Method, device and equipment for constructing three-dimensional geological model fused with multi-element data and medium

Abstract

This invention discloses a method, apparatus, equipment, and medium for constructing a three-dimensional geological model integrating multivariate data, applicable to the field of geological exploration. The method includes acquiring two-dimensional profile detection data of a target geological area and expanding it into three-dimensional training data; extracting spatial geological constraints; constructing a multivariate clustering pattern library based on the three-dimensional training data, velocity model, and density model; constructing a candidate pattern library based on a pre-defined simulation path and spatial geological constraints; calculating a probability matrix based on the candidate pattern library; selecting patterns based on the probability matrix and cross-entropy to obtain an initial three-dimensional geological model; and performing multi-scale iterative optimization on the initial three-dimensional geological model to obtain an optimized three-dimensional geological model. This invention effectively improves the accuracy and reliability of three-dimensional geological models, better handles data sparsity and geological complexity issues, and reduces the uncertainty of simulation results.

Description

Technical Field

[0001] This invention relates to the field of geological exploration technology, and in particular to a method, apparatus, equipment and medium for constructing a three-dimensional geological model that integrates multi-source data. Background Technology

[0002] With the continuous development of geological exploration technology, three-dimensional geological modeling is increasingly widely used in resource development and engineering surveys. However, existing technologies still face many challenges in constructing complex geological structures.

[0003] While traditional multipoint statistical (MPS) methods can simulate geological structures, they suffer from significant uncertainty and error when data is sparse and geological constraints are insufficient. Furthermore, most existing MPS methods extract pattern features from single training images, failing to capture the global characteristics of complex geological structures, leading to insufficient reliability and accuracy in simulation results. Simultaneously, when fusing multivariate geological data, existing techniques lack effective global feature extraction and geological semantic considerations, making it difficult to effectively reduce model uncertainty.

[0004] Therefore, how to effectively integrate diverse geological data in 3D geological modeling to improve the accuracy of modeling has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and medium for constructing a three-dimensional geological model by integrating multi-source data, so as to achieve accurate model construction for a target geological area.

[0006] To address the aforementioned technical problems, embodiments of the present invention provide a method for constructing a three-dimensional geological model by fusing multi-source data, comprising:

[0007] Two-dimensional profile detection data of the target geological area is obtained, and the two-dimensional profile detection data is expanded into three-dimensional training data.

[0008] Spatial geological constraints are extracted from the two-dimensional profile detection data.

[0009] A multivariate clustering pattern library is constructed based on the three-dimensional training data, velocity model, and density model.

[0010] Based on the pre-defined simulation path and the spatial geological constraints, a candidate pattern library is constructed using the multi-element clustering pattern library.

[0011] The probability matrix is ​​calculated based on the candidate pattern library. Pattern selection is performed based on the probability matrix and cross-entropy to obtain the initial three-dimensional geological model.

[0012] The three-dimensional training data is used to perform multi-scale iterative optimization on the initial three-dimensional geological model to obtain the optimized three-dimensional geological model.

[0013] Furthermore, the two-dimensional profile detection data includes geological profile data, P-wave velocity profile data, and density profile data.

[0014] The step of expanding the two-dimensional profile detection data into three-dimensional training data includes:

[0015] The geological profile data, P-wave velocity profile data, and density profile data are imported into a simulation grid and randomly expanded within the simulation grid. The geological profile data, the P-wave velocity profile data, and the density profile data are all expanded into three-dimensional profiles with a certain width to obtain three-dimensional training data.

[0016] Further, the extraction of spatial geological constraints from the two-dimensional profile detection data includes:

[0017] Stratigraphic sequence information and stratigraphic thickness information are extracted from the geological profile data.

[0018] Based on the geological profile data, the spatial contact relationships between geological fault data and various geological bodies are identified.

[0019] The stratigraphic sequence information, stratigraphic thickness information, geological fault data, and spatial contact relationships are used as spatial geological constraints for the target geological region.

[0020] Further, the step of constructing a multivariate clustering pattern library based on the three-dimensional training data, velocity model, and density model includes:

[0021] A multivariate pattern library is constructed based on the aforementioned 3D training data, velocity model, and density model.

[0022] The patterns in the multivariate pattern library are clustered according to a similarity threshold. Based on the analysis results, the multivariate pattern library is classified to obtain a multivariate clustered pattern library.

[0023] Furthermore, the step of constructing a candidate pattern library using the multivariate clustering pattern library based on the pre-defined simulation path and the spatial geological constraints includes:

[0024] The positions of the nodes to be simulated are determined sequentially according to the set simulation path.

[0025] For each node to be simulated, several patterns associated with the location of the node to be simulated are extracted from the multivariate clustering pattern library.

[0026] Based on the aforementioned spatial geological constraints, each model is screened to obtain the target model.

[0027] A library of alternative modes is obtained based on the target modes of each node to be simulated.

[0028] Further, the step of calculating the probability matrix based on the candidate pattern library, and selecting a pattern based on the probability matrix and cross-entropy to obtain a three-dimensional geological initial model includes:

[0029] The lithological data similarity of each target model in the candidate model library is calculated using the Hamming distance function, and the velocity data similarity and density data similarity of each target model in the candidate model library are calculated using the Euclidean distance function.

[0030] A similarity matrix is ​​obtained based on the similarity of the lithological data, velocity data, and density data.

[0031] The data in the similarity matrix is ​​normalized to obtain the probability matrix.

[0032] The probability weight coefficients of each target mode are calculated using cross-entropy.

[0033] The probability matrix is ​​adjusted by weighting the probability matrix according to the probability weight coefficients to obtain the comprehensive probability matrix.

[0034] The target pattern with the highest probability value is selected from the comprehensive probability matrix and pasted onto the node to be simulated to obtain the initial three-dimensional geological model.

[0035] Furthermore, the step of using the three-dimensional training data to perform multi-scale iterative optimization on the initial three-dimensional geological model to obtain the optimized three-dimensional geological model includes:

[0036] The initial three-dimensional geological model was divided into multiple scales to obtain several different resolution levels.

[0037] At each resolution level, the three-dimensional training data is used to iteratively optimize the three-dimensional geological initial model, and the optimized three-dimensional geological initial model is upsampled to the next resolution level.

[0038] The iterative optimization process is repeated until the initial three-dimensional geological model reaches the highest resolution, resulting in a fully optimized three-dimensional geological model.

[0039] Another embodiment of the present invention provides a three-dimensional geological model construction device that integrates multi-source data, comprising:

[0040] The data expansion module is used to acquire two-dimensional profile detection data of the target geological area and expand the two-dimensional profile detection data into three-dimensional training data.

[0041] The constraint extraction module is used to extract spatial geological constraints from the two-dimensional profile detection data.

[0042] The pattern library construction module is used to construct a multivariate clustering pattern library based on the three-dimensional training data, velocity model, and density model.

[0043] The pattern library screening module is used to construct a candidate pattern library based on the set simulation path and the spatial geological constraints, using the multivariate clustering pattern library.

[0044] The model building module is used to calculate the probability matrix based on the candidate pattern library, and to select a pattern based on the probability matrix and cross-entropy to obtain a three-dimensional geological initial model.

[0045] The model optimization module is used to perform multi-scale iterative optimization of the initial three-dimensional geological model using the three-dimensional training data to obtain the optimized three-dimensional geological model.

[0046] Another embodiment of the present invention provides a computer device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the method for constructing a three-dimensional geological model by fusing multi-source data as described above.

[0047] In another embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, the method for constructing a three-dimensional geological model by fusing multi-source data as described above is implemented.

[0048] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following:

[0049] By integrating multi-dimensional data such as geological profiles, velocity, and density to construct a multi-dimensional clustering pattern library, and combining this with spatial geological constraints for pattern selection, the accuracy and reliability of the 3D geological model are effectively improved. The introduction of cross-entropy and probability weight calculations during the simulation process better addresses data sparsity and geological complexity, reducing model uncertainty. Furthermore, a multi-scale iterative optimization algorithm is employed to progressively refine the model resolution, avoiding the computational burden of high-resolution data processing and improving modeling efficiency. This method is not only suitable for modeling complex geological structures but also enhances the global expressiveness of geological features, significantly improving the overall performance of the model. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating the steps of a method for constructing a three-dimensional geological model by fusing multi-source data in one embodiment of the present invention.

[0051] Figure 2 This is a schematic diagram illustrating the process of constructing a three-dimensional geological model that integrates multi-source data in one embodiment of the present invention;

[0052] Figure 3 This is a structural block diagram of a three-dimensional geological model construction device that integrates multi-source data in one embodiment of the present invention;

[0053] Figure 4 A structural diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0055] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0056] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. The terms "vertical," "horizontal," "left," "right," "upper," "lower," and similar expressions used herein are for illustrative purposes only and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0057] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0058] The core of reconstructing three-dimensional geological structures lies in finding spatial pattern relationships between known geological data and inferring the possible distribution of geological attribute values ​​at unsampled locations. Geostatistics theory offers a possible solution to this problem. It utilizes sparse data and, according to a certain spatial continuity model, constructs multiple possible geological models to quantitatively describe all possible spatial distribution characteristics (Mariethoz & Caers, 2014). Among these, Multiple-Point Statistics (MPS) is a modeling method that has developed rapidly in the last two decades and has been widely used in various scientific and applied research fields, such as energy exploration, mineral resource surveys, and engineering exploration. The core of MPS lies in exploring the correlation expression between multiple points in space and introducing geological model control and geological knowledge into the simulation process, so that the simulation results not only faithfully reflect the modeling data but also effectively describe the complex spatial geometry of geological bodies.

[0059] However, the MPS method essentially considers the similarity between data events and patterns, without taking into account geological knowledge such as stratigraphic sequence and the period of geological events during the simulation process. Furthermore, most methods extract pattern features from only one training image, which inevitably leads to problems such as insufficient model constraints and an inadequate number of candidate spatial distribution patterns when data is sparse or knowledge is limited. Therefore, one embodiment of this invention provides a method for constructing a three-dimensional geological model by fusing multivariate data. For details, please refer to [link to specific documentation]. Figure 1 and Figure 2 , Figure 1 The diagram shown is a flowchart illustrating the steps of a method for constructing a three-dimensional geological model by fusing multi-source data in one embodiment of the present invention. Figure 2 The diagram illustrates the process of constructing a three-dimensional geological model by fusing multi-source data in one embodiment of the present invention, including:

[0060] S11. Obtain two-dimensional profile detection data of the target geological area, and expand the two-dimensional profile detection data into three-dimensional training data.

[0061] In data preprocessing, in order to obtain a three-dimensional spatial model, it is necessary to acquire all the profile data involved in the modeling, including geological profile data, P-wave velocity profile data, and density profile data. All of the above data are two-dimensional profile detection data.

[0062] After obtaining the two-dimensional data of the geological profile, P-wave velocity profile, and density profile mentioned above, it is necessary to expand them into three-dimensional training data (TD), specifically:

[0063] The geological profile data, P-wave velocity profile data, and density profile data are randomly expanded to obtain three-dimensional geological profile data, three-dimensional velocity profile data, and three-dimensional density profile data. Then, these three-dimensional data are spatially aligned and integrated into three-dimensional training data. Spatial alignment can provide basic data support for subsequent three-dimensional geological model construction, while ensuring data consistency and integrity.

[0064] S12. Extract spatial geological constraints from the two-dimensional profile detection data.

[0065] In constructing a three-dimensional geological model, accurately describing the spatial structure and interrelationships of geological bodies is crucial. To achieve this, key spatial geological information needs to be extracted from geological profile data. This information serves as the fundamental constraints for model construction, ensuring that the model accurately reflects the actual characteristics of the geological bodies.

[0066] First, stratigraphic sequence information and stratigraphic thickness information are extracted from geological profile data.

[0067] Geological profile data is two-dimensional data obtained by vertically cutting a geological region, recording the distribution sequence of strata and the thickness of each layer. Stratigraphic sequence information reflects the vertical arrangement of different strata; for example, some strata may be above or below others. This sequence is crucial for understanding geological history and tectonic movements. Stratigraphic thickness information provides the specific thickness of each layer, which has practical value for assessing underground resource reserves and planning the depth of engineering projects. By extracting this information, a basic framework can be provided for constructing three-dimensional geological models, clarifying the spatial location and extent of each layer.

[0068] Next, the spatial contact relationships between geological fault data and various geological bodies are identified based on geological profile data.

[0069] Geological faults are areas where strata fracture and shift during tectonic movements, affecting the continuity and integrity of geological bodies. By analyzing geological profile data, the location, strike, and dip of faults can be identified. Spatial contact relationships between different geological bodies describe their relative positions and contact patterns, such as whether some bodies may be in contact, cut, or overlap each other.

[0070] Finally, the extracted stratigraphic sequence information, stratigraphic thickness information, geological fault data, and spatial contact relationships are used as spatial geological constraints for the target geological region. In the process of constructing the three-dimensional geological model, the spatial geological constraints provide clear boundary conditions and structural framework for the model construction, ensuring that the model can realistically reflect the actual characteristics and interrelationships of geological bodies in space.

[0071] For example, during model building, the location and extent of each layer can be determined based on stratigraphic sequence information and thickness information; the continuity and integrity of the strata can be adjusted based on geological fault data; and the interaction between different geological bodies can be handled based on spatial contact relationships.

[0072] S13. Construct a multi-clustering pattern library based on the three-dimensional training data, velocity model, and density model.

[0073] A multivariate pattern library is constructed based on the aforementioned 3D training data, velocity model, and density model:

[0074] A multivariate model library was constructed using 3D training data TD, velocity model V, and density model ρ as raw data. The velocity model can be constructed based on P-wave velocity profile data, which reflects the propagation velocity characteristics of P-waves in the subsurface medium. By analyzing the propagation velocity of P-waves in different geological media, the physical properties and structural changes of the strata can be inferred. The velocity model integrates this velocity information into a 3D data volume to describe the propagation velocity distribution of the subsurface medium. The density model can be constructed based on density profile data, which reflects the density characteristics of the subsurface medium. By analyzing the density changes of different geological media, the physical properties and structural characteristics of the strata can be further understood. The density model integrates density information into a 3D data volume to describe the density distribution of the subsurface medium.

[0075] Iterate through each node of TD. If the current node has no null values ​​within the template size, extract it as a multi-variable pattern and store it in three lists. In the middle. Simultaneously, extract the coordinates of the node and store them in the coordinate list L. z In this example, the data in the list is stored as a NumPy array.

[0076] Cluster analysis is performed on the patterns in the multivariate pattern library according to a similarity threshold. Based on the analysis results, the multivariate pattern library is classified to obtain a multivariate clustered pattern library:

[0077] After constructing the multivariate pattern library, cluster analysis was further performed on these patterns. To more comprehensively capture the spatial characteristics of the patterns, the 3D patterns were divided into overlapping regions in six directions for calculation. Different patterns within the cluster are clustered according to their corresponding similarity functions. Patterns with similarity values ​​less than a set threshold are grouped together, and the original list index is used as the key data for storage. This results in a multivariate clustering pattern library. Each multivariate clustering pattern library contains 6 sub-pattern libraries, each corresponding to a pattern classification in one of the 6 directions.

[0078] S14. Based on the set simulation path and the spatial geological constraints, construct a candidate pattern library using the multi-element clustering pattern library.

[0079] In 3D geological modeling, setting the simulation path is crucial for model construction. Following the pre-defined simulation path, the positions of the nodes to be simulated are determined sequentially. This path can be random, scanning and extracting the set of nodes with null values ​​from the initial model, randomly rearranging the order of the node set but prioritizing the simulation of nodes with more conditional data; or it can be arranged in a certain order, i.e., scanning the initial model sequentially to obtain the path of the nodes to be simulated. The specific choice depends on the modeling needs and objectives.

[0080] For each node to be simulated, several patterns associated with the location of the node are extracted from the multivariate clustering pattern library. The selected patterns are spatially similar to or correlated with the node to be simulated. The correlation may be reflected in geological attributes, velocity characteristics or density characteristics, depending on the data type and geological background used in the modeling.

[0081] After extracting relevant models, these models are screened based on spatial geological constraints to obtain target models. Spatial geological constraints include stratigraphic sequence, stratigraphic thickness, geological fault data, and spatial contact relationships between different geological bodies. These constraints ensure that the selected models are geologically reasonable and conform to known geological laws and characteristics. For example, if a model does not match the known location of geological faults or contradicts the sedimentary sequence of strata, then this model may be excluded.

[0082] After screening, the target patterns will be used to build a candidate pattern library, which contains patterns that meet the geological constraints and are candidate solutions in the subsequent modeling process.

[0083] In this way, it can be ensured that each node to be simulated can select the most suitable mode from multiple reasonable candidate modes during the simulation process, thereby improving the accuracy and reliability of the model.

[0084] S15. Calculate the probability matrix based on the candidate pattern library, and select a pattern based on the probability matrix and cross-entropy to obtain a three-dimensional geological initial model.

[0085] First, the lithological data similarity of each target model in the candidate model library is calculated using the Hamming distance function. Hamming distance is suitable for categorical variables and can measure the difference between two models in lithological classification. For velocity and density data, since they are continuous variables, the Euclidean distance function is used to calculate their similarity, which can effectively measure the degree of difference between numerical data.

[0086] The calculated similarities of lithological data, velocity data, and density data are integrated into a single similarity matrix, which contains similarity information of all target patterns across different data types.

[0087] To enable the similarity of different data types to be compared on a uniform scale, the data in the similarity matrix needs to be normalized to obtain a probability matrix. The normalized data range is set between 0 and 1 to avoid calculation bias caused by differences in data units and orders of magnitude, and to ensure that each data type has the same weight in subsequent calculations.

[0088] Cross-entropy is a statistic that measures the difference between two probability distributions. It can reflect the degree of matching between a target model and known geological data. The probability weight coefficients calculated by cross-entropy can quantify the importance of each target model in geological modeling.

[0089] The normalized similarity matrix is ​​adjusted based on the calculated probability weight coefficients to obtain a comprehensive probability matrix. This further optimizes the selection of the target pattern so that the final selected pattern is more consistent with geological patterns and data characteristics.

[0090] The target pattern with the highest probability value is selected from the comprehensive probability matrix and then pasted onto the node to be simulated. This ensures that the selected pattern is optimal at each node to be simulated, thereby gradually constructing the entire three-dimensional geological initial model.

[0091] Preferably, in the process of constructing the three-dimensional geological initial model, this embodiment introduces a fully connected artificial neural network (FCN) to predict the elevation of various geological surfaces. The kernel function parameters of the FCN are characterized by the geological surface elevation, and the distribution pattern of geological surfaces is learned through training data. The specific steps are as follows:

[0092] Extract stratigraphic sequence, stratigraphic thickness, and spatial geological constraints between geological objects from geological profile data, and store this information in a database.

[0093] FCNs are used to predict the elevation of various geological surfaces. Specifically, for each type of lithology, two FCNs need to be built and trained, one for predicting the elevation of the top surface and the other for predicting the elevation of the bottom surface.

[0094] The initial model is obtained by generating the top (Surtop) and bottom (Surbott) surfaces of the geological body using FCN and assigning corresponding lithological properties between the predicted geological surfaces.

[0095] S16. The three-dimensional training data is used to perform multi-scale iterative optimization on the initial three-dimensional geological model to obtain the optimized three-dimensional geological model.

[0096] After the initial model is built, in order to further improve the accuracy and reliability of the model, it is necessary to divide the initial model into multiple scales to obtain several different resolution levels. This multi-scale division method allows the model to be gradually optimized at different scales, thereby better capturing geological features and changes from macroscopic to microscopic levels.

[0097] At each resolution level, the initial model is iteratively optimized using 3D training data. During the optimization process, the distribution of geological attributes in the model is gradually adjusted by comparing the differences between the initial model and the training data, including the boundaries of geological bodies, the distribution of geological attributes, and the spatial relationships between geological bodies.

[0098] After optimizing the current resolution level, the optimized model is upsampled to the next resolution level. Upsampling increases the model's resolution while preserving the already optimized geological features. At the new resolution level, iterative optimization continues to refine the model's details.

[0099] Repeat the iterative optimization and upsampling process described above to gradually increase the model's resolution until the highest resolution level is reached. During this process, the model is gradually improved from macroscopic structure to microscopic details, ultimately resulting in an optimized 3D geological model.

[0100] This invention presents a method for constructing 3D geological models by integrating multi-source data, including geological profiles, velocity, and density, to build a multi-source clustering pattern library. By combining spatial geological constraints with pattern selection, the accuracy and reliability of the 3D geological model are effectively improved. The introduction of cross-entropy and probability weight calculations during simulation better addresses data sparsity and geological complexity, reducing model uncertainty. Furthermore, a multi-scale iterative optimization algorithm is employed to progressively refine the model resolution, avoiding the computational burden of high-resolution data processing and improving modeling efficiency. This method is not only suitable for modeling complex geological structures but also enhances the global expressiveness of geological features, significantly improving the overall performance of the model.

[0101] This invention also provides a three-dimensional geological model construction apparatus that integrates multi-source data, used to execute the three-dimensional geological model construction method that integrates multi-source data as described above. Figure 3This is a structural block diagram of a three-dimensional geological model construction device that integrates multi-source data according to an embodiment of the present invention. The device includes:

[0102] The data expansion module 21 is used to acquire two-dimensional profile detection data of the target geological area and expand the two-dimensional profile detection data into three-dimensional training data.

[0103] The constraint extraction module 22 is used to extract spatial geological constraints from the two-dimensional profile detection data.

[0104] The pattern library construction module 23 is used to construct a multi-clustering pattern library based on the three-dimensional training data, velocity model, and density model.

[0105] The pattern library screening module 24 is used to construct a candidate pattern library based on the set simulation path and the spatial geological constraints using the multi-dimensional clustering pattern library.

[0106] The model building module 25 is used to calculate the probability matrix based on the candidate pattern library, select patterns based on the probability matrix and cross-entropy, and obtain a three-dimensional geological initial model.

[0107] The model optimization module 26 is used to perform multi-scale iterative optimization of the initial three-dimensional geological model using the three-dimensional training data to obtain the optimized three-dimensional geological model.

[0108] The technical features and effects of the device proposed in the embodiments of the present invention are the same as those of the method proposed in the embodiments of the present invention, and will not be repeated here. Each module in the above-described device can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0109] See Figure 4 This is a structural block diagram of a computer device provided in an embodiment of the present invention. The computer device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps described in the above embodiment of the method for constructing a three-dimensional geological model integrating multi-source data. Figure 1 The steps S11 to S16 described above; or, when the processor executes the computer program, it implements the functions of each module in the above-described device embodiments, such as modules 21 to 26 of the three-dimensional geological model construction device that integrates multi-source data.

[0110] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the computer device.

[0111] The computer device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the schematic diagram is merely an example of a computer device and does not constitute a limitation on the computer device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0112] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the computer device, connecting various parts of the computer device via various interfaces and lines.

[0113] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0114] If the modules integrated into the computer device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0115] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0116] Accordingly, embodiments of the present invention provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform steps in the three-dimensional geological model construction method for fusing multi-source data as described in the above embodiments, for example... Figure 1 Steps S11 to S16 as described above.

[0117] In summary, compared with the prior art, the method, apparatus, computer equipment, and computer-readable storage medium for constructing a three-dimensional geological model integrating multi-source data provided by the embodiments of the present invention have the following beneficial effects:

[0118] By integrating multi-dimensional data such as geological profiles, velocity, and density to construct a multi-dimensional clustering pattern library, and combining this with spatial geological constraints for pattern selection, the accuracy and reliability of the 3D geological model are effectively improved. The introduction of cross-entropy and probability weight calculations during the simulation process better addresses data sparsity and geological complexity, reducing model uncertainty. Furthermore, a multi-scale iterative optimization algorithm is employed to progressively refine the model resolution, avoiding the computational burden of high-resolution data processing and improving modeling efficiency. This method is not only suitable for modeling complex geological structures but also enhances the global expressiveness of geological features, significantly improving the overall performance of the model.

[0119] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for constructing a three-dimensional geological model integrating multi-source data, characterized in that, include: Two-dimensional profile detection data of the target geological area is obtained, and the two-dimensional profile detection data is expanded into three-dimensional training data; Extract spatial geological constraints from the two-dimensional profile detection data; A multivariate clustering pattern library is constructed based on the aforementioned three-dimensional training data, velocity model, and density model. Based on the pre-defined simulation path and the spatial geological constraints, a candidate pattern library is constructed using the multi-dimensional clustering pattern library. The probability matrix is ​​calculated based on the candidate pattern library. The pattern is selected based on the probability matrix and cross-entropy to obtain the initial three-dimensional geological model. The three-dimensional training data is used to perform multi-scale iterative optimization on the initial three-dimensional geological model to obtain the optimized three-dimensional geological model. The step of constructing a candidate pattern library using the multivariate clustering pattern library based on the pre-defined simulation path and the spatial geological constraints includes: The positions of the nodes to be simulated are determined sequentially according to the set simulation path; For each node to be simulated, several patterns associated with the location of the node to be simulated are extracted from the multivariate clustering pattern library; Based on the aforementioned spatial geological constraints, each model is screened to obtain the target model; A candidate mode library is obtained based on the target mode of each node to be simulated; The process of calculating a probability matrix based on the candidate pattern library, selecting a pattern based on the probability matrix and cross-entropy, and obtaining a three-dimensional geological initial model includes: The lithological data similarity of each target model in the candidate model library is calculated using the Hamming distance function, and the velocity data similarity and density data similarity of each target model in the candidate model library are calculated using the Euclidean distance function. A similarity matrix is ​​obtained based on the similarity of the lithological data, velocity data, and density data. The data in the similarity matrix are normalized to obtain the probability matrix; The probability weight coefficients of each target pattern are calculated using cross-entropy. The probability matrix is ​​adjusted according to the probability weight coefficients to obtain the comprehensive probability matrix. The target pattern with the highest probability value is selected from the comprehensive probability matrix and pasted onto the node to be simulated to obtain the initial three-dimensional geological model.

2. The method for constructing a three-dimensional geological model by integrating multi-source data as described in claim 1, characterized in that, The two-dimensional profile detection data includes geological profile data, P-wave velocity profile data, and density profile data; The step of expanding the two-dimensional profile detection data into three-dimensional training data includes: The geological profile data, P-wave velocity profile data, and density profile data are imported into a simulation grid and randomly expanded within the simulation grid to form a three-dimensional profile with a certain width, thus obtaining three-dimensional training data.

3. The method for constructing a three-dimensional geological model by integrating multi-source data as described in claim 2, characterized in that, The extraction of spatial geological constraints from the two-dimensional profile detection data includes: Stratigraphic sequence information and stratigraphic thickness information are extracted from the geological profile data; Based on the geological profile data, the spatial contact relationship between geological fault data and various geological bodies was identified; The stratigraphic sequence information, stratigraphic thickness information, geological fault data, and spatial contact relationships are used as spatial geological constraints for the target geological region.

4. The method for constructing a three-dimensional geological model by integrating multi-source data as described in claim 1, characterized in that, The construction of a multi-clustering pattern library based on the three-dimensional training data, velocity model, and density model includes: A multivariate pattern library is constructed based on the aforementioned three-dimensional training data, velocity model, and density model; The patterns in the multivariate pattern library are clustered according to a similarity threshold. Based on the analysis results, the multivariate pattern library is classified to obtain a multivariate clustered pattern library.

5. The method for constructing a three-dimensional geological model by integrating multi-source data as described in claim 1, characterized in that, The process of using the three-dimensional training data to perform multi-scale iterative optimization on the initial three-dimensional geological model to obtain the optimized three-dimensional geological model includes: The initial three-dimensional geological model is divided into multiple scales to obtain several different resolution levels; At each resolution level, the three-dimensional training data is used to iteratively optimize the three-dimensional geological initial model, and the optimized three-dimensional geological initial model is upsampled to the next resolution level. The iterative optimization process is repeated until the initial three-dimensional geological model reaches the highest resolution, resulting in a fully optimized three-dimensional geological model.

6. A device for constructing a three-dimensional geological model by integrating multi-source data, characterized in that, include: The data expansion module is used to acquire two-dimensional profile detection data of the target geological area and expand the two-dimensional profile detection data into three-dimensional training data. The constraint extraction module is used to extract spatial geological constraints from the two-dimensional profile detection data; The pattern library construction module is used to construct a multi-clustering pattern library based on the three-dimensional training data, velocity model, and density model. The pattern library screening module is used to construct a candidate pattern library based on the set simulation path and the spatial geological constraints using the multi-element clustering pattern library; The model building module is used to calculate a probability matrix based on the candidate pattern library, select a pattern based on the probability matrix and cross-entropy, and obtain a three-dimensional geological initial model. The model optimization module is used to perform multi-scale iterative optimization of the initial three-dimensional geological model using the three-dimensional training data to obtain the optimized three-dimensional geological model. The step of constructing a candidate pattern library using the multivariate clustering pattern library based on the pre-defined simulation path and the spatial geological constraints includes: The positions of the nodes to be simulated are determined sequentially according to the set simulation path; For each node to be simulated, several patterns associated with the location of the node to be simulated are extracted from the multivariate clustering pattern library; Based on the aforementioned spatial geological constraints, each model is screened to obtain the target model; A candidate mode library is obtained based on the target mode of each node to be simulated; The process of calculating a probability matrix based on the candidate pattern library, selecting a pattern based on the probability matrix and cross-entropy, and obtaining a three-dimensional geological initial model includes: The lithological data similarity of each target model in the candidate model library is calculated using the Hamming distance function, and the velocity data similarity and density data similarity of each target model in the candidate model library are calculated using the Euclidean distance function. A similarity matrix is ​​obtained based on the similarity of the lithological data, velocity data, and density data. The data in the similarity matrix are normalized to obtain the probability matrix; The probability weight coefficients of each target pattern are calculated using cross-entropy. The probability matrix is ​​adjusted according to the probability weight coefficients to obtain the comprehensive probability matrix. The target pattern with the highest probability value is selected from the comprehensive probability matrix and pasted onto the node to be simulated to obtain the initial three-dimensional geological model.

7. A computer device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the method for constructing a three-dimensional geological model by fusing multi-source data as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the method for constructing a three-dimensional geological model by fusing multi-source data as described in any one of claims 1 to 5.

Images