Gravity adaptive focusing inversion method and system based on predicted field offset and dynamic weight, storage medium and equipment

Abstract

The invention discloses a gravity adaptive focusing inversion method and system based on predictive field offset and dynamic weight, a storage medium and equipment, and belongs to the technical field of gravity inversion. According to the method, offset imaging information is converted into an adaptive weight matrix through dynamic closed-loop iteration of forward offset weight construction inversion; and coupling with a traditional depth weighting matrix and a focusing constraint matrix to form a comprehensive weight matrix, embedding the comprehensive weight matrix into a regularization inversion objective function, and iteratively updating the density model until convergence. According to the method, under the condition of not depending on a high-precision initial model, the deep resolution and the abnormal body boundary description capability of gravity inversion are synchronously improved, and the problems of model smoothness and boundary fuzziness caused by a depth attenuation effect in a traditional method are solved. The method has the advantages of strong anti-noise capability, wide applicability and the like, and is suitable for fine geological interpretation in deep mineral and oil-gas exploration.

Description

Technical Field

[0001] This invention relates to the field of gravity inversion technology, and in particular to a gravity adaptive focusing inversion method, system, storage medium and device based on predicted field offset and dynamic weight. Background Technology

[0002] Three-dimensional gravity inversion is a key technology in deep mineral and oil and gas exploration. Currently, mainstream methods include Tikhonov regularization inversion, depth-weighted inversion, focusing inversion, and migration-based fast imaging methods. Among these, Tikhonov regularization introduces regularization parameters to constrain the solution space; depth-weighted inversion constructs an adaptive depth-weighting matrix based on parameter-integrated sensitivity; focusing inversion proposes a minimum gradient-supported focusing method to sharpen boundaries; and migration-based fast imaging methods have developed potential field transfer methods to achieve rapid imaging without prior knowledge.

[0003] Despite continuous development, existing gravity inversion methods still face the following bottlenecks: 1. The Tikhonov regularization framework laid the mathematical foundation for inversion, but it cannot effectively compensate for the characteristic that gravity signals decay sharply with depth, which can easily lead to insufficient deep resolution and insignificant density recovery. 2. Although the depth-weighted method suppresses the "skin effect" through geometric weights, its weights only depend on the depth parameter and do not take into account the actual material property distribution. While alleviating the shallow slope, it causes the overall model to be smooth and the boundaries to be blurred. 3. Focused inversion methods are dedicated to sharpening the boundaries of anomalies, but they have poor inversion stability and are prone to over-focusing or geologically unreasonable solutions; 4. Although migration imaging methods do not require prior models and are computationally stable, they have low imaging resolution and weak boundary delineation capabilities, making it difficult to meet the needs of detailed geological interpretation.

[0004] Overall, existing methods struggle to achieve a balance between improved depth resolution and clear boundary characterization, still relying heavily on prior information, which limits the precise application of gravity inversion under complex geological conditions. Summary of the Invention

[0005] The purpose of this invention is to overcome the problems of model smoothing and boundary blurring caused by the depth attenuation effect in the prior art. It provides a gravity adaptive focusing inversion method, system, storage medium and device based on prediction field offset and dynamic weights, which improves the depth resolution and anomalous body boundary characterization ability of gravity inversion without relying on a high-precision initial model.

[0006] The objective of this invention is achieved through the following technical solution: A first aspect of the present invention provides a gravity adaptive focusing inversion method based on predicted field migration and dynamic weights, comprising the following steps: S1. Gravity forward modeling calculation: The underground space is discretized into multiple regular units, and the predicted gravity field is calculated based on the current predicted density model; S2. Gravity migration imaging: Using the predicted gravity field obtained in step S1, migration imaging is performed to obtain the initial migration density field; S3. Construct the offset driving weight matrix: Extract the diagonal elements of the initial offset density field and take the reciprocal of each diagonal element to obtain the offset driving weight matrix; and perform natural logarithmic transformation, Z-score normalization and Sigmoid function mapping on the offset driving weight matrix in sequence. S4. Construction of the comprehensive weight matrix: The offset-driven weight matrix is ​​coupled with the traditional sensitivity-based depth weighting matrix and focus constraint matrix to form a comprehensive weight matrix; S5. Regularized Inversion Solution: The comprehensive weight matrix is ​​embedded into the regularization objective function for inversion solution to obtain the updated prediction density model; S6. Iteration and Convergence Judgment: Substitute the updated predicted density model into step S1 to calculate the new predicted gravity field, and calculate the relative error between the new predicted gravity field and the observed gravity field; if the relative error is not less than the set threshold, return to step S2 to continue iterating; if the relative error is less than the set threshold, output the final inversion model.

[0007] In some embodiments, step S1 specifically includes: Discretize the underground space into Using regular cuboid elements, the vertical component of the gravitational field is predicted based on an integral formula: In the formula At the observation point The vertical component of the gravity anomaly; It is the gravitational constant; It represents the total number of cuboid units after discretizing the target region; It is the first The constant density value of each cuboid element; The coordinates of the source point currently being integrated; the integration kernel. It describes the geometric attenuation relationship between the source point and the observation point, and is the essential manifestation of the sensitivity of the vertical component of the gravitational field.

[0008] In some embodiments, step S2 employs an adjoint operator to perform offset imaging on the predicted gravity field.

[0009] In some embodiments, the weight elements in the offset-driven weight matrix are defined as: in, Represents the diagonal elements of the initial offset density field. .

[0010] In some embodiments, the comprehensive weight matrix is ​​calculated using the following formula: in, This represents the offset-driven weight matrix. This represents the depth-weighted matrix of system sensitivity. This represents the focus constraint matrix.

[0011] In some embodiments, the regularization objective function is: in, For the orthogonal operator, For the prior density model, For regularization parameters, This represents the prediction density model updated in the (n+1)th iteration. This represents the observed gravity data vector.

[0012] In some embodiments, the threshold value in step S6 is set to 0.005 or 0.01.

[0013] A second aspect of the present invention provides a gravity adaptive focusing inversion system based on predicted field migration and dynamic weights, comprising: The gravity forward modeling module is used to discretize the underground space into multiple regular units and calculate the predicted gravity field based on the current predicted density model. The gravity migration imaging module is used to perform migration imaging using a predicted gravity field to obtain an initial migration density field; The offset-driven weight matrix construction module is used to extract the diagonal elements of the initial offset density field, and take the reciprocal of each diagonal element to obtain the offset-driven weight matrix; and to perform natural logarithmic transformation, Z-score normalization and Sigmoid function mapping on the offset-driven weight matrix in sequence. The comprehensive weight matrix construction module is used to couple the offset-driven weight matrix with the traditional sensitivity-based depth weighting matrix and focus constraint matrix to form a comprehensive weight matrix; The regularization inversion solution module is used to embed the comprehensive weight matrix into the regularization objective function for inversion solution to obtain the updated prediction density model. The iteration and convergence judgment module is used to calculate the new predicted gravity field using the updated prediction density model, and to calculate the relative error between the new predicted gravity field and the observed gravity field. If the relative error is not less than a set threshold, it returns to the gravity migration imaging module to continue iterating. If the relative error is less than the set threshold, it outputs the final inversion model.

[0014] A third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the gravity adaptive focusing inversion method based on predicted field offset and dynamic weights as described in the first aspect.

[0015] In a fourth aspect, the present invention provides an electronic device including a memory and a processor, wherein the memory stores computer instructions executable on the processor, and the processor executes the gravity adaptive focusing inversion method based on predicted field offset and dynamic weights as described in the first aspect when executing the computer instructions.

[0016] It should be further noted that the technical features corresponding to the above-mentioned options and embodiments can be combined or substituted with each other to form new technical solutions without conflict.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. Improved depth resolution: By dynamically constructing weights through migration imaging, the influence of gravity signal attenuation with depth is suppressed, thereby improving the recovery capability of deep anomalies. Taking the reciprocal of the migration density field, a constraint strategy is implemented to assign smaller weights (weak constraints) to high migration density regions (corresponding to real anomalies) and larger weights (strong constraints) to low migration density regions, effectively guiding the inversion to focus on the anomaly region and suppressing false anomalies.

[0018] 2. Enhancing Boundary Sharpness: The migration results provide inversion constraints, combined with traditional sensitivity-based depth-weighted matrices and focus constraint matrices, to improve the sharpness of anomalous body boundaries and the accuracy of geometric morphology restoration. Specifically, the traditional sensitivity-based depth-weighted matrix compensates for depth attenuation effects, the focus constraint matrix sharpens anomalous body boundaries, and the migration-driven weighting matrix adaptively guides imaging information, ultimately achieving an organic fusion of these three mechanisms. This design simultaneously improves the restoration capability and boundary delineation accuracy of deep anomalous bodies without relying on a high-precision initial model.

[0019] 3. Not dependent on initial models and prior information: Weights are updated iteratively and adaptively to reduce reliance on prior information and make it better suited for exploring new areas; 4. Strong noise resistance: By using the prediction field instead of the residual field for migration, noise amplification is avoided, thereby improving the stability of the inversion iteration. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the gravity adaptive focusing inversion method based on predicted field migration and dynamic weights, as shown in an embodiment of the present invention. Figure 2The diagram illustrates the specific iterative inversion method of this invention. The red part represents the innovative steps of this method, and the black part represents the traditional steps of focused inversion based on the Tikhonov regularization framework. Figure 3 This is a three-dimensional residual density model of a double cube in space, as shown in an embodiment of the present invention. Figure 4 This is a schematic diagram of the XZ direction density profile at Y=1000 m, as shown in an embodiment of the present invention. Figure 5 This is a schematic diagram of the vertical component of gravity anomaly field generated on the Earth's surface by the bicubic residual density model as shown in an embodiment of the present invention. Figure 6 This is a schematic diagram comparing the inversion results of a bicubic model using three different imaging methods, as shown in an embodiment of the present invention. Detailed Implementation

[0021] The technical solution 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, not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] It should be noted that the defects in the solutions in the prior art are all the results of the inventors' practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the embodiments of this application in the following text should be the inventors' contributions to this application in the process of invention and creation, and should not be understood as technical content known to those skilled in the art.

[0023] In view of the technical problems pointed out in the background art, the present invention provides the following embodiments: In one exemplary embodiment, a gravity adaptive focusing inversion method based on predicted field shift and dynamic weights is provided, such as... Figure 1 As shown, it includes the following steps: S1. Gravity forward modeling: The underground space is discretized into multiple regular units, and the predicted gravity field is calculated based on the current predicted density model to provide input for subsequent gravity migration imaging; S2. Gravity migration imaging: Using the predicted gravity field obtained in step S1, migration imaging is performed to obtain the initial migration density field; S3. Construct the offset driving weight matrix: Extract the diagonal elements of the initial offset density field and take the reciprocal of each diagonal element to obtain the offset driving weight matrix; and perform natural logarithmic transformation, Z-score normalization and Sigmoid function mapping on the offset driving weight matrix in sequence. S4. Construction of the comprehensive weight matrix: The offset-driven weight matrix is ​​coupled with the traditional sensitivity-based depth weighting matrix and focus constraint matrix to form a comprehensive weight matrix; S5. Regularized Inversion Solution: The comprehensive weight matrix is ​​embedded into the regularization objective function for inversion solution to obtain the updated prediction density model; S6. Iteration and Convergence Judgment: Substitute the updated predicted density model into step S1 to calculate the new predicted gravity field, and calculate the relative error between the new predicted gravity field and the observed gravity field; if the relative error is not less than the set threshold, return to step S2 to continue iterating; if the relative error is less than the set threshold, output the final inversion model.

[0024] Specifically, the following combination Figure 2 Explain the specific implementation method of each step.

[0025] In step S1, the underground space is discretized into Given a regular cuboid element, assuming uniform density within the element, the vertical component of the gravitational field is predicted using an integral formula: In the formula At the observation point The vertical component of the gravity anomaly; gravitational constant ; It represents the total number of cuboid units after discretizing the target region; It is the first The constant density value of each cuboid element; The coordinates of the source point currently being integrated; the integration kernel. It describes the geometric attenuation relationship between the source point and the observation point, and is the essential manifestation of the sensitivity of the vertical component of the gravitational field.

[0026] In step S2, a fast migration imaging is performed using the predicted gravity field obtained in step S1 to obtain a density distribution estimate reflecting the spatial location and boundary morphology of the anomaly, thus obtaining the initial migration density field (the first iteration is called the initial migration density field, and subsequent iterations are simply called the migration density field). The predicted gravity field is chosen instead of the traditional observation and prediction residual field to avoid the constraint decay and noise amplification problems in the later stages of the traditional residual migration, so that the migration results can continuously provide effective spatial orientation as the model gradually approaches the true solution.

[0027] In this embodiment, the offset process employs an adjoint operator. For predicting the gravitational field To perform imaging: The approximate solution for a single offset is: in Represents the initial offset density field. The depth-weighted matrix of the vertical component of the gravitational field. The correlation coefficient.

[0028] For example, step S3 specifically includes: Initial weight generation: Take the diagonal matrix of the offset density field and take the reciprocal of each element. Specifically, extract the offset density field. diagonal elements Construct initial weight elements : in Used to avoid division by zero errors.

[0029] It should be noted that, since high offset density regions are more likely to correspond to real anomalies in the model domain regularization term, they should be given smaller weights (weak constraints) to allow the region to adjust freely according to the data; low offset density regions should be given larger weights (strong constraints) to suppress false anomalies, so the offset density should be taken as the reciprocal. Numerical compression and standardization: Perform a natural logarithmic transformation on the initial weight elements to compress the extreme value range that the reciprocal might cause: In the formula, in the formula These are the transformed weight elements.

[0030] Next, Z-score standardization is performed to eliminate differences in units and improve the comparability of data from different regions: In the formula, These are the standardized weight values; The mean is the log-weighted average. Its standard deviation, The total number of elements in the diagonal matrix Range mapping and clipping: Finally, the weights are mapped to a preset range using the Sigmoid function. While controlling the weight range, the directional characteristics of the anomalous structure are preserved to avoid over-reliance on imaging information due to limited offset accuracy. The cropping operation ensures that all weights strictly fall within this range. In the formula This indicates that the input value will be clipped to a range. The operator.

[0031] Constructing a matrix: Placed on the diagonal, forming a diagonal weight matrix , used for the Constraints of the next iteration.

[0032] In step S4, the combined weight matrix simultaneously enhances both depth resolution and boundary delineation capabilities: in, This represents the offset-driven weight matrix. This represents the depth-weighted matrix of system sensitivity. This represents the focus constraint matrix.

[0033] In step S5, the comprehensive weight matrix is ​​embedded with the following regularization objective function: in, For the orthogonal operator, For the prior density model, For regularization parameters, This represents the prediction density model updated in the (n+1)th iteration. This represents the observed gravity data vector.

[0034] During the inversion solution, the conjugate gradient method is used to solve the minimization problem and update the prediction density model.

[0035] In step S6, iterative convergence control is performed: after each iteration, the relative error between the predicted field and the observed field is calculated. If the error is less than a preset threshold (usually set to 0.005 or 0.01), the iteration stops and the model is output; otherwise, the process returns to step S1 and continues the loop using the updated prediction density model.

[0036] Furthermore, the optimal implementation parameters for this method are recommended as follows: The Sigmond mapping interval is selected as [0.2, 10], which is the empirically optimal interval and can balance constraint strength and numerical stability in most models; regularization parameters... It is recommended to use the L-curve method or an adaptive descent strategy to dynamically adjust during iteration; the iteration termination threshold can be 0.005 for synthetic data and 0.05 for measured noisy data.

[0037] To verify the effectiveness of this method in inverting high-density anomalies, a three-dimensional bicubic density model was constructed. This model extends 2000 meters in both the X and Y directions, with a vertical depth of 1000 meters. It is discretized using a regular mesh, with 40 meshes uniformly divided in both the X and Y directions and 20 meshes in the Z direction, resulting in 32,000 cubic elements, each 50 meters × 50 meters × 50 meters. The model contains two cubic anomalies with a residual density of +1.0 g / cm³, with the top buried at a depth of 250 meters and the bottom at a depth of 500 meters. Figure 3 For a three-dimensional representation of the model, Figure 4 This is the density profile along the XZ direction at Y=1000 meters. Figure 5 This corresponds to the distribution of the vertical component of gravity anomaly observed on the Earth's surface.

[0038] Furthermore, Figure 6 This paper presents a comparison of the inversion results of the bicubic model using three different imaging methods. The results obtained using the traditional regularized focusing method (…) Figure 6 a) shows that its focusing effect is limited, the boundaries of the inverted anomalies are blurred, some exceed the preset model boundaries (shown in purple boxes), and the recovered residual density values ​​are significantly lower than the values ​​set in the true model. The static weighted inversion method based on a single migration result ( Figure 6 (b) Improvements have been made in the characterization of anomalous body boundaries, with the inverted anomalous body central region basically matching the preset model range. However, the recovered residual density values ​​still show a significant difference from the true values. The offset-driven weighted adaptive focusing inversion method proposed in this invention ( Figure 6 c) It exhibits superior performance in both boundary focusing and property recovery: the morphology of the inverted anomaly matches the boundary of the preset model well, and the residual density value of its core region is closer to the true set value, indicating that the method effectively improves the recovery accuracy of density parameters while enhancing the clarity of the boundary.

[0039] In another exemplary embodiment, based on the same inventive concept as the method embodiment, a gravity adaptive focusing inversion system based on predicted field shift and dynamic weights is provided, comprising: The gravity forward modeling module is used to discretize the underground space into multiple regular units and calculate the predicted gravity field based on the current predicted density model. The gravity migration imaging module is used to perform migration imaging using a predicted gravity field to obtain an initial migration density field; The offset-driven weight matrix construction module is used to extract the diagonal elements of the initial offset density field, and take the reciprocal of each diagonal element to obtain the offset-driven weight matrix; and to perform natural logarithmic transformation, Z-score normalization and Sigmoid function mapping on the offset-driven weight matrix in sequence. The comprehensive weight matrix construction module is used to couple the offset-driven weight matrix with the traditional sensitivity-based depth weighting matrix and focus constraint matrix to form a comprehensive weight matrix; The regularization inversion solution module is used to embed the comprehensive weight matrix into the regularization objective function for inversion solution to obtain the updated prediction density model. The iteration and convergence judgment module is used to calculate the new predicted gravity field using the updated prediction density model, and to calculate the relative error between the new predicted gravity field and the observed gravity field. If the relative error is not less than a set threshold, it returns to the gravity migration imaging module to continue iterating. If the relative error is less than the set threshold, it outputs the final inversion model.

[0040] Specifically, firstly, an initial subsurface model is constructed using the gravity forward modeling module, and observed gravity data is acquired. In the first iteration, the observed data is input into the gravity migration imaging module to obtain the initial migration density field. Subsequently, the migration-driven weight matrix construction module sequentially performs reciprocal, natural logarithmic transformation, Z-score normalization, and Sigmoid function mapping on the density field to construct the migration-driven weight matrix. The comprehensive weight matrix construction module combines it with the traditional sensitivity depth weighting matrix. and focus constraint matrix Coupling to form a comprehensive weight matrix Next, the regularization inversion solution module will... The inversion solution is performed within a regularized objective function to obtain an updated predicted density model. Subsequently, an iteration and convergence judgment module calculates the residuals between the predicted gravity field and the observed field. If the convergence threshold is not met, the predicted gravity field is obtained through forward modeling based on the current prediction model and used as the input for the next migration. This process of migration imaging, weight matrix update, and inversion iteration is repeated until the residuals converge, ultimately outputting a high-resolution subsurface density structure model. This workflow achieves adaptive fusion of migration density and inversion constraints through a closed-loop iteration of "forward modeling, migration weights, and inversion construction."

[0041] In another exemplary embodiment, based on the same inventive concept as the method embodiment, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which, when executed by a processor, implements the gravity adaptive focusing inversion method based on predicted field offset and dynamic weights provided in this embodiment of the invention. Based on this understanding, the technical solution of this embodiment, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0042] In another exemplary embodiment, based on the same inventive concept as the method embodiment, an electronic device is provided, including a memory and a processor. The memory stores computer instructions that can be executed on the processor. When the processor executes the computer instructions, it performs the gravity adaptive focusing inversion method based on predicted field offset and dynamic weight provided in the embodiment of the present invention.

[0043] The processor may be a single-core or multi-core central processing unit or a specific integrated circuit, or one or more integrated circuits configured to implement the present invention.

[0044] The embodiments of the subject matter and functional operation described in this specification can be implemented in: tangibly embodied computer software or firmware, computer hardware including the structures disclosed in this specification and their structural equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible, non-transitory program carrier for execution by a data processing device or for controlling the operation of a data processing device. Alternatively or additionally, the program instructions may be encoded on artificially generated propagation signals, such as machine-generated electrical, optical, or electromagnetic signals, which are generated to encode information and transmit it to a suitable receiving device for execution by the data processing device.

[0045] The processing and logic flow described in this specification can be executed by one or more programmable computers that execute one or more computer programs to perform corresponding functions by operating on input data and generating output. The processing and logic flow can also be executed by dedicated logic circuitry—such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits), and the device can also be implemented as dedicated logic circuitry.

[0046] Suitable processors for executing computer programs include, for example, general-purpose and / or special-purpose microprocessors, or any other type of central processing unit. Typically, the central processing unit receives instructions and data from read-only memory and / or random access memory. The basic components of a computer include a central processing unit for implementing or executing instructions and one or more memory devices for storing instructions and data. Typically, a computer will also include one or more mass storage devices for storing data, such as disks, magneto-optical disks, or optical disks, or the computer will be operatively coupled to such mass storage devices to receive data from or transfer data to them, or both. However, a computer is not required to have such devices. Furthermore, a computer can be embedded in another device, such as a mobile phone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a global positioning system (GPS) receiver, or a portable storage device such as a universal serial bus (USB) flash drive, to name a few.

[0047] It should be understood that each block in a flowchart or block diagram can represent a module, segment, or portion of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0048] The above detailed embodiments are a description of the present invention. It should not be considered that the specific embodiments of the present invention are limited to these descriptions. For those skilled in the art, several simple deductions and substitutions can be made without departing from the concept of the present invention, and all of these should be considered to fall within the protection scope of the present invention.

Claims

1. A gravity adaptive focusing inversion method based on predicted field migration and dynamic weights, characterized in that, Includes the following steps: S1. Gravity forward modeling calculation: The underground space is discretized into multiple regular units, and the predicted gravity field is calculated based on the current predicted density model; S2. Gravity migration imaging: Using the predicted gravity field obtained in step S1, migration imaging is performed to obtain the initial migration density field; S3. Construct the offset driving weight matrix: Extract the diagonal elements of the initial offset density field and take the reciprocal of each diagonal element to obtain the offset driving weight matrix; and perform natural logarithmic transformation, Z-score normalization and Sigmoid function mapping on the offset driving weight matrix in sequence. S4. Construction of the comprehensive weight matrix: The offset-driven weight matrix is ​​coupled with the traditional sensitivity-based depth weighting matrix and focus constraint matrix to form a comprehensive weight matrix; S5. Regularized Inversion Solution: The comprehensive weight matrix is ​​embedded into the regularization objective function for inversion solution to obtain the updated prediction density model; S6. Iteration and Convergence Judgment: Substitute the updated predicted density model into step S1 to calculate the new predicted gravity field, and calculate the relative error between the new predicted gravity field and the observed gravity field; if the relative error is not less than the set threshold, return to step S2 to continue iterating; if the relative error is less than the set threshold, output the final inversion model.

2. The gravity adaptive focusing inversion method based on predicted field migration and dynamic weights according to claim 1, characterized in that, Step S1 specifically includes: Discretize the underground space into Using regular cuboid elements, the vertical component of the gravitational field is predicted based on an integral formula: In the formula At the observation point The vertical component of the gravity anomaly; It is the gravitational constant; It represents the total number of cuboid units after discretizing the target region; It is the first The constant density value of each cuboid element; The coordinates of the source point currently being integrated; the integration kernel It describes the geometric attenuation relationship between the source point and the observation point, and is the essential manifestation of the sensitivity of the vertical component of the gravitational field.

3. The gravity adaptive focusing inversion method based on predicted field migration and dynamic weights according to claim 1, characterized in that, In step S2, the adjoint operator is used to perform migration imaging on the predicted gravity field.

4. The gravity adaptive focusing inversion method based on predicted field migration and dynamic weights according to claim 1, characterized in that, The weight elements in the offset-driven weight matrix are defined as follows: in, Represents the diagonal elements of the initial offset density field. .

5. The gravity adaptive focusing inversion method based on predicted field migration and dynamic weights according to claim 1, characterized in that, The comprehensive weight matrix is ​​calculated using the following formula: in, This represents the offset-driven weight matrix. This represents the depth-weighted matrix of system sensitivity. This represents the focus constraint matrix.

6. The gravity adaptive focusing inversion method based on predicted field migration and dynamic weights according to claim 1, characterized in that, The regularization objective function is: in, For the orthogonal operator, For the prior density model, For regularization parameters, This represents the prediction density model updated in the (n+1)th iteration. This represents the observed gravity data vector.

7. The gravity adaptive focusing inversion method based on predicted field migration and dynamic weights according to claim 1, characterized in that, In step S6, the threshold is set to 0.005 or 0.

01.

8. A gravity adaptive focusing inversion system based on predicted field migration and dynamic weights, characterized in that, include: The gravity forward modeling module is used to discretize the underground space into multiple regular units and calculate the predicted gravity field based on the current predicted density model. The gravity migration imaging module is used to perform migration imaging using a predicted gravity field to obtain an initial migration density field; The offset-driven weight matrix construction module is used to extract the diagonal elements of the initial offset density field, and take the reciprocal of each diagonal element to obtain the offset-driven weight matrix; and to perform natural logarithmic transformation, Z-score normalization and Sigmoid function mapping on the offset-driven weight matrix in sequence. The comprehensive weight matrix construction module is used to couple the offset-driven weight matrix with the traditional sensitivity-based depth weighting matrix and focus constraint matrix to form a comprehensive weight matrix; The regularization inversion solution module is used to embed the comprehensive weight matrix into the regularization objective function for inversion solution to obtain the updated prediction density model. The iteration and convergence judgment module is used to calculate the new predicted gravity field using the updated prediction density model, and to calculate the relative error between the new predicted gravity field and the observed gravity field. If the relative error is not less than a set threshold, it returns to the gravity migration imaging module to continue iterating. If the relative error is less than the set threshold, it outputs the final inversion model.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the gravity adaptive focusing inversion method based on predicted field offset and dynamic weight as described in any one of claims 1-7.

10. An electronic device comprising a memory and a processor, wherein the memory stores computer instructions executable by the processor, characterized in that, When the processor executes computer instructions, it performs the gravity adaptive focusing inversion method based on predicted field offset and dynamic weights as described in any one of claims 1-7.

Images