based on multiple inferred geological information corresponding to the other voxels includes: extracting attributes from the multiple inferred geological information corresponding to the other voxels to obtain multiple attributes corresponding to the other voxels; calculating the initial information confidence level corresponding to each attribute based on the multiple attributes; and taking the maximum initial information confidence level among all initial information confidence levels as the target information confidence level corresponding to the other voxels.

[0011] Optionally, determining the unreasonable score corresponding to the random combination based on the random combination and at least one target geological information corresponding to each second other voxel includes: for each second other voxel, determining the deviation score of the inferred geological information corresponding to the second other voxel in the random combination; based on the inferred geological information corresponding to the second other voxel in the random combination and at least one target geological information corresponding to the second other voxel, determining the consistency score of the inferred geological information corresponding to the second other voxel in the random combination; determining the sum of the deviation score and the consistency score as the unreasonable score of the inferred geological information corresponding to the second other voxel in the random combination; and determining the sum of the unreasonable scores of all inferred geological information corresponding to the random combination as the unreasonable score of the random combination.

[0012] Based on the same inventive concept, this application also provides a geological space modeling apparatus, comprising: a screening module configured to perform multiple rounds of geological information screening operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space, wherein each set of geological information includes actual geological information corresponding to some voxels; a first determining module configured to determine multiple inferred geological information for each other voxel other than all some voxels in the multiple voxels based on the entire set of geological information; and a second determining module configured to determine, for each other voxel, based on the multiple inferred geological information corresponding to the other voxel, the other... The system comprises: a target information confidence level corresponding to a voxel, and a determination, based on the target information confidence level, whether to update multiple inferred geological information corresponding to other voxels; an update module, configured to, in response to determining to update multiple inferred geological information corresponding to at least one other voxel, update the multiple inferred geological information corresponding to the at least one other voxel to its corresponding updated geological information based on the multiple inferred geological information corresponding to the at least one other voxel, the target information confidence level, and the multiple sets of geological information; and a modeling module, configured to model the predetermined geological space based on all actual geological information, all inferred geological information, and all updated geological information.

[0013] Based on the same inventive concept, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method described above when executing the computer program.

[0014] Based on the same inventive concept, this application also provides a non-transitory computer-readable storage medium that stores computer instructions for causing a computer to perform the method described above.

[0015] As can be seen from the above, the geological spatial modeling method, apparatus, electronic device, and storage medium provided in this application include a method that involves performing multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space. Each set of geological information includes actual geological information corresponding to some voxels, avoiding the uniformity of the geological information set and laying a data foundation for subsequent cross-validation of drilling data. Based on the entire geological information set, multiple inferred geological information for each other voxel (excluding all partial voxels) is determined, effectively reducing errors and biases that may arise from a single information source. For each other voxel, based on the multiple inferred geological information corresponding to the other voxel, the confidence level of the target information corresponding to the other voxel is determined, and based on the confidence level of the target information, it is determined whether to update the multiple inferred geological information corresponding to the other voxel. By determining the confidence level of the target information corresponding to other voxels, the reliability of the multiple inferred geological information corresponding to other voxels is accurately quantified, solving the core problem in existing geological modeling techniques where the reliability of inferred geological information cannot be quantified. Furthermore, it eliminates the need for human verification of the accuracy of the inferred geological information, avoiding the tediousness and uncertainty associated with manual intervention, and effectively improving the efficiency and accuracy of geological modeling. In response to determining and updating multiple inferred geological information corresponding to at least one other voxel, based on the confidence level of the multiple inferred geological information corresponding to the at least one other voxel and the multiple sets of geological information, the multiple inferred geological information corresponding to the at least one other voxel is updated to its corresponding updated geological information. Utilizing appropriate technical means and data from multiple sources ensures the accuracy of determining the multiple inferred geological information corresponding to at least one other voxel. Based on all actual geological information, all inferred geological information, and all updated geological information, the predetermined geological space is modeled, ensuring the accuracy of modeling the predetermined geological space. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the geological spatial modeling method according to an embodiment of this application;

[0018] Figure 2 This is a flowchart illustrating a geological spatial modeling method according to another embodiment of this application; Figure 3This is a schematic diagram of the geological space modeling device according to an embodiment of this application; Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0020] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0021] As described in the background section, with tunnel engineering expanding to deeper, longer, larger cross-sections and higher-risk geological areas, factors such as the degree of fragmentation, water-bearing capacity, and distribution of weak interlayers in the surrounding rock structure exhibit strong spatial uncertainties, significantly increasing construction safety risks. Traditional geological exploration methods (such as single drilling) are limited by exploration depth and construction interference, making it difficult to obtain continuous and detailed geological information. Engineering practice heavily relies on geological modeling technology for advanced prediction. However, "unclear geology and insufficient advanced prediction" have become the core root cause of engineering accidents such as collapses and water inrushes. Therefore, the accuracy of geological spatial modeling is crucial.

[0022] The existing geological modeling process is as follows: ① Determine the modeling scope and divide it into grids; ② Generate a single 3D geological model based on limited drilling information using spatial interpolation (such as kriging); ③ Manually adjust obviously unreasonable areas in the model (such as lithological abrupt changes) using ground-penetrating radar or drilling data; ④ Output a static model for construction guidance. This is essentially a "single geological information set". Manual correction The aforementioned geological modeling techniques have the following defects: (I) Existing technologies are based on a single geological information set, relying on a single geological information set or a single data source. Because drilling information is a discrete truth point, a single geological information set cannot cover the impact of missing borehole locations. The lack of multiple sets of cross-validation results in the reliability of voxel attributes not being quantifiable. Construction teams cannot determine which areas in the model are reliable and which are questionable, leading to blind reliance on the model and significantly increasing decision-making risks. The absence or deviation of a single borehole location directly results in poor geological modeling accuracy. (II) Existing technologies handle areas with uncertain attributes (such as sparse borehole areas) through manual correction or static methods. The lack of dynamic assignment, coupled with the absence of constraints from surrounding high-reliability voxels and statistical optimization methods, and the absence of a standardized process, leads to highly subjective and biased optimization results. This compromises geological continuity and further reduces the accuracy of geological modeling. Furthermore, the lack of a dynamic convergence mechanism makes it difficult to ensure the model reaches optimal reliability levels. The absence of deep integration with reliability quantification mechanisms also hinders the full utilization of complementary data. Moreover, multi-source data is not integrated into the reliability quantification system, serving only as an auxiliary correction tool, thus failing to fully leverage data complementarity. This results in insufficient overall model confidence, making it difficult to support accurate decision-making in high-risk tunnel engineering. These deficiencies in geological modeling techniques contribute to poor geological modeling accuracy.

[0023] In view of this, embodiments of this application propose a geological spatial modeling method, referring to... Figure 1 This includes the following steps: Step 101: Perform multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space, wherein each set of geological information includes actual geological information corresponding to some voxels.

[0024] In this step, the predetermined geological space is the underground space to be constructed. It is continuous, complex, and invisible, and is determined according to engineering requirements. Directly modeling the geological space would lead to chaotic geological information throughout the space. Therefore, the predetermined geological space is first spatially voxelized to obtain multiple voxels corresponding to the predetermined geological space. Spatial voxelization involves dividing the predetermined geological space into a predetermined number of subspaces, each of which is a voxel. The predetermined geological space is 3D, and a "voxel" is a small cube. For example, the underground space is divided into thousands of uniformly sized "small cubes (voxels)". This transforms the originally vague and continuous predetermined geological space into "digital grids" that can be individually labeled and have information stored. Each voxel corresponds to the geological information of its location. An attribute array is defined for each voxel unit, containing two fields: geological attribute value and confidence level.

[0025] The multiple actual geological information corresponding to the predetermined geological space are real data obtained through exploration, using multiple boreholes to explore the predetermined geological space. It should be noted that when modeling the predetermined geological space, it is impossible to actually drill all spaces within the predetermined geological space; the geological information of un-drilled geological spaces depends on the inference from the geological information of drilled geological spaces. In existing technologies, geological modeling is based on a single set of geological information, that is, using only one set of geological information containing all actual geological data to directly infer the geological information of other un-drilled geological spaces within the predetermined geological space. Because drilling data are discrete ground truth points, a single set of geological information cannot cover the influence of missing borehole locations and lacks multiple sets of cross-validation. This application, however, performs multiple rounds of geological information filtering operations on the predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space. Each set of geological information includes actual geological information corresponding to some voxels, and some voxels in each set are not completely identical. Each set of geological information can be obtained by spatial interpolation of borehole combinations that lack single borehole information from all drilling data. Each set also includes actual geological information obtained from the actual drilling corresponding to some voxels. This application employs a combined sampling strategy to construct multiple sets of geological information, avoiding the uniformity of the sets and laying a data foundation for subsequent cross-validation of drilling data. It also avoids model failure caused by single-hole location bias and provides a statistical basis for confidence level calculations.

[0026] Step 102: Based on the complete set of geological information, determine multiple inferred geological information for each other voxel in the plurality of voxels, excluding all partial voxels.

[0027] In this step, each voxel in each geological information set corresponds to actual geological information, which is data obtained through actual drilling. For each set of geological information, one inferred geological information can be deduced for each remaining voxel other than the voxel corresponding to that geological information set. Since each remaining voxel corresponding to that geological information set may be a voxel corresponding to another geological information set, there is no need to further verify the reliability of the inferred geological information of the aforementioned voxels. Therefore, by removing the voxels corresponding to other geological information sets from each remaining voxel corresponding to that geological information set, the other voxels corresponding to that geological information set are obtained, and the inferred geological information corresponding to the other voxels is retained. Since one inferred geological information for other voxels can be determined based on a set of geological information, multiple inferred geological information for each other voxel other than all voxels in multiple voxels are determined based on the entire geological information set.

[0028] Each set of geological information infers the inferred geological information of other voxels based on the actual geological information of different partial voxels, and the distribution of partial voxels varies among different sets of geological information. Compared with the prior art, which only determines one inferred geological information for each other voxel, this application can effectively reduce the errors and biases that may be caused by a single information source by determining multiple inferred geological information for each other voxel.

[0029] Step 103: For each other voxel, based on the multiple speculative geological information corresponding to the other voxel, determine the confidence level of the target information corresponding to the other voxel, and based on the confidence level of the target information, determine whether to update the multiple speculative geological information corresponding to the other voxel.

[0030] In this step, for each other voxel, since there are multiple inferred geological information corresponding to that voxel, some of these inferred geological information may be inaccurate. Therefore, it is necessary to verify the accuracy of the multiple inferred geological information corresponding to that other voxel. Based on the multiple inferred geological information corresponding to other voxels, the confidence level of the target information corresponding to other voxels is determined, and the reliability of the multiple inferred geological information corresponding to other voxels is precisely quantified. The information confidence level is used to characterize the reliability of the multiple inferred geological information corresponding to other voxels. By utilizing the multiple inferred geological information corresponding to other voxels to determine the confidence level of the target information corresponding to other voxels, the problem of "lack of multiple sets of verification" in existing technologies is solved. The core drawback of "unquantifiable reliability" is addressed by this application, which utilizes multiple sets of geological information to cross-validate multiple inferred geological information corresponding to other voxels. Furthermore, a confidence quantification mechanism is proposed to transform voxel reliability from "fuzzy judgment" to "precise numbers." Construction teams can intuitively identify high / low reliability areas, avoid blindly relying on inaccurate geological models, and significantly reduce decision-making risks.

[0031] If multiple inferred geological information corresponding to other voxels is unreliable, it needs to be updated. Based on the confidence level of the target information, determine whether to update the multiple inferred geological information corresponding to other voxels. If the confidence level of the target information is greater than or equal to a first predetermined confidence level, it indicates that the accurate inferred geological information corresponding to other voxels can be determined based on the multiple inferred geological information, and there is no need to update the multiple inferred geological information corresponding to other voxels. If the confidence level of the target information is less than the first predetermined confidence level, it indicates that the accurate inferred geological information corresponding to other voxels cannot be determined based on the multiple inferred geological information corresponding to other voxels, and the multiple inferred geological information corresponding to other voxels needs to be updated.

[0032] By determining the confidence level of target information corresponding to other voxels, the reliability of multiple inferred geological information corresponding to other voxels can be accurately quantified, solving the core problem of the inability to quantify the reliability of inferred geological information in existing geological modeling techniques. Furthermore, it eliminates the need for human judgment on the accuracy of inferred geological information, avoiding the tediousness and uncertainty caused by manual intervention, and effectively improving the efficiency and accuracy of geological modeling.

[0033] Step 104: In response to determining that at least one other voxel corresponds to multiple speculative geological information, based on the multiple speculative geological information corresponding to the at least one other voxel and the target information confidence level, and the multiple sets of geological information, update the multiple speculative geological information corresponding to the at least one other voxel to its corresponding updated geological information.

[0034] In this step, given that multiple inferred geological information corresponding to at least one other voxel are determined to be updated, the multiple inferred geological information corresponding to at least one other voxel are updated to their corresponding updated geological information based on the confidence level of the target information and multiple sets of geological information. The confidence level of the target information corresponding to the other voxel determines which technique to use to update the multiple inferred geological information corresponding to the other voxel. Multiple sets of geological information provide real data for the update, and the multiple inferred geological information corresponding to the other voxel provides reference data for the update. By utilizing appropriate technical means and data from multiple sources, the accuracy of determining the multiple inferred geological information corresponding to at least one other voxel is ensured.

[0035] Step 105: Based on all actual geological information, all inferred geological information, and all updated geological information, model the predetermined geological space.

[0036] In this step, all actual geological information refers to data obtained from actual drilling, all inferred geological data refers to data with high information confidence after verification, and all updated geological data refers to data obtained by updating inferred geological update data with low information confidence. All three types of geological information are reliable. Therefore, modeling the predetermined geological space based on all actual geological information, all inferred geological information, and all updated geological information ensures the accuracy of the modeling of the predetermined geological space. It should be noted that the reliability of multiple inferred geological information corresponding to each other voxel is determined by comparing its corresponding target information confidence with a first predetermined confidence level. If the target information confidence is not 100% reliable, before modeling the predetermined geological space based on all actual geological information, all inferred geological information, and all updated geological information, it is necessary to discard multiple inferred geological information corresponding to other voxels based on the target information confidence level, retaining only the high-reliability inferred geological information corresponding to other voxels and discarding the low-reliability inferred geological information corresponding to other voxels.

[0037] The most important technical problem that this invention aims to solve is: how to achieve cross-validation through multiple sets of geological information. Voxel confidence statistical quantification Level 3 optimization A closed-loop technical solution of "convergent iterative correction" is adopted to ensure the accuracy of geological modeling, construct a high-confidence geological model, and achieve quantifiable, verifiable, and optimizable reliability of voxel attributes, providing accurate and reliable geological support for tunnel construction. Specific objectives include: ① Constructing a preliminary model set through multiple sets of geological information to achieve cross-validation of voxel attributes, converting reliability into a quantifiable confidence level of 0-1; ② Establishing a standardized low-reliability voxel optimization process, combined with constraints from surrounding high-reliability voxels and consistency checks, to avoid human intervention; ③ Introducing an iterative convergence mechanism to ensure the model reaches the optimal reliability level, ultimately outputting a high-confidence geological model consisting of "attribute value + confidence level + reliability level," providing quantifiable and reliable geological decision-making basis for tunnel construction.

[0038] The above scheme involves multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels, resulting in multiple sets of geological information corresponding to the predetermined geological space. Each set of geological information includes actual geological information corresponding to some voxels, avoiding the uniformity of the geological information set and laying a data foundation for subsequent cross-validation of drilling data. Based on the entire geological information set, multiple inferred geological information for each other voxel (excluding all partial voxels) is determined, effectively reducing the errors and biases that may be caused by a single information source. For each other voxel, based on the multiple inferred geological information corresponding to the other voxel, the confidence level of the target information corresponding to the other voxel is determined, and based on the confidence level of the target information, it is determined whether to update the multiple inferred geological information corresponding to the other voxel. By determining the confidence level of the target information corresponding to the other voxel, the reliability of the multiple inferred geological information corresponding to the other voxel is accurately quantified, solving the core problem of the inability to quantify the reliability of inferred geological information in existing geological modeling techniques. Moreover, it eliminates the need for manual judgment of the accuracy of inferred geological information, avoiding the tediousness and uncertainty caused by manual intervention, and effectively improving the efficiency and accuracy of geological modeling. In response to determining and updating multiple inferred geological information corresponding to at least one other voxel, based on the multiple inferred geological information corresponding to the at least one other voxel and the target information confidence level, as well as the multiple sets of geological information, the multiple inferred geological information corresponding to the at least one other voxel is updated to its corresponding updated geological information. By utilizing appropriate technical means and data from multiple sources, the accuracy of determining the multiple inferred geological information corresponding to at least one other voxel is ensured. Based on all actual geological information, all inferred geological information, and all updated geological information, the predetermined geological space is modeled, ensuring the accuracy of modeling the predetermined geological space.

[0039] In some embodiments, the updated geological information is first updated geological information or second updated geological information; the step of updating the multiple speculative geological information corresponding to the at least one other voxel to its corresponding updated geological information based on the confidence levels of multiple speculative geological information and target information corresponding to the at least one other voxel, and the multiple sets of geological information, includes: for each other voxel in the at least one other voxel, in response to determining that the confidence level of the target information corresponding to the other voxel is less than a first predetermined confidence level and greater than or equal to a second predetermined confidence level, determining the other voxel as a first other voxel, or, in response to determining that the confidence level of the target information is less than the second predetermined confidence level, determining the other voxel as a second other voxel; taking each actual geological information and each speculative geological information in the multiple sets of geological information as target geological information; for For each first other voxel, acquire at least one target geological information within a sphere centered on the first other voxel and with a first predetermined distance as its radius; based on the at least one target geological information corresponding to the first other voxel, determine the first updated geological information corresponding to the first other voxel using a first predetermined algorithm, and replace the multiple speculative geological information corresponding to the first other voxel with the first updated geological information; for each second other voxel, acquire at least one target geological information within a sphere centered on the second other voxel and with a second predetermined distance as its radius; based on the at least one target geological information and multiple speculative geological information corresponding to each second other voxel, perform multiple rounds of information update operations to obtain the second updated geological information corresponding to each second other voxel, and replace the multiple speculative geological information corresponding to each second other voxel with its corresponding second updated geological information.

[0040] In this embodiment, for each of the at least one other voxels, if the confidence level of the target information corresponding to the other voxel is less than a first predetermined confidence level but greater than or equal to a second predetermined confidence level, it indicates that the multiple speculative geological information corresponding to that other voxel belongs to multiple speculative geological information of moderate reliability, and the other voxel is identified as the first other voxel. Alternatively, if the confidence level of the target information is less than the second predetermined confidence level, it indicates that the multiple geological speculative information corresponding to that other voxel belongs to multiple speculative geological information of low reliability, and the other voxel is identified as the second other voxel. By distinguishing other voxels, the purpose of distinguishing the multiple speculative geological information corresponding to other voxels according to their reliability is achieved. The predetermined confidence level is the basis for deciding whether to update at least one other voxel. The confidence level of the target information is compared with two preset thresholds of the first and second predetermined confidence levels. According to the comparison result, the other voxels are divided into three categories: high reliability, moderate reliability, and low reliability, and processed differently. The first and second predetermined confidence levels are determined based on historical experience. For example, the first predetermined confidence level is 0.9 and the second predetermined confidence level is 0.6.

[0041] Each actual geological information and each inferred geological information from multiple sets of geological information are used as target geological information. For each first other voxel, at least one target geological information is acquired within a sphere centered on the first other voxel and with a first predetermined distance as its radius. By acquiring at least one target geological information surrounding the space where the first other voxel is located, updated inferred geological information corresponding to the first other voxel is inferred. The sphere represents a three-dimensional geological space. It should be noted that before determining the first updated geological information corresponding to the first other voxel using a first predetermined algorithm based on at least one target geological information corresponding to the first other voxel, the inferred geological information in the at least one target geological information corresponding to the first other voxel is filtered. In response to the determination that the confidence level of the target information in the at least one target geological information corresponding to the first other voxel is less than the first predetermined confidence level, the inferred geological information is removed from the at least one target geological information, ensuring the reliability of the remaining at least one target geological information. Since the multiple inferred geological information data corresponding to the first other voxel are of moderate reliability, although the multiple inferred geological information data corresponding to the first other voxel do not form an absolute consensus, the risk of deviation is controllable. Therefore, based on at least one target geological information corresponding to the first other voxel, a first predetermined algorithm is used to determine the first updated geological information corresponding to the first other voxel, and multiple inferred geological information corresponding to the first other voxel are replaced with the first updated geological information. For example, the first predetermined algorithm is a local interpolation or weighted fusion algorithm. By focusing on relatively reliable target geological information around the first other voxel and comprehensively analyzing this information through a specific algorithm, errors caused by data dispersion or local anomalies can be effectively reduced, making the updated geological information closer to the actual situation and ensuring the accuracy of the determined first updated geological information.

[0042] Multiple inferred geological information corresponding to the first other voxel is accurately updated through "local modeling": taking the first other voxel as the center, the local modeling range is defined (usually covering the area where the voxel itself and the surrounding high-reliability voxels are located), and the drilling data and high-confidence voxel attribute information within this range are integrated. Local interpolation or weighted fusion algorithms are used to redetermine and update the inferred geological information.

[0043] For each second other voxel, at least one target geological information is acquired within a sphere centered on the second other voxel and with a second predetermined distance as its radius. By acquiring at least one target geological information surrounding the second other voxel, updated inferred geological information corresponding to that second other voxel is inferred. Since the multiple inferred geological information data corresponding to the second other voxel are of low reliability, multiple rounds of information update operations are performed based on the at least one target geological information and multiple inferred geological information corresponding to each second other voxel to obtain the second updated geological information corresponding to each second other voxel. By fully utilizing surrounding known information and iterative inference results, the continuity and reliability of the obtained second updated geological information corresponding to each second other voxel are ensured. The core logic is to first find "reliable neighbor references" for the multiple inferred geological information corresponding to the second other voxel, and then perform multiple rounds of information update operations to test data among the multiple inferred geological information corresponding to each second other voxel until the geological attributes both match the references and conform to geological laws. Then, the multiple inferred geological information corresponding to each second other voxel is replaced with its corresponding second updated geological information.

[0044] It should be noted that before performing multiple rounds of information update operations based on at least one target geological information and multiple inferred geological information corresponding to each second other voxel, the inferred geological information in the at least one target geological information corresponding to the second other voxel is screened. In response to the determination that the target information confidence of the inferred geological information in the at least one target geological information corresponding to the second other voxel is less than a first predetermined confidence level, the inferred geological information is removed from the at least one target geological information, thus ensuring the reliability of at least one target geological information.

[0045] In some embodiments, the step of performing multiple rounds of information update operations based on at least one target geological information and multiple inferred geological information corresponding to each second other voxel to obtain second updated geological information corresponding to each second other voxel includes: each round of information update operations is performed as follows: randomly selecting one inferred geological information from the multiple inferred geological information corresponding to each second other voxel to obtain a random combination containing multiple inferred geological information corresponding to the second other voxel; determining an unreasonable score corresponding to the random combination based on the random combination and at least one target geological information corresponding to each second other voxel; updating the unreasonable score to the predetermined optimal benchmark value in the next round of information update operations in response to determining that the unreasonable score is less than the predetermined optimal benchmark value in the current round of information update operations; or, in response to determining that the unreasonable score is greater than or equal to the predetermined optimal benchmark value in the current round of information update operations, using the predetermined optimal benchmark value in the current round of information update operations as the next round of information update operations. The process involves updating a predetermined optimal baseline value in the update operation; in response to determining that the number of completed information update operations is less than or equal to a first predetermined number, executing the next round of information update operations; in response to determining that the number of completed information update operations is greater than the first predetermined number, detecting whether there are any target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal baseline value, wherein the multiple rounds of target information update operations include the current round of information update operations and a predetermined number of consecutive historical information update operations adjacent to it; in response to determining that there are target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal baseline value, executing the next round of information update operations; in response to determining that there are no target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal baseline value, taking the inferred geological information corresponding to each second other voxel contained in the random combination of the current round of information update operations as the second updated geological information corresponding to each second other voxel, and exiting at least one round of information update operations.

[0046] In this embodiment, each inferred geological information corresponding to each second other voxel is determined through each set of geological information. Currently, there is no explicit second updated geological information for the second other voxel; the second updated geological information corresponding to the second other voxel may be one of its multiple inferred geological information. To ensure the accuracy of the second updated geological information corresponding to each second other voxel, the second updated geological information corresponding to each second other voxel is determined at the overall level. Therefore, each round of information update operation is performed as follows: one inferred geological information is randomly selected from the multiple inferred geological information corresponding to each second other voxel, resulting in a random combination containing multiple inferred geological information corresponding to the second other voxel. Since multiple rounds of geological information filtering operations are performed, multiple sets of geological information are obtained. Based on each set of geological information, the inferred geological information corresponding to the second other voxel can be inferred; therefore, each second other voxel corresponds to multiple inferred geological information. For each second other voxel, one inferred geological information is randomly selected from the multiple inferred geological information corresponding to that second other voxel. All the inferred geological information corresponding to the second other voxels forms a random combination. The accuracy of multiple inferred geological information in the random combination is then verified. Based on the random combination and at least one target geological information corresponding to each second other voxel, the unreasonable score corresponding to the random combination is determined. The higher the unreasonable score, the lower the reliability of the random combination.

[0047] The unreasonable score is compared with the predetermined optimal benchmark value in the current round of information update. If the unreasonable score is lower than the predetermined optimal benchmark value, the unreasonable score is updated to the predetermined optimal benchmark value for the next round of information update. When the unreasonable score corresponding to the random combination in this round is lower than the predetermined optimal benchmark value for the current round, it indicates that a more reasonable geological information combination has been discovered than the optimal result at the current stage. In this case, the predetermined optimal benchmark value for the next round is updated to this lower unreasonable score, so that the next round of screening uses the new optimal benchmark as the judgment standard. If the unreasonable score is greater than or equal to the predetermined optimal benchmark value, the predetermined optimal benchmark value in the current round of information update is used as the predetermined optimal benchmark value for the next round of information update. When the unreasonable score corresponding to the random combination in this round is greater than or equal to the predetermined optimal benchmark value for the current round, it indicates that no better geological information combination has been discovered in this round. The current predetermined optimal benchmark value remains the optimal judgment benchmark at this stage, so this benchmark value is maintained unchanged, and the next round will still use it as the reasonableness judgment standard. It should be noted that the predetermined optimal benchmark value in the first round of information update is preset in advance, for example, the predetermined optimal benchmark value is 4.2. For example, if the unreasonable score in the first round of information update operation is 2, since 2 is less than 4.2, then 2 will be used as the predetermined optimal benchmark value in the second round of information update operation.

[0048] The two-layer progressive iterative termination determination operation, based on the number of iterative operations and the predetermined optimal baseline value update state, is a scientific determination mechanism designed for the random iterative update process of geological information. Its core principle follows the design logic of "first ensuring the sufficiency of basic iterations, then verifying the convergence stability of the optimal solution, and terminating promptly after convergence stability." This avoids missing optimal solutions due to insufficient iterations and prevents the waste of computational resources caused by meaningless and ineffective iterations. Furthermore, it achieves objective determination of iteration termination through quantitative indicators. Since the inferred geological information of the second other voxels has multiple random combinations, if the number of iterations is insufficient, the sampling cannot cover multi-dimensional geological information combinations. This can easily lead to the accidental sampling of a locally superior combination, which may be mistaken for the global optimal solution, resulting in a significant deviation between the final output geological information and the actual geological characteristics. Therefore, if the number of multiple rounds of target information update operations is less than or equal to the first predetermined number, the next round of information update operations must be executed. These multiple rounds of target information update operations include the current round of information update operations and a predetermined number of consecutive historical information update operations adjacent to it.

[0049] When the number of target information update operations exceeds the first predetermined number, the predetermined optimal benchmark value serves as the optimal criterion for judging the rationality of geological information combinations in each iteration. Its update status directly represents the "possibility of discovering a better geological information combination". If there are cases where the predetermined optimal benchmark value has not been updated in multiple rounds of target information update operations, it indicates that there is still a possibility of discovering a better geological information combination, the optimal solution has not yet stabilized, and iteration needs to continue to fully explore it. If there are no cases where the predetermined optimal benchmark value has not been updated in multiple rounds of target information update operations, that is, all target rounds have completed the benchmark value update, it indicates that the state of finding a better solution in each round has terminated under the premise of sufficient basic iteration rounds. The geological information combination corresponding to the current predetermined optimal benchmark value is the global optimal solution that can be explored at this stage, and the optimal solution tends to stabilize with no further optimization space. The inferred geological information corresponding to each second other voxel contained in the random combination obtained in the current round of information update operations is taken as the second updated geological information corresponding to each second other voxel, and the iteration is exited. Under the premise of ensuring the accuracy of the final result, invalid iteration operations are minimized, the overall computational efficiency of the geological information update process is significantly improved, and the consumption of computing resources is reduced.

[0050] By employing multiple rounds of information updates, randomly selecting combinations from various inferred geological information in each round, a comprehensive exploration of all possible combinations of geological information can be achieved. Geological spaces are complex and varied, and different combinations of geological information may correspond to different actual geological conditions. This random selection method avoids overlooking some potentially reasonable combinations, thus increasing the likelihood of finding the information combination that best approximates the actual geological situation, thereby improving the accuracy of the second updated geological information. This method avoids a comprehensive and detailed sorting and comparison of all random combinations, significantly reducing the computational load and the number of comparisons, thereby quickly selecting relatively high-quality random combinations and improving the efficiency of information screening.

[0051] The process objective is to: lock in reference neighbors (i.e., at least one target geological information corresponding to each second other voxel). Trial selection of attributes (i.e., a random combination of inferred geological information corresponding to multiple second other voxels). Quantify the rationality of the test (i.e., calculate the unreasonable scores corresponding to the random combination). Iterative optimization allows these second, unconsensual voxels to possess reasonable properties, which are then incorporated into the final geological model.

[0052] In another embodiment provided in this application, step 1: Classification candidate set statistics + filtering (locking out the second other voxels to be optimized + finding reliable neighbors corresponding to the second other voxels). This step is the "reference anchoring" stage of optimization, clarifying "which second other voxels need to be optimized" and "which neighbors can be referenced". The set of low-reliability second other voxels to be optimized includes: The reference neighbor set of the second other voxels to be optimized: ,in, The confidence level of the target information corresponding to the kth other voxel. Let j be the j-th voxel, which does not belong to the second set of other voxels. For the kth second other voxel, This is the second predetermined distance. For example, suppose the predetermined geological space has 10 voxels: the confidence level of the target information corresponding to voxel A is 0.5 (<0.6), and the confidence level of the target information corresponding to voxel B is 0.4 (<0.6). These two voxels belong to The coordinates of voxel A are (5, 5, 5), and r is set to 2 (only look for neighboring voxels within 2 squares). The neighboring voxels with a distance ≤ 2 are voxels C (coordinates (6, 5, 5), target information confidence 0.7, belonging to the reference set); voxel C is included. (Reliable neighbor cluster of voxel A). First, pick out the "voxels without consensus", and then find the "relatively reliable neighbors" for each cell to avoid incorrect determination of subsequent optimization attributes.

[0053] Step 2: Random generation of candidate sets (trial selection of attributes corresponding to voxels within the historical modeling range) + quantification of reasonableness. This step is the "trial selection of attributes" stage—the attribute range of the candidate set is clearly defined as all inferred geological information that has actually appeared in multiple sets of geological information in the previous stages for the low-reliability voxel. Random generation is performed based on historically observed inferred geological information, fundamentally limiting the reasonable boundaries of attributes and ensuring that the trial selection results conform to the probability of the voxel's geological attributes. Keeping the attribute range of the candidate set unchanged, new attribute combinations are randomly generated again; the unreasonableness score is calculated. The purpose of this process is to replace "gut feeling" with "quantified scores," ensuring that the new attributes neither deviate from the original reliable information nor contradict the spatial continuity of subsurface geology (consistency of neighboring attributes).

[0054] Step 3: Iteration Condition Judgment. The essence of the "iteration condition" is "convergence judgment." The core rule is: with a period of "q consecutive random combinations generated", calculate the unreasonable score D(c) after each generation. If the unreasonable score D(c) no longer decreases after these q generation cycles, it is determined that the condition is "satisfied," and the iteration terminates; if it continues to decrease, the condition is "not satisfied," and the process returns to regeneration. D(c) is the set of all low-reliability voxels (the set of low-reliability voxels Ω). p The larger the total unreasonable score of the summation, the worse the matching degree between the attributes of the entire low-reliability voxel group and the surrounding reliable voxels and the original confidence. The judgment node is "whether the iteration condition is met?", which points to "the D(c) after this attribute generation no longer decreases compared with the historical best value", and the execution steps are as follows: (1): the first round of random attribute generation and calculation of the baseline D(c): for Ω p For all low-reliability voxels, a set of attributes is randomly generated based on their respective surrounding candidate sets (voxel attribute range with confidence ≥ 0.6); the first round of overall consistency function value is calculated and recorded as the initial value, which is used as the predetermined optimal baseline value for iteration. (2): Attribute generation + comparison of D(c) changes: Keep the attribute range of the candidate set unchanged and randomly generate a new attribute combination again; calculate D(c) this time and compare it with the current predetermined optimal baseline value: if D(c) this time < predetermined optimal baseline value: it means that the attribute combination is more reasonable, update the predetermined optimal baseline value to D(c) this time, and return to the "randomly generate attributes" step to continue iterating; if D(c) this time ≥ predetermined optimal baseline value: it means that the attribute combination is not optimized, do not update the predetermined optimal baseline value for the time being, and continue to return to generation. (3): Triggering termination conditions: as the iteration progresses, due to the limited attribute combinations of the candidate set, the situation of "D(c) always ≥ the current predetermined optimal baseline value after multiple attribute generation" will eventually occur. At this time, it is determined that "the iteration condition is met" and the iteration is terminated. (4): Update the final model: determine the final attribute: select the set of attributes corresponding to the predetermined optimal baseline value as Ω. pThe final geological properties of all low-reliability voxels are entered into the model and labeled: the final properties and confidence levels are entered into the corresponding spatial locations of the final geological model and uniformly labeled as "low-reliability optimization level".

[0055] For example, suppose Ω p Containing two low-reliability voxels, the candidate set attributes are only "sandstone, mudstone", and the predetermined weight λ corresponding to the consistency score is 1.2: First generation: k1=mudstone, k2=mudstone → initial (optimal baseline); Second generation: k1=sandstone, k2=mudstone → D(c)=2.0 (D(c)<4.2, the predetermined optimal baseline value is 2.0, continue iterating); Third generation: k1=mudstone, k2=sandstone → D(c)=3.1 (D(c)≥2 0.0, no update, continue iteration); 4th generation: k1=sandstone, k2=sandstone → D(c)=2.0 (D(c)≥2.0, no update); 5th generation: k1=sandstone, k2=mudstone → D(c)=2.0 (D(c)≥2.0); At this time, the D(c) generated in multiple iterations is not lower than the predetermined optimal benchmark value of 2.0, which satisfies the iteration condition, terminates the iteration, and selects the attribute combination of "k1=sandstone, k2=mudstone" to update the model. Convergent iterative optimization method for low-reliability voxels: with "candidate set constraint → probabilistic random generation → consistency check → the predetermined optimal benchmark value D(c) does not decrease after q consecutive times" as a closed loop, combined with multi-source data weighted fusion, to ensure that the attributes of low-reliability voxels conform to geological laws and spatial continuity.

[0056] In some embodiments, performing multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space includes: obtaining a drilling combination containing multiple drilling information; each round of geological information filtering operation is performed as follows: determining the drilling information of each predetermined number in the drilling combination as a set of geological information; in response to determining that the number of all geological information sets is less than or equal to a second predetermined number, updating some predetermined numbers in all predetermined numbers in the current round of geological information filtering operation, and performing the next round of geological information filtering operation; in response to determining that the number of all geological information sets is greater than the second predetermined number, exiting at least one round of geological information filtering operation.

[0057] In this embodiment, a drilling combination containing multiple drilling information is obtained, and iterative sampling and interpolation loops are performed on the drilling combination using multiple iterative information. This process is the basic model generation step of the entire multi-source information fusion geological modeling method, through "drilling combination sampling". Spatial interpolation modeling The closed-loop process of "model set accumulation" yields multiple sets of geological information, providing a sample basis for subsequent information confidence analysis. Each round of geological information screening is performed as follows: Drilling information for each predetermined sequence number in the drilling combination is determined as a geological information set. For example, if the drilling combination is {Drilling Information 1, Drilling Information 2, Drilling Information 3}, and the drilling information with predetermined sequences 1 and 2 in the drilling combination is determined as a geological information set, then the geological information set is {Drilling Information 1, Drilling Information 2}. If the total number of geological information sets is less than or equal to a second predetermined number, it indicates that the number of geological information sets is insufficient. In this case, some predetermined sequences from all predetermined sequences in the current round of geological information screening are updated, and the next round of screening is executed. Geological information filtering operation. For example, the predetermined sequence number 2 in the drilling combination is updated to 3, and the next round of geological information filtering operation is performed. The next round of geological information set will be {drilling information 1, drilling information 3}. If the total number of geological information sets is greater than a second predetermined number, it indicates that the number of geological information sets is sufficient, and at least one round of geological information filtering operation is exited. The second predetermined number is determined based on historical experience. By filtering the drilling information in the drilling combination, multiple sets of geological information sets can be cross-validated. A combination sampling strategy for each missing drilling information is adopted to construct multiple sets of geological information sets, achieving cross-validation of borehole data, avoiding model failure caused by single borehole location bias, and providing a statistical basis for confidence calculation.

[0058] For example, five actual boreholes (denoted as boreholes 1-5, with corresponding data as drilling information 1-5, all of which are "spatial truth information points" possessing precise geological attributes) are deployed in the predetermined geological space. The sampling rule is clearly defined as "selecting four from the five drilling information points each time to form one set of geological information." Based on the combination logic, the five drilling information points can generate a total of five sets of borehole combinations, and each set is missing information from only one drilling information point. The specific combinations are as follows: ① Geological Information Set 1: Drilling Information 1 + Drilling Information 2 + Drilling Information 3 + Drilling Information 4 (Drilling Information 5 is missing); ② Geological information set 2: Drilling information 1 + Drilling information 2 + Drilling information 3 + Drilling information 5 (Drilling information 4 is missing); ③ Geological information set 3: Drilling information 1 + Drilling information 2 + Drilling information 4 + Drilling information 5 (Drilling information 3 is missing); ④ Geological information set 4: Drilling information 1 + Drilling information 3 + Drilling information 4 + Drilling information 5 (Drilling information 2 is missing); ⑤ Geological Information Set 5: Drilling Information 2 + Drilling Information 3 + Drilling Information 4 + Drilling Information 5 (Drilling Information 1 is missing).

[0059] This sampling method ensures that each drilling information can be verified in different geological information sets, while covering all scenarios of "missing single borehole information".

[0060] Loop Trigger: If the number of geological information sets generated so far has not reached the second predetermined number (i.e., no of the above 5 sets of geological information have been generated), a loop is triggered, returning to the "Geological Information Filtering Operation" step, selecting the next set of borehole combinations that have not undergone interpolation modeling, and repeating the "Sampling" process. interpolation Modeling The process of "accumulation".

[0061] Loop Termination: When all 5 borehole combinations have completed interpolation modeling and the geological information set has contained all 5 preliminary models, the loop terminates and the process proceeds to the next stage (voxel consistency statistics and confidence calculation).

[0062] The core value of this iterative sampling-interpolation loop lies in: ① making full use of limited drilling data: by combining designs with "missing single boreholes", the influence weight of each borehole on the geological modeling of the surrounding area is verified, avoiding modeling blind spots caused by missing information from a single borehole location; ② reducing the bias risk of a single geological information set: multiple sets of geological information can reflect the differences in geological extrapolation under different borehole combinations, and subsequent consistency comparison can filter out abnormal extrapolation results, improving the reliability of the final model; ③ laying the foundation for confidence calculation: the existence of multiple sets of geological information makes "voxel attribute consistency statistics" possible, which is the core prerequisite for quantifying the "reliability" of each voxel.

[0063] In some embodiments, determining multiple inferred geological information for each other voxel (excluding all partial voxels) among the plurality of voxels based on the entire geological information set includes: for each geological information set, determining the inferred geological information for other voxels corresponding to the geological information set using a second predetermined algorithm, so as to obtain multiple inferred geological information for each other voxel.

[0064] In this embodiment, for each geological information set, based on the geological information set, a second predetermined algorithm is used to determine the inferred geological information of other voxels corresponding to the geological information set. For example, the second predetermined algorithm is a spatial interpolation algorithm. Each geological information set includes actual geological information corresponding to some voxels, and the precise geological attributes contained therein are known samples. The spatial interpolation algorithm is used to assign attribute values ​​across the entire range to the predetermined geological space that has been spatially voxelized. Simply put, the geological information set is equivalent to clearly defining the underground geological coordinates. The spatial interpolation algorithm will "reasonably deduce" the geological attributes of all other voxel units based on the clear spatial positional relationships and the correlation of geological attributes, transforming discrete drilling information into a continuous and complete three-dimensional preliminary geological model, ensuring the accuracy of the inferred geological information corresponding to each geological information set. Since the actual geological information corresponding to each geological information set is partially different, the inferred geological information corresponding to the same other voxels in each geological information set may be different.

[0065] In some embodiments, determining the target information confidence level corresponding to the other voxels based on multiple inferred geological information corresponding to the other voxels includes: extracting attributes from the multiple inferred geological information corresponding to the other voxels to obtain multiple attributes corresponding to the other voxels; calculating the initial information confidence level corresponding to each attribute based on the multiple attributes; and taking the maximum initial information confidence level among all initial information confidence levels as the target information confidence level corresponding to the other voxels.

[0066] In this embodiment, based on multiple sets of geological information, the "consistency degree" of the geological attributes of each other voxel is quantified to calculate its "information confidence," providing a quantitative basis for subsequent model optimization. The core logic is: the more consistent the geological attribute corresponding to another voxel is across multiple sets of geological information, the higher its information confidence (the more reliable); the less consistent it is, the lower its information confidence (the more it needs optimization). Voxel consistency statistics are a prerequisite for information confidence calculation. Essentially, it involves "statistically determining whether the geological attributes of the same voxel are the same and how many times they are the same among multiple inferred geological information corresponding to other voxels." The specific statistical logic is as follows: for each other voxel within a predetermined geological space (traversing them one by one without omission), its corresponding geological attributes in multiple sets of geological information are extracted. Attribute extraction is performed on multiple inferred geological information corresponding to other voxels to obtain multiple attributes corresponding to other voxels. The consistency count of a voxel's attribute is recorded as "number of times" (i.e., how many sets of geological information assign the same geological attribute to another voxel; for example, if 4 out of 5 sets of geological information identify the voxel as "sandstone," the consistency count is 4). This is recorded in the form of "voxel-attribute-consistency count," forming a "voxel consistency statistics matrix" (e.g., voxel 1: sandstone, consistency count 4; voxel 2: mudstone, consistency count 2, etc.). For example: Suppose there are 5 sets of geological information (M1~M5), and voxel A's attribute is sandstone in M1~M4 and mudstone in M5. Then the consistency statistics for voxel A are: consistency count for attribute "sandstone" = 4, consistency count for attribute "mudstone" = 1.

[0067] Based on multiple attributes, the initial information confidence level for each attribute is calculated. The confidence level calculation converts the "consistency statistics result" into a quantitative value (confidence level) between 0 and 1. The closer the value is to 1, the more reliable the geological attributes of that voxel; the closer it is to 0, the less reliable it is. The initial information confidence level for each attribute is determined using the following formula: Formula 1, where, This refers to the confidence level of the kth other voxel corresponding to "attribute m" (e.g., the confidence level of other voxel A corresponding to "sandstone"), and n refers to the number of geological information sets (n=5 in the example). This refers to summing the results of "all n sets of geological information", δ It is a "judgment switch function" (δ=1 when the condition in parentheses is true, δ=0 when it is false, i.e., "compliance count 1, non-compliance count 0"). For the first k Individual phenotypic modeling attributes Stratigraphic properties mFor example, calculating "other voxels A": n=5, target attribute m=sandstone, determine whether the attribute of voxel A in each set of geological information is equal to "sandstone". M1~M4 all satisfy the condition (δ = 1 for each), M5 does not satisfy the condition (δ = 0), and calculate the initial information confidence level. Similarly, the initial information confidence level of voxel A corresponding to "mudstone" is 0.2, and the final candidate attribute of voxel A is "sandstone", with a target information confidence level of 0.8 (representing that the "reliability score" of sandstone is 80%).

[0068] The highest initial information confidence score among all initial information confidence scores is used as the target information confidence score for other voxels. The calculation rule is supplemented as follows: for each voxel, confidence scores are calculated for all possible geological attributes (such as sandstone, mudstone, limestone, etc.), and the attribute with the highest confidence score is selected as the "candidate attribute" for that voxel. Its corresponding initial information confidence score is the "core confidence score" for that voxel. By calculating the target information confidence scores for other voxels, the reliability of multiple inferred geological information corresponding to other voxels is quantified. The fuzzy judgment of "whether multiple inferred geological information corresponding to the same other voxels in multiple sets of geological information are consistent" is transformed into a specific value of 0-1, providing a clear standard for subsequent model optimization (e.g., target information confidence scores > 0.9 are directly retained, < 0.6 require optimization). Using the highest initial information confidence score among all initial information confidence scores as the target information confidence score for other voxels ensures the accuracy of the target information confidence scores for other voxels. At this point, the inferred geological information corresponding to other voxels is the inferred geological information based on the target information confidence scores for other voxels. Simultaneously connecting with subsequent processes, the confidence result serves as the core basis for subsequent "model optimization"—high-confidence voxels directly confirm attributes, while low-confidence voxels need further correction through candidate set screening, random generation, and other methods to ensure the reliability of the final model.

[0069] In some embodiments, determining the unreasonable score corresponding to the random combination based on the random combination and at least one target geological information corresponding to each second other voxel includes: for each second other voxel, determining the deviation score of the inferred geological information corresponding to the second other voxel in the random combination; based on the inferred geological information corresponding to the second other voxel in the random combination and at least one target geological information corresponding to the second other voxel, determining the consistency score of the inferred geological information corresponding to the second other voxel in the random combination; determining the sum of the deviation score and the consistency score as the unreasonable score of the inferred geological information corresponding to the second other voxel in the random combination; and determining the sum of the unreasonable scores of all inferred geological information corresponding to the random combination as the unreasonable score of the random combination.

[0070] In this embodiment, the random combination is validated for reasonableness. An unreasonable score is used to determine whether the new attribute matches both the original reliable score and the neighboring attributes. The unreasonable score for the random combination is determined using the following formula: (the smaller the score, the more reasonable the attribute): Formula 2, where, For the unreasonable scores corresponding to random combinations, This is the sum of the deviation score and the consistency score. δ represents the predetermined weight corresponding to the consistency score. It is a "judgment switch function" (δ=1 when the condition in parentheses is true, δ=0 when it is false, i.e., "compliance count 1, non-compliance count 0"). For the first k Individual phenotypic modeling attributes For the first j Individual phenotypic modeling attributes For the second set of other voxels with low reliability to be optimized, The reference neighbor set of the second other voxel to be optimized is the set of at least one target geological information corresponding to the second other voxel.

[0071] The deviation score of the inferred geological information corresponding to the second other voxel in a random combination is determined, and the consistency score of the inferred geological information corresponding to the second other voxel in the random combination is determined based on the inferred geological information corresponding to the second other voxel in the random combination and at least one target geological information corresponding to the second other voxel. For each second other voxel, the deviation score of its corresponding inferred geological information in the random combination is first determined. This step aims to consider the degree of deviation between the inferred geological information and the expected state. Due to the complexity and uncertainty of geological space, the inferred geological information may deviate from the actual geological conditions. By quantifying this deviation, the accuracy of the inferred results can be preliminarily understood. Next, the consistency score is determined based on the inferred geological information of the second other voxel in the random combination and at least one corresponding target geological information. The target geological information can be regarded as a relatively accurate and reliable geological data reference. Calculating the consistency score can measure the degree of fit between the inferred geological information and these reliable data, reflecting the degree of consistency between the inferred results and the actual situation. The summation of the deviation score and the consistency score yields the unreasonable score for the second other voxel in the random combination. This comprehensive approach, considering both deviation and consistency, avoids the one-sidedness of a single indicator evaluation and more comprehensively reflects the degree of unreasonableness of the inferred geological information under this random combination. Finally, the unreasonable scores of all inferred geological information corresponding to the random combination are added together to obtain the unreasonable score of the entire random combination. This allows for an overall assessment of the rationality of the random combination, ensuring the accuracy of the unreasonable score corresponding to the random combination.

[0072] For example, the second other voxel A (original reliable component P) k =0.5), neighbor heap Ω k It is voxel C (attribute "sandstone"), the new attribute is selected as "sandstone", and λ=1.2: First item (deviation score): 1 - 0.5 = 0.5; Second item (consistency score): Sandstone 1.2 × 0 = 0; The sum of the unreasonable scores of all inferred geological information corresponding to the random combination is determined as the unreasonable score of the random combination. Total unreasonable score: D(c) = 0.5 + 0 = 0.5 (the score is very low, and the attribute is very reasonable).

[0073] If you select "Mudstone" as the new attribute: Second item (consistency score): Mudstone and Sandstone 1.2 × 1 = 1.2; Total unreasonable score: D(c) = 0.5 + 1.2 = 1.7 (high score, unreasonable attribute).

[0074] In another embodiment provided in this application, such as Figure 2 As shown, Figure 2This is a flowchart illustrating a geological space modeling method according to another embodiment of this application. A predetermined geological space is spatially voxelized to obtain a predetermined geological space containing multiple voxels. Multiple drilling information in a drilling combination corresponds one-to-one with actual boreholes. Drilling information corresponding to n-1 boreholes is extracted, and spatial interpolation is performed on the drilling information corresponding to the n-1 boreholes. It is determined whether to traverse all borehole combinations. If not, drilling information extraction continues. If all borehole combinations are traversed, multiple sets of geological information are obtained. Each set of geological information can infer a geological model, which includes inferred geological information corresponding to multiple other voxels. Voxel consistency statistics are performed on the multiple inferred geological information corresponding to each other voxel in the multiple sets of geological information, and the maximum confidence level (i.e., target information confidence level) is calculated. If the maximum confidence level corresponding to other voxels is greater than 0.9, the final geological space is modeled using the multiple inferred geological information corresponding to other voxels. When the maximum confidence level for other voxels is less than or equal to 0.9 and greater than or equal to 0.6, constraints are applied to multiple inferred geological information corresponding to other voxels using each actual geological information and each inferred geological information from multiple sets of geological information. A first predetermined algorithm is used for local modeling to determine the updated geological information corresponding to other voxels, and this updated geological information is used to model the final geological space. When the maximum confidence level for other voxels is less than 0.6, a classification candidate set containing multiple inferred geological information corresponding to all other voxels with a maximum confidence level less than 0.6 is statistically analyzed. This candidate set is then filtered, and random combinations are randomly generated based on it. Consistency checks are performed on these random combinations. If the iteration conditions are met, the inferred geological information corresponding to each other voxel in the random combination is used as the updated geological information corresponding to each other voxel, and this updated geological information is used to model the final geological space. If the iteration conditions are not met, a random combination generated based on the classification candidate set is returned. Finally, high-confidence geological modeling is performed on the predetermined geological space using all actual geological information, all inferred geological information, and all updated geological information.

[0075] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0076] Based on the same inventive concept, corresponding to any of the above embodiments, this application also provides a geological space modeling device.

[0077] refer to Figure 3 The geological space modeling device includes: The filtering module 10 is configured to perform multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space, wherein each set of geological information includes actual geological information corresponding to some voxels.

[0078] The first determining module 20 is configured to determine multiple inferred geological information for each of the multiple voxels other than all partial voxels, based on the entire set of geological information.

[0079] The second determining module 30 is configured to, for each other voxel, determine the confidence level of the target information corresponding to the other voxel based on the multiple speculative geological information corresponding to the other voxel, and determine whether to update the multiple speculative geological information corresponding to the other voxel based on the confidence level of the target information.

[0080] The update module 40 is configured to, in response to determining that updating multiple speculative geological information corresponding to at least one other voxel, update the multiple speculative geological information corresponding to the at least one other voxel to its corresponding updated geological information based on the multiple speculative geological information corresponding to the at least one other voxel, the target information confidence level, and the multiple sets of geological information.

[0081] Modeling module 50 is configured to model the predetermined geological space based on all actual geological information, all inferred geological information, and all updated geological information.

[0082] Using the aforementioned device, multiple rounds of geological information filtering operations are performed on a predetermined geological space containing multiple voxels, resulting in multiple sets of geological information corresponding to the predetermined geological space. Each set of geological information includes actual geological information corresponding to some voxels, avoiding the uniformity of the geological information set and laying a data foundation for subsequent cross-validation of drilling data. Based on the entire geological information set, multiple inferred geological information for each other voxel (excluding all partial voxels) is determined, effectively reducing errors and biases that may be caused by a single information source. For each other voxel, based on the multiple inferred geological information corresponding to the other voxel, the confidence level of the target information corresponding to the other voxel is determined, and based on the confidence level of the target information, it is determined whether to update the multiple inferred geological information corresponding to the other voxel. By determining the confidence level of the target information corresponding to the other voxel, the reliability of the multiple inferred geological information corresponding to the other voxel is accurately quantified, solving the core problem of the inability to quantify the reliability of inferred geological information in existing geological modeling techniques. Moreover, there is no need for human judgment on the accuracy of inferred geological information, avoiding the tediousness and uncertainty caused by manual intervention, and effectively improving the efficiency and accuracy of geological modeling. In response to determining and updating multiple inferred geological information corresponding to at least one other voxel, based on the multiple inferred geological information corresponding to the at least one other voxel and the target information confidence level, as well as the multiple sets of geological information, the multiple inferred geological information corresponding to the at least one other voxel is updated to its corresponding updated geological information. By utilizing appropriate technical means and data from multiple sources, the accuracy of determining the multiple inferred geological information corresponding to at least one other voxel is ensured. Based on all actual geological information, all inferred geological information, and all updated geological information, the predetermined geological space is modeled, ensuring the accuracy of modeling the predetermined geological space.

[0083] In some embodiments, the updated geological information is first updated geological information or second updated geological information; the update module 40 is further configured to, for each of the at least one other voxels, in response to determining that the confidence level of the target information corresponding to the other voxel is less than a first predetermined confidence level and greater than or equal to a second predetermined confidence level, determine the other voxel as a first other voxel, or, in response to determining that the confidence level of the target information is less than the second predetermined confidence level, determine the other voxel as a second other voxel; take each actual geological information and each inferred geological information in the multiple sets of geological information as target geological information; for each first other voxel, obtain at least one target geological information within a sphere with the first other voxel as the center and a first predetermined distance as the radius; based on the at least one target geological information corresponding to the first other voxel, use a first predetermined algorithm to determine the first updated geological information corresponding to the first other voxel, and replace the multiple inferred geological information corresponding to the first other voxel with the first updated geological information; For each second other voxel, acquire at least one target geological information within a sphere centered on the second other voxel and with a second predetermined distance as its radius; based on the at least one target geological information and multiple inferred geological information corresponding to each second other voxel, perform multiple rounds of information update operations to obtain the second updated geological information corresponding to each second other voxel, and replace the multiple inferred geological information corresponding to each second other voxel with its corresponding second updated geological information.

[0084] In some embodiments, the update module 40 is further configured to perform the following operations in each round of information update: randomly select one speculative geological information from multiple speculative geological information corresponding to each second other voxel to obtain a random combination containing multiple speculative geological information corresponding to the second other voxels; determine an unreasonable score corresponding to the random combination based on the random combination and at least one target geological information corresponding to each second other voxel; update the unreasonable score to the predetermined optimal benchmark value in the next round of information update operation in response to determining that the unreasonable score is less than the predetermined optimal benchmark value in the current round of information update operation; or, in response to determining that the unreasonable score is greater than or equal to the predetermined optimal benchmark value in the current round of information update operation, use the predetermined optimal benchmark value in the current round of information update operation as the predetermined optimal benchmark value in the next round of information update operation; and in response to determining that the completed information update operation has been performed... If the number of new operations is less than or equal to a first predetermined number, execute the next round of information update operations; in response to determining that the number of completed information update operations is greater than the first predetermined number, detect whether there are any target information update operations that have not updated the predetermined optimal benchmark value in the multiple rounds of target information update operations, wherein the multiple rounds of target information update operations include the current round of information update operations and a predetermined number of consecutive historical information update operations adjacent to it; in response to determining that there are target information update operations that have not updated the predetermined optimal benchmark value in the multiple rounds of target information update operations, execute the next round of information update operations; in response to determining that there are no target information update operations that have not updated the predetermined optimal benchmark value in the multiple rounds of target information update operations, take the inferred geological information corresponding to each second other voxel contained in the random combination of the current round of information update operations as the second updated geological information corresponding to each second other voxel, and exit at least one round of information update operations.

[0085] In some embodiments, the filtering module 10 is further configured to acquire a drilling combination containing multiple drilling information; each round of geological information filtering operation is performed as follows: the drilling information of each predetermined number in the drilling combination is determined as a set of geological information; in response to determining that the number of all geological information sets is less than or equal to a second predetermined number, a portion of the predetermined numbers in all predetermined numbers in the current round of geological information filtering operation is updated, and the next round of geological information filtering operation is performed; in response to determining that the number of all geological information sets is greater than the second predetermined number, at least one round of geological information filtering operation is exited.

[0086] In some embodiments, the first determining module 20 is further configured to, for each geological information set, determine the inferred geological information of other voxels corresponding to the geological information set using a second predetermined algorithm, so as to obtain multiple inferred geological information for each other voxel.

[0087] In some embodiments, the second determining module 30 is further configured to extract attributes from the multiple speculative geological information corresponding to the other voxels to obtain multiple attributes corresponding to the other voxels; calculate the initial information confidence level corresponding to each attribute based on the multiple attributes; and take the maximum initial information confidence level among all initial information confidence levels as the target information confidence level corresponding to the other voxels.

[0088] In some embodiments, the updating module 40 is further configured to, for each second other voxel, determine a deviation score of the inferred geological information corresponding to the second other voxel in the random combination; determine a consistency score of the inferred geological information corresponding to the second other voxel in the random combination based on the inferred geological information corresponding to the second other voxel in the random combination and at least one target geological information corresponding to the second other voxel; determine the sum of the deviation score and the consistency score as the unreasonable score of the inferred geological information corresponding to the second other voxel in the random combination; and determine the sum of the unreasonable scores of all inferred geological information corresponding to the random combination as the unreasonable score corresponding to the random combination.

[0089] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0090] The apparatus described above is used to implement the corresponding geological space modeling method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0091] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the geological space modeling method as described in any of the above embodiments.

[0092] Figure 4 This embodiment illustrates a more specific hardware structure of an electronic device. The device may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0093] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0094] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0095] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0096] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0097] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0098] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0099] The electronic devices described above are used to implement the corresponding geological space modeling methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0100] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to execute the geological space modeling method as described in any of the above embodiments.

[0101] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0102] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the geological space modeling method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0103] Based on the same concept, corresponding to the methods of any of the above embodiments, this application also provides a computer program product, including computer program instructions. When the computer program instructions are run on a computer, the computer causes the computer to execute the geological space modeling method as described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0104] It should be noted that the embodiments of this application can also be further described in the following ways: It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0105] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0106] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0107] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0108] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application is limited to these examples; under the concept of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in detail for the sake of brevity.

[0109] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0110] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0111] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A geological spatial modeling method, characterized in that, include: Multiple rounds of geological information filtering operations are performed on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space, wherein each set of geological information includes actual geological information corresponding to some voxels. Based on the complete set of geological information, multiple inferred geological information for each other voxel in the plurality of voxels, excluding all partial voxels, are determined; For each other voxel, based on the multiple inferred geological information corresponding to the other voxel, the confidence level of the target information corresponding to the other voxel is determined, and based on the confidence level of the target information, it is determined whether to update the multiple inferred geological information corresponding to the other voxel. In response to determining to update multiple speculative geological information corresponding to at least one other voxel, based on the multiple speculative geological information corresponding to the at least one other voxel and the target information confidence, and the multiple sets of geological information, the multiple speculative geological information corresponding to the at least one other voxel is updated to its corresponding updated geological information. Based on all actual geological information, all inferred geological information, and all updated geological information, the predetermined geological space is modeled. The updated geological information is either first updated geological information or second updated geological information; the step of updating the multiple inferred geological information corresponding to the at least one other voxel to its corresponding updated geological information based on the confidence levels of the multiple inferred geological information and target information corresponding to the at least one other voxel, and the multiple sets of geological information, includes: for each other voxel in the at least one other voxel, in response to determining that the confidence level of the target information corresponding to the other voxel is less than a first predetermined confidence level and greater than or equal to a second predetermined confidence level, determining the other voxel as a first other voxel, or, in response to determining that the confidence level of the target information is less than the second predetermined confidence level, determining the other voxel as a second other voxel; taking each actual geological information and each inferred geological information in the multiple sets of geological information as target geological information; for each first For each second other voxel, acquire at least one target geological information within a sphere centered on the first other voxel and with a radius of a first predetermined distance. Based on the at least one target geological information corresponding to the first other voxel, determine the first updated geological information corresponding to the first other voxel using a first predetermined algorithm, and replace the multiple speculative geological information corresponding to the first other voxel with the first updated geological information. For each second other voxel, acquire at least one target geological information within a sphere centered on the second other voxel and with a radius of a second predetermined distance. Based on the at least one target geological information and multiple speculative geological information corresponding to each second other voxel, perform multiple rounds of information update operations to obtain the second updated geological information corresponding to each second other voxel, and replace the multiple speculative geological information corresponding to each second other voxel with its corresponding second updated geological information. The process of performing multiple rounds of information update operations based on at least one target geological information and multiple inferred geological information corresponding to each second other voxel to obtain second updated geological information corresponding to each second other voxel includes: each round of information update operations is performed as follows: randomly selecting one inferred geological information from the multiple inferred geological information corresponding to each second other voxel to obtain a random combination containing multiple inferred geological information corresponding to the second other voxel; determining an unreasonable score corresponding to the random combination based on the random combination and at least one target geological information corresponding to each second other voxel; updating the unreasonable score to the predetermined optimal benchmark value in the next round of information update operations in response to determining that the unreasonable score is less than the predetermined optimal benchmark value in the current round of information update operations; or, in response to determining that the unreasonable score is greater than or equal to the predetermined optimal benchmark value in the current round of information update operations, using the predetermined optimal benchmark value in the current round of information update operations as the next round of information update operations. The predetermined optimal benchmark value is determined. In response to determining that the number of completed information update operations is less than or equal to a first predetermined number, the next round of information update operations is executed. In response to determining that the number of completed information update operations is greater than the first predetermined number, it is detected whether there are any target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal benchmark value, wherein the multiple rounds of target information update operations include the current round of information update operations and a predetermined number of consecutive historical information update operations adjacent to it. In response to determining that there are target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal benchmark value, the next round of information update operations is executed. In response to determining that there are no target information update operations in the multiple rounds of target information update operations that have not updated the predetermined optimal benchmark value, the inferred geological information corresponding to each second other voxel included in the random combination of the current round of information update operations is used as the second updated geological information corresponding to each second other voxel, and at least one round of information update operations is exited.

2. The method according to claim 1, characterized in that, The process of performing multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space includes: Obtain drilling combinations that contain information on multiple drilling operations; Each round of geological information screening is performed as follows: The drilling information of each predetermined number in the drilling combination is determined as a set of geological information. In response to the determination that the total number of geological information sets is less than or equal to the second predetermined number, some predetermined numbers in all predetermined numbers in the current round of geological information screening operation are updated, and the next round of geological information screening operation is executed. In response to the determination that the total number of geological information sets is greater than the second predetermined number, exit at least one round of geological information screening operation.

3. The method according to claim 1, characterized in that, The determination of multiple inferred geological information for each voxel other than all partial voxels, based on the entire geological information set, includes: For each set of geological information, based on the set of geological information, a second predetermined algorithm is used to determine the inferred geological information of other voxels corresponding to the set of geological information, so as to obtain multiple inferred geological information for each other voxel.

4. The method according to claim 1, characterized in that, The process of determining the confidence level of the target information corresponding to the other voxels based on multiple inferred geological information includes: Attribute extraction is performed on multiple inferred geological information corresponding to the other voxels to obtain multiple attributes corresponding to the other voxels; Based on the multiple attributes, calculate the initial information confidence level corresponding to each attribute; The highest initial information confidence score among all initial information confidence scores is taken as the target information confidence score for the other voxels.

5. The method according to claim 1, characterized in that, The determination of unreasonable scores corresponding to the random combination based on the random combination and at least one target geological information corresponding to each second other voxel includes: For each second other voxel, determine the deviation score of the inferred geological information corresponding to the second other voxel in the random combination; Based on the inferred geological information corresponding to the second other voxel in the random combination and at least one target geological information corresponding to the second other voxel, determine the consistency score corresponding to the inferred geological information of the second other voxel in the random combination. The sum of the deviation score and the consistency score is determined as the unreasonable score of the inferred geological information corresponding to the second other voxel in the random combination; The sum of the unreasonable scores of all speculative geological information corresponding to the random combination is determined as the unreasonable score corresponding to the random combination.

6. A geological spatial modeling device, characterized in that, include: The filtering module is configured to perform multiple rounds of geological information filtering operations on a predetermined geological space containing multiple voxels to obtain multiple sets of geological information corresponding to the predetermined geological space, wherein each set of geological information includes actual geological information corresponding to some voxels. The first determining module is configured to determine multiple inferred geological information for each of the multiple voxels other than all partial voxels, based on the entire set of geological information. The second determining module is configured to, for each other voxel, determine the confidence level of the target information corresponding to the other voxel based on multiple inferred geological information corresponding to the other voxel, and determine whether to update the multiple inferred geological information corresponding to the other voxel based on the confidence level of the target information. The update module is configured to, in response to determining that multiple speculative geological information corresponding to at least one other voxel is to be updated, update the multiple speculative geological information corresponding to the at least one other voxel to its corresponding updated geological information based on the multiple speculative geological information corresponding to the at least one other voxel and the target information confidence level, and the multiple sets of geological information. The modeling module is configured to model the predetermined geological space based on all actual geological information, all inferred geological information, and all updated geological information. The updated geological information is either first updated geological information or second updated geological information; the updating module is further configured to, for each of the at least one other voxels, in response to determining that the confidence level of the target information corresponding to the other voxel is less than a first predetermined confidence level and greater than or equal to a second predetermined confidence level, determine the other voxel as a first other voxel, or, in response to determining that the confidence level of the target information is less than the second predetermined confidence level, determine the other voxel as a second other voxel; take each actual geological information and each inferred geological information in the multiple sets of geological information as target geological information; for each first other voxel, acquire at least one voxel within a sphere centered on the first other voxel and with a first predetermined distance as its radius. Target geological information; based on at least one target geological information corresponding to the first other voxel, a first predetermined algorithm is used to determine the first updated geological information corresponding to the first other voxel, and multiple inferred geological information corresponding to the first other voxel are replaced with the first updated geological information; for each second other voxel, at least one target geological information is obtained within a sphere with the second other voxel as the center and a second predetermined distance as the radius; based on at least one target geological information and multiple inferred geological information corresponding to each second other voxel, multiple rounds of information update operations are performed to obtain the second updated geological information corresponding to each second other voxel, and multiple inferred geological information corresponding to each second other voxel are replaced with its corresponding second updated geological information; The update module is further configured to perform the following operations in each round of information update: randomly select one inferred geological information from multiple inferred geological information corresponding to each second other voxel to obtain a random combination containing multiple inferred geological information corresponding to the second other voxels; based on the random combination and at least one target geological information corresponding to each second other voxel, determine the unreasonable score corresponding to the random combination; in response to determining that the unreasonable score is less than the predetermined optimal benchmark value in the current round of information update, update the unreasonable score to the predetermined optimal benchmark value in the next round of information update; or, in response to determining that the unreasonable score is greater than or equal to the predetermined optimal benchmark value in the current round of information update, use the predetermined optimal benchmark value in the current round of information update as the predetermined optimal benchmark value in the next round of information update; in response to determining the number of completed information update operations... If the quantity is less than or equal to a first predetermined quantity, execute the next round of information update operation; in response to determining that the number of completed information update operations is greater than the first predetermined quantity, detect whether there are any target information update operations that have not updated the predetermined optimal benchmark value in the multiple rounds of target information update operations, wherein the multiple rounds of target information update operations include the current round of information update operation and a predetermined number of consecutive historical information update operations adjacent to it; in response to determining that there are target information update operations that have not updated the predetermined optimal benchmark value in the multiple rounds of target information update operations, execute the next round of information update operation; in response to determining that there are no target information update operations that have not updated the predetermined optimal benchmark value in the multiple rounds of target information update operations, take the inferred geological information corresponding to each second other voxel contained in the random combination of the current round of information update operations as the second updated geological information corresponding to each second other voxel, and exit at least one round of information update operation.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method described in any one of claims 1 to 5.

Images