A comprehensive inversion method suitable for fine analysis of three-dimensional magnetic structure of earth's crust

Abstract

The application provides a comprehensive inversion method suitable for fine analysis of three-dimensional magnetic structure of the earth's crust, which comprises the following steps: obtaining regional aeromagnetic data of a study area, and fusing the regional aeromagnetic data to form an aeromagnetic total field anomaly covering the study area; constructing a magnetic gradient tensor according to three magnetic anomaly components of the aeromagnetic total field anomaly, and calculating normalized magnetic source strength; performing weighted focusing three-dimensional inversion according to the normalized magnetic source strength to obtain a shallow three-dimensional magnetic structure of the study area; calculating a magnetic anomaly modulus according to the three magnetic anomaly components of the aeromagnetic total field anomaly; and solving a comprehensive inversion target function by taking the shallow three-dimensional magnetic structure as an a priori model to obtain a three-dimensional magnetic structure of the study area. The embodiment of the application fully utilizes the complementary advantages that the normalized magnetic source strength is weakly sensitive to the magnetization direction and can distinguish the shallow part, and the magnetic anomaly modulus can retain the deep low-frequency signal, and overcomes the bottleneck that a single conversion quantity inversion cannot depict a multi-scale magnetic body.

Description

Technical Field

[0001] This invention relates to the fields of geophysics and geochemistry, and in particular to a comprehensive inversion method suitable for fine analysis of the three-dimensional magnetic structure of the Earth's crust. Background Technology

[0002] Magnetic exploration boasts advantages such as high exploration efficiency, low economic cost, wide operating range, and numerous applications, making it an important tool for basic geological surveys, deep resource exploration, geothermal research, and structural analysis. However, in mid- and low-latitude regions (such as the Tibetan Plateau orogenic belt), magnetic bodies often exhibit significant remanence, with the direction of remanence inconsistent with the direction of induced magnetization. This causes the morphology of the total magnetic anomaly to deviate significantly from the true location of the magnetic body, resulting in distortion of conventional inversion results and an inability to accurately characterize the location, morphology, and physical properties of the magnetic body.

[0003] To mitigate the influence of magnetization direction on inversion results, the industry commonly uses magnetic conversion quantities that are insensitive or weakly sensitive to magnetization direction to replace the total field anomaly for inversion. Among these, the two most widely used conversion quantities are Normalized Source Strength (NSS) and Total Magnetic Field Anomaly (Ta). NSS is composed of the eigenvalues ​​of the magnetic gradient tensor matrix. It is completely independent of the magnetization direction for the dipole field and is weakly sensitive to the magnetization direction in the case of three-dimensional field sources, thus gaining widespread application.

[0004] However, NSS is inversely proportional to the fourth power of distance, and the field source decays extremely rapidly, resulting in single NSS inversions primarily responding to shallow magnetic bodies, with severe loss of deep information. Ta is synthesized from three components of magnetic anomaly, is inversely proportional to the cube of distance, and is of the same order of magnitude as the magnetic anomaly, preserving long-wavelength low-frequency information reflecting deep geological structures; however, single Ta inversions result in the coupling of shallow high frequencies and deep low frequencies, making it difficult to distinguish the position and morphology of magnetic bodies at different scales. Summary of the Invention

[0005] This invention provides a comprehensive inversion method suitable for fine analysis of the three-dimensional magnetic structure of the Earth's crust. Its main purpose is to give full play to the complementary advantages of normalized magnetic source intensity being weakly sensitive to magnetization direction and distinguishing shallow magnetic anomaly modulus while preserving deep low-frequency signals, thus overcoming the bottleneck that single conversion inversion cannot characterize multi-scale magnetic bodies.

[0006] In a first aspect, embodiments of the present invention provide a comprehensive inversion method suitable for fine analysis of the three-dimensional magnetic structure of the Earth's crust, comprising:

[0007] S1, acquire regional aeromagnetic data of the study area, and fuse the regional aeromagnetic data to form a total aeromagnetic field anomaly covering the study area;

[0008] S2, Based on the three magnetic anomaly components of the aeromagnetic total field anomaly, construct the magnetic gradient tensor and calculate the normalized magnetic source intensity;

[0009] S3. Based on the normalized magnetic source intensity, a weighted focusing three-dimensional inversion is performed to obtain the shallow three-dimensional magnetic structure of the study area;

[0010] S4. The magnetic anomaly modulus is calculated based on the three magnetic anomaly components of the aeromagnetic total field anomaly.

[0011] S5. Using the shallow three-dimensional magnetic structure as a priori model, a comprehensive inversion objective function is constructed by incorporating the magnetic anomaly modulus in the form of a reference model deviation term. The comprehensive inversion objective function is solved to obtain the three-dimensional magnetic structure of the study area.

[0012] Further, step S1 includes:

[0013] S11, For the initial aeromagnetic data of each measurement area in the study area, the low-altitude aeromagnetic data in the initial aeromagnetic data is unified to the reference altitude through the frequency domain upward extension operator to obtain the upwardly extended aeromagnetic data;

[0014] S12. Based on the aeromagnetic data after upward extension of the overlapping zone in adjacent survey areas, a univariate linear regression model is constructed to solve the offset coefficient and the proportional coefficient, and the benchmark correction transformation formula between adjacent survey areas is obtained.

[0015] S13, using the reference correction transformation formula, perform reference correction on the aeromagnetic data extended upward in the adjacent survey areas to obtain the reference-corrected aeromagnetic data for each survey area.

[0016] S14. By interpolation, the scale of the aeromagnetic data after benchmark correction for each survey area is unified to obtain the aeromagnetic data after unified scale.

[0017] S15. By using a stitching method or a hybrid method, the aeromagnetic data of each survey area at a unified scale are automatically stitched together to obtain the regional aeromagnetic data of the study area.

[0018] Further, step S2 includes:

[0019] S21, convert the aeromagnetic total field anomaly into three magnetic anomaly components;

[0020] S22, calculate the spatial partial derivatives of the three magnetic anomaly components within the frequency range, construct the second-order gradient tensor, and obtain the tensor matrix;

[0021] S23, by performing eigenvalue decomposition on the tensor matrix, three eigenvalues ​​are obtained, and the normalized magnetic source intensity is calculated based on the three eigenvalues.

[0022] Further, step S3 includes:

[0023] S31. Based on the basic theory of Tikhonov regularization inversion, a model weighting function is constructed, and the subsurface medium below the surface of the study area is divided into M uniform orthogonal subsurface grid units in three-dimensional space. The continuous magnetic susceptibility in each grid is discretized and expressed, thereby constructing a model space of length M, where M is a positive integer.

[0024] S32 employs a focused support functional stabilizer and deep weighting to construct a model weighting matrix;

[0025] S33, the transformation of weighted model parameters depends on the introduced model weighting matrix, and a deep weighting matrix is ​​constructed;

[0026] S34. Based on the model weighting matrix and the depth weighting matrix, construct the objective function for shallow inversion, and solve the objective function for shallow inversion to obtain the shallow three-dimensional magnetic structure.

[0027] Further, in step S34, solving the objective function of the shallow inversion to obtain the shallow three-dimensional magnetic structure includes:

[0028] The objective function for the shallow inversion is iteratively optimized using the conjugate gradient method: in each iteration, the response of the shallow three-dimensional magnetic structure is calculated and continuously updated by comparing it with the observed normalized magnetic source intensity until the data fitting difference meets the set threshold, and finally the shallow three-dimensional magnetic structure is obtained.

[0029] Furthermore, step S4 is achieved through the following formula:

[0030] ;

[0031] in, This represents the magnetic anomaly modulus. , , This indicates the components of the aeromagnetic total field anomaly in the X, Y, and Z directions.

[0032] Furthermore, the calculation method for the comprehensive inversion objective function in step S5 is as follows:

[0033] ;

[0034] ;

[0035] in, Denotes the comprehensive inversion objective function, This represents the model parameters to be solved. The nonlinear forward modeling operator representing the magnetic anomaly modulus, This refers to the shallow three-dimensional magnetic structure. Represents a data weighting matrix. Represents the depth-weighted matrix. Represents the regularization factor. This represents the magnetic anomaly modulus;

[0036] Indicates the first The forward modeling response of the magnetic anomaly modulus at each observation point , and These are the sensitivity matrices corresponding to the X, Y, and Z directions, respectively. , and They represent the first At each observation point, the model... The generated magnetic field components in three directions It is a positive integer.

[0037] Further, in step S5, the step of solving the comprehensive inversion objective function includes:

[0038] The first derivative of the comprehensive inversion objective function is calculated and set to zero. The inversion is started with a preset initial value during the iteration process. The conjugate gradient method is used to solve the comprehensive inversion objective function, and finally the three-dimensional magnetic structure of the inversion model is obtained.

[0039] The data weighting matrix is ​​a unit diagonal matrix, and the regularization factor is an adaptive regularization factor used for inversion calculation.

[0040] Furthermore, the regularization factor is calculated using an adaptive regularization factor, including:

[0041] The first iteration of the steepest descent of the conjugate gradient method in Under the circumstances, in which, This represents the initial value of the regularization factor;

[0042] The initial value of the regularization parameter is determined after the first iteration according to the following formula:

[0043] ;

[0044] in, The regularization factor during the first iteration. This represents the current model during the first iteration. This represents the normalized magnetic source strength. This represents the model weighting matrix during the first iteration;

[0045] The regularization factor is adjusted in later iterations according to the following formula:

[0046] , ;

[0047] in, Indicates the first The regularization factor at the next iteration Indicates the first The regularization factor at the next iteration This represents the iteration factor.

[0048] Secondly, embodiments of the present invention provide a comprehensive inversion system suitable for fine analysis of the three-dimensional magnetic structure of the Earth's crust, comprising:

[0049] The acquisition module is used to acquire regional aeromagnetic data of the study area and fuse the regional aeromagnetic data to form a total aeromagnetic field anomaly covering the study area.

[0050] The normalized magnetic source intensity module is used to construct a magnetic gradient tensor based on the three magnetic anomaly components of the aeromagnetic total field anomaly and to calculate the normalized magnetic source intensity.

[0051] The weighted inversion module is used to perform weighted focused three-dimensional inversion based on the normalized magnetic source intensity to obtain the shallow three-dimensional magnetic structure of the study area;

[0052] The magnetic anomaly modulus module is used to calculate the magnetic anomaly modulus based on the three magnetic anomaly components of the aeromagnetic total field anomaly.

[0053] The comprehensive inversion module is used to construct a comprehensive inversion objective function by using the shallow three-dimensional magnetic structure as a priori model and incorporating the magnetic anomaly modulus in the form of a reference model deviation term. The comprehensive inversion objective function is then solved to obtain the three-dimensional magnetic structure of the study area.

[0054] Thirdly, embodiments of the present invention provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-described comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust.

[0055] Fourthly, embodiments of the present invention provide a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust.

[0056] This invention proposes a comprehensive inversion method suitable for fine analysis of the three-dimensional magnetic structure of the Earth's crust, namely the NSS-Ta comprehensive inversion imaging method. It innovatively uses the depth-weighted focusing inversion results of NSS as a priori model and incorporates them into the Ta inversion objective function in the form of a "reference model deviation term" to achieve multi-scale inversion and integrated reconstruction of shallow high-frequency information (anchored by NSS) and deep low-frequency information (driven by Ta data).

[0057] The integrated inversion method fully leverages the complementary advantages of NSS (slightly sensitive to magnetization direction and able to distinguish shallow parts) and Ta (retaining deep low-frequency signals), overcoming the bottleneck that single-conversion inversion cannot characterize multi-scale magnetic bodies. It can be widely applied to the fine characterization of three-dimensional magnetic structures in basin basements, orogenic belts, and marine areas, providing technical support for deep mineral resource exploration, geothermal research, and structural analysis. Attached Figure Description

[0058] Figure 1 A flowchart of a comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust, provided in an embodiment of the present invention;

[0059] Figure 2 A flowchart of a comprehensive inversion system for fine analysis of the three-dimensional magnetic structure of the Earth's crust, provided as an embodiment of the present invention.

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0062] To enable those skilled in the art to better understand the solutions of this application, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0063] In the embodiments of this application, "at least one" refers to one or more; "multiple" refers to two or more. In the description of this application, terms such as "first," "second," and "third" are used only for descriptive purposes and should not be construed as indicating or implying relative importance or order. Furthermore, the terms "first" and "second" 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 as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0064] References such as “one embodiment” or “some embodiments” as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the terms “comprising,” “including,” “having,” and variations thereof, as used in this specification, mean “including, but not limited to,” unless otherwise specifically emphasized.

[0065] Figure 1 A flowchart of a comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust, provided as an embodiment of the present invention, is shown below. Figure 1 As shown, the method includes:

[0066] S1, acquire regional aeromagnetic data of the study area, and fuse the regional aeromagnetic data to form a total aeromagnetic field anomaly covering the study area;

[0067] Since flight data from a single measurement area is usually insufficient to fully cover the study area, it is necessary to spatially stitch together aeromagnetic data from multiple measurement areas to construct a basic data volume that can be used for subsequent conversion, processing, and inversion.

[0068] Given the objective differences in flight altitude, background reference, and grid resolution between different periods of aerial surveys, a processing procedure of "altitude normalization, reference correction, and grid fusion" is required.

[0069] Specifically, step S1 includes:

[0070] S11, For the initial aeromagnetic data of each measurement area in the study area, the low-altitude aeromagnetic data in the initial aeromagnetic data is unified to the reference altitude through the frequency domain upward extension operator to obtain the upwardly extended aeromagnetic data;

[0071] In this embodiment of the invention, initial aerial survey data for each survey area within the study area are first collected. The study area is the overall target area for this exploration and geological analysis, the complete geographical range to be fully imaged by the project, and the boundary of the entire work area. Dividing the entire study area into different survey areas is to facilitate field aerial surveying.

[0072] During the data collection process, all magnetic field anomaly data measured by magnetometers on all aircraft are collectively referred to as initial aeromagnetic data, which is further subdivided into two categories based on flight altitude: high-altitude aeromagnetic data and low-altitude aeromagnetic data. High-altitude aeromagnetic data has a large flight altitude and a wide measurement area coverage, serving as a reference benchmark for a unified altitude. Low-altitude detection data is low-altitude flight aeromagnetic measurement data, which has higher resolution, but the observation surface is lower, and the anomalous amplitude will be distorted due to terrain undulations and altitude differences.

[0073] First, to eliminate abnormal amplitude distortion caused by undulations in the observation surface, data from the survey area at a relatively high flight altitude was selected as the target reference surface. For low-altitude aeromagnetic data... The frequency domain upward extension operator is used to unify it to the reference height:

[0074] (1)

[0075] in, This represents the aeromagnetic data after upward extension; This is the upper extension height, which is the difference between the lower height data and the reference height; ,in, , These are the frequencies along the X and Y axes, respectively. Indicates Fourier transform; This represents the inverse Fourier transform.

[0076] S12. Based on the aeromagnetic data after upward extension of the overlapping zone in adjacent survey areas, a univariate linear regression model is constructed to solve the offset coefficient and the proportional coefficient, and the benchmark correction transformation formula between adjacent survey areas is obtained.

[0077] Secondly, in order to correct the background reference drift caused by the long-term variation of the geomagnetic field in different survey areas, a univariate linear regression model was established in the overlapping zone of adjacent survey areas.

[0078] In practice, magnetic anomaly values ​​at the same location in two sets of survey areas can only be extracted if there is an overlapping coverage area between adjacent survey areas. Then, a univariate linear regression is constructed using paired observations within the overlapping zone to solve for the offset coefficient and the scaling factor, thereby obtaining the benchmark correction transformation formula between the two sets of survey blocks. The correction coefficients obtained from the regression are then applied to all survey points in the entire survey area (not just the overlapping zone) to complete the background benchmark translation correction for all aeromagnetic data in that block.

[0079] definition For reference area data, The spatial correlation of the data to be corrected can be expressed as:

[0080] (2)

[0081] in, This is the intercept, i.e., the offset coefficient; This is the regression coefficient, also known as the proportionality coefficient; This is the residual term.

[0082] S13, using the reference correction transformation formula, perform reference correction on the aeromagnetic data extended upward in the adjacent survey areas to obtain the reference-corrected aeromagnetic data for each survey area.

[0083] The intercept is obtained by fitting based on the least squares criterion. With regression coefficient ( After the residual term, for Implement datum transformation:

[0084] (3)

[0085] After the standardization and Located under the same benchmark, This refers to the aeromagnetic data after benchmark correction for each survey area.

[0086] S14. By interpolation, the scale of the aeromagnetic data after benchmark correction for each survey area is unified to obtain the aeromagnetic data after unified scale.

[0087] The aeromagnetic data after benchmark correction is unified to the target scale by interpolation method, resulting in aeromagnetic data with a unified scale.

[0088] S15. By using a stitching method or a hybrid method, the aeromagnetic data of each survey area at a unified scale are automatically stitched together to obtain the regional aeromagnetic data of the study area.

[0089] After the scale was standardized, the regional aeromagnetic and magnetic field data covering the study area were obtained directly by using the GridKnitting data stitching module in the geophysical data processing software Oasis Montaj, selecting either the stitching method or a hybrid method for automatic stitching. .

[0090] S2, Based on the three magnetic anomaly components of the aeromagnetic total field anomaly, construct the magnetic gradient tensor and calculate the normalized magnetic source intensity;

[0091] The normalized magnetic source intensity is a magnetic field conversion quantity that is weakly sensitive to the magnetization direction and is inversely proportional to the fourth power of the source distance. It can highlight how shallow magnetic anomalies suppress information from deeper anomalies. The normalized magnetic source intensity can be obtained through the total field intensity. The eigenvalues ​​of the gradient tensor matrix are derived from this.

[0092] It should be noted that the regional aeromagnetic data, after undergoing highly unified, long-term variable benchmark correction, and stitching, forms a complete grid / measuring point dataset. The original records are observations of the total geomagnetic field intensity, comprising three parts: the main magnetic field, the varying field, and the total magnetic anomaly. Their specific meanings are as follows:

[0093] Main magnetic field: Deep Earth dipole magnetic field (IGRF International Geomagnetic Reference Field);

[0094] Variation fields: external interference fields such as diurnal variations and long-term variations;

[0095] ΔT: Total magnetic anomaly (a useful signal generated by underground magnetic rock mass).

[0096] Step S1 only performs data fusion, height reduction, and baseline drift correction, without separating the main magnetic field and interference fields. Therefore, this set of regional aeromagnetic data is the complete original dataset of the total observation field, not a pure anomaly.

[0097] From the regional aeromagnetic total field observations in step S1, after deducting the IGRF main magnetic field and eliminating external magnetic fields such as diurnal variations, the residual field caused only by the magnetism of underground geological bodies, namely the total magnetic anomaly, is extracted. This is the basic input for gradient tensor and NSS calculations.

[0098] Specifically, step S2 includes:

[0099] S21, convert the aeromagnetic total field anomaly into three magnetic anomaly components;

[0100] In this embodiment of the invention, the aeromagnetic total field anomaly is converted into three magnetic anomaly components using the following formula. , and :

[0101] (4)

[0102] in, Background field direction factor , , Represents the north, east, and vertical direction cosines determined by the geomagnetic inclination angle I and the geomagnetic declination angle D, where, , , .

[0103] S22, calculate the spatial partial derivatives of the three magnetic anomaly components within the frequency range, construct the second-order gradient tensor, and obtain the tensor matrix;

[0104] Secondly, the spatial partial derivatives of the three components are obtained in the frequency domain to construct a second-order gradient tensor. Since the external space of the geomagnetic field source is a source-free field, it satisfies the Laplace equation, and the tensor matrix is... as follows:

[0105] (5)

[0106] in, , , , , These are the five independent elements of the magnetic gradient tensor, which are also the three components of the magnetic anomaly. , , The corresponding spatial partial derivatives.

[0107] S23, by performing eigenvalue decomposition on the tensor matrix, three eigenvalues ​​are obtained, and the normalized magnetic source intensity is calculated based on the three eigenvalues.

[0108] By performing eigenvalue decomposition on a tensor matrix, we can obtain the three eigenvalues ​​corresponding to the tensor matrix. , and (in, Then, the normalized magnetic source strength is calculated by combining the eigenvalue invariants. :

[0109] (6)

[0110] in, It is inversely proportional to the fourth power of the distance, weakly sensitive to the magnetization direction, and completely unaffected by the magnetization direction in the case of a dipole field. It is the ideal conversion quantity for inverting shallow magnetic bodies.

[0111] S3. Based on the normalized magnetic source intensity, a weighted focusing three-dimensional inversion is performed to obtain the shallow three-dimensional magnetic structure of the study area;

[0112] The results obtained in step S2 The data was subjected to deep weighted focusing 3D inversion to obtain the shallow 3D magnetic structure of the study area.

[0113] In one implementation, step S3 includes:

[0114] S31. Based on the basic theory of Tikhonov regularization inversion, a model weighting function is constructed, and the subsurface medium below the surface of the study area is divided into M uniform orthogonal subsurface grid units in three-dimensional space. The continuous magnetic susceptibility in each grid is discretized and expressed, thereby constructing a model space of length M, where M is a positive integer.

[0115] Its stable functional is: (7)

[0116] in, It is a weighted Tikhonov regularization parameter functional. It is a weighted model-stable functional. It is a regularization parameter or regularization factor; The normalized magnetic source strength obtained in step S2 data; For weighted forward mapping matrices; Weighted matrix of data (superscript) (represents matrix transpose); These are the weighted model space parameters.

[0117] S32 employs a focused support functional stabilizer and deep weighting to construct a model weighting matrix;

[0118] By employing a focused support functional stabilizer and depth weighting, a clear imaging effect with well-defined property boundaries is achieved. A model weighting matrix is ​​introduced, and the data-fitted functional is as follows:

[0119] (8)

[0120] Where M is the total number of grid cells in the above underground subdivision; Weighted model vector The physical property value of the j-th grid cell in the equation; ò is a small threshold parameter to prevent the denominator from being zero and to control the focusing intensity. This functional can effectively sharpen the boundary of a geological body by minimizing the volume of a non-zero physical property anomaly.

[0121] S33, the transformation of weighted model parameters depends on the introduced model weighting matrix, and a deep weighting matrix is ​​constructed;

[0122] To overcome the inherent skin effect in potential field data inversion (i.e., the large physical property values ​​obtained from the inversion tend to be non-physically concentrated at the surface and do not match the actual deep geological conditions), the above-mentioned weighted model parameters are transformed ( It depends on the introduced model weighting matrix (i.e., the depth weighting matrix). :

[0123] (9)

[0124] In the formula, The center depth of the j-th model mesh; To prevent singular, tiny constants; To compensate for the field attenuation exponent of deep mesh sensitivity (taking the rapid attenuation characteristics of NSS parameters, we take...) ).

[0125] S34. Based on the model weighting matrix and the depth weighting matrix, construct the objective function for shallow inversion, and solve the objective function for shallow inversion to obtain the shallow three-dimensional magnetic structure.

[0126] After the objective functional is constructed, iterative solutions are obtained using a model-weighted conjugate gradient algorithm. The objective function consists of two parts: a data-fitting functional and a model-stabilizing functional, both of which are adjusted by regularization.

[0127] When the regularization factor is small, the effect of the model's stabilizing functional is weak, meaning the regularization effect is poor. This makes the inversion results more susceptible to data noise, and the results are usually unstable and inaccurate. Conversely, when the regularization factor is large, the data fitting term has a smaller effect, making it difficult to reduce data errors during the inversion process.

[0128] Further, in step S34, solving the objective function of the shallow inversion to obtain the shallow three-dimensional magnetic structure includes:

[0129] The objective function for the shallow inversion is iteratively optimized using the conjugate gradient method: in each iteration, the response of the shallow three-dimensional magnetic structure is calculated and continuously updated by comparing it with the observed normalized magnetic source intensity until the data fitting difference meets the set threshold, and finally the shallow three-dimensional magnetic structure is obtained.

[0130] To simplify the process and save computation time, this embodiment of the invention employs an adaptive regularization factor for inversion calculation. The first iteration of the steepest descent using the conjugate gradient method is in The regularization parameter is initially determined after the first iteration (Formula 10), and subsequent iterations adjust the regularization factor according to (Formula 11). Additionally, during the iteration process, due to numerical algorithm limitations, the solution space may contain negative values ​​or positive values ​​exceeding a reasonable range, requiring truncation of these negative and positive values.

[0131] (10)

[0132] , (11)

[0133] Among them, the conjugate gradient method is used to iteratively optimize the objective function: in each iteration, the current model is used as... The NSS response is calculated by comparing it with the observed d (calculated in step S2). The data is continuously updated after comparison until the data fitting difference meets the set threshold, ultimately obtaining the three-dimensional magnetic structure of the shallow subsurface. This shallow magnetic model serves as the prior constraint model for the multi-scale integrated inversion of shallow and deep magnetic fields in step S5.

[0134] S4. The magnetic anomaly modulus is calculated based on the three magnetic anomaly components of the aeromagnetic total field anomaly.

[0135] In this embodiment of the invention, the magnetic anomaly modulus It is another type of magnetic conversion quantity that is weakly sensitive to the magnetization direction. It is inversely proportional to the cube of the distance and of the same order of magnitude as the magnetic anomaly. It does not significantly amplify noise during the conversion process and can effectively preserve long-wavelength low-frequency information, thus providing a reliable data foundation for the inversion of deep magnetic structures.

[0136] Magnetic anomaly modulus Synthesized from the three components of the magnetic anomaly:

[0137] (12)

[0138] in, This represents the magnetic anomaly modulus. , , This indicates the components of the aeromagnetic total field anomaly in the X, Y, and Z directions.

[0139] S5. Using the shallow three-dimensional magnetic structure as a priori model, a comprehensive inversion objective function is constructed by incorporating the magnetic anomaly modulus in the form of a reference model deviation term. The comprehensive inversion objective function is solved to obtain the three-dimensional magnetic structure of the study area.

[0140] Inversion using a single normalized magnetic source intensity (NSS) can only yield shallow magnetic bodies and single... The inversion results suffer from shallow-deep coupling, making it difficult to distinguish magnetic bodies of different scales. This invention addresses the problem of shallow magnetic structures obtained by inverting the normalized magnetic source intensity obtained in step S3. As a prior model, it is incorporated in the form of a deviation term from the reference model. The objective function for inversion is used to construct the NSS-Ta comprehensive inversion:

[0141] (13)

[0142] This is magnetic anomaly modulus data; These are the model parameters to be solved in the inversion process; for Nonlinear forward operators, This is the shallow prior model obtained from the inversion in step S3; Weighted matrix of data, It is a depth-weighted matrix. This is the regularization factor (same as formula 10-11).

[0143] forward operator It can be represented as:

[0144] (14)

[0145] in, Indicates the first At each observation point Orthogonal response; , and These are the sensitivity matrices corresponding to the three directional components; , and They represent the first At each observation point, the model... The generated magnetic field components are in three directions. The squares and square roots in equation (14) are calculated point-by-point. The depth weighting relationship is determined by... Characterization, in which the depth-weighted index The value is 3, corresponding to The relationship between the cubic decay of distance and distance.

[0146] The key design of formula (13) lies in: the model regularization term in the traditional objective function. Replace with a penalty term that deviates from the prior model The magnetic materials that were clearly distinguished in the shallow NSS were... Anchoring, making The degrees of freedom for inversion are primarily granted to mid-to-deep long-wavelength signals, thereby avoiding a single... The mutual masking of shallow high-frequency and deep low-frequency information during inversion.

[0147] Taking the first derivative of the objective function (13) and setting it to zero, we can obtain the following optimality conditions:

[0148] (15)

[0149] in, for forward operator For model parameters The Jacobian matrix, due to It has a non-linear relationship with the model and needs to be updated in each iteration. :

[0150] (16)

[0151] During the iteration process, the inversion is started with a small initial model (usually 0.001), and the objective function (13) is solved using the conjugate gradient method.

[0152] Data weighting matrix Typically, a unit diagonal matrix is ​​used (Equation 9); regularization factor An adaptive regularization factor is used for inversion calculation (Equations 10 and 11).

[0153] The final inversion model It contains two types of information: high-frequency details are... Provided, medium-deep long-wavelength structures are made of Data-driven inversion is provided.

[0154] The comprehensive inversion results are the multi-scale three-dimensional magnetic structure of the study area, which can be used for the identification of magnetic units of different tectonic units (such as orogenic belts, basin basements, ocean-continent transition zones, etc.), the statistical analysis of magnetic layer thickness, the interpretation of magnetic stripes and geological bodies such as seamounts, and the analysis of deep geological structures by combining Curie data, heat flow data, etc.

[0155] This invention proposes a comprehensive inversion method suitable for fine analysis of the three-dimensional magnetic structure of the Earth's crust, namely the NSS-Ta comprehensive inversion imaging method. It innovatively uses the depth-weighted focusing inversion results of NSS as a priori model and incorporates them into the Ta inversion objective function in the form of a "reference model deviation term" to achieve multi-scale inversion and integrated reconstruction of shallow high-frequency information (anchored by NSS) and deep low-frequency information (driven by Ta data).

[0156] The integrated inversion method fully leverages the complementary advantages of NSS (slightly sensitive to magnetization direction and able to distinguish shallow parts) and Ta (retaining deep low-frequency signals), overcoming the bottleneck that single-conversion inversion cannot characterize multi-scale magnetic bodies. It can be widely applied to the fine characterization of three-dimensional magnetic structures in basin basements, orogenic belts, and marine areas, providing technical support for deep mineral resource exploration, geothermal research, and structural analysis.

[0157] Figure 2 A schematic diagram of a comprehensive inversion system for fine analysis of the three-dimensional magnetic structure of the Earth's crust, provided in an embodiment of the present invention, is shown below. Figure 2 As shown, the system includes:

[0158] The acquisition module 210 is used to acquire regional aeromagnetic data of the study area and fuse the regional aeromagnetic data to form a total aeromagnetic field anomaly covering the study area.

[0159] The normalized magnetic source strength module 220 is used to construct a magnetic gradient tensor based on the three magnetic anomaly components of the aeromagnetic total field anomaly and calculate the normalized magnetic source strength.

[0160] The weighted inversion module 230 is used to perform weighted focused three-dimensional inversion based on the normalized magnetic source intensity to obtain the shallow three-dimensional magnetic structure of the study area.

[0161] The magnetic anomaly modulus module 240 is used to calculate the magnetic anomaly modulus based on the three magnetic anomaly components of the aeromagnetic total field anomaly.

[0162] The comprehensive inversion module 250 is used to construct a comprehensive inversion objective function by using the shallow three-dimensional magnetic structure as a prior model and incorporating the magnetic anomaly modulus in the form of a reference model deviation term, and to solve the comprehensive inversion objective function to obtain the three-dimensional magnetic structure of the study area.

[0163] This embodiment is a system embodiment corresponding to the above method embodiment. Its specific implementation process is the same as that of the above method embodiment, and this system embodiment does not make any specific limitations.

[0164] The modules in the comprehensive inversion system applicable to the fine analysis of the three-dimensional magnetic structure of the Earth's crust described above can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware within or independently of the processor in a computer device, or stored in software within the memory of a computer device, so that the processor can call and execute the corresponding operations of each module.

[0165] This invention provides a computer device, which may be a server. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a computer storage medium and internal memory. The computer storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the computer storage medium. The database stores data generated or acquired during the execution of a comprehensive inversion method applicable to the three-dimensional magnetic structure of the Earth's crust, such as regional aeromagnetic data and three magnetic anomaly components. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a comprehensive inversion method applicable to the three-dimensional magnetic structure of the Earth's crust.

[0166] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a comprehensive inversion method for three-dimensional magnetic structures of the Earth's crust as described in the above embodiment. Alternatively, when the processor executes the computer program, it implements the functions of various modules / units in this embodiment of a comprehensive inversion system for three-dimensional magnetic structures of the Earth's crust.

[0167] In one embodiment, a computer storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the steps of the comprehensive inversion method for three-dimensional magnetic structures of the Earth's crust described in the above embodiment. Alternatively, when executed by a processor, the computer program implements the functions of each module / unit in the comprehensive inversion system for three-dimensional magnetic structures of the Earth's crust described in this embodiment.

[0168] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0169] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0170] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A comprehensive inversion method suitable for fine analysis of the three-dimensional magnetic structure of the Earth's crust, characterized in that, include: S1, acquire regional aeromagnetic data of the study area, and fuse the regional aeromagnetic data to form a total aeromagnetic field anomaly covering the study area; S2, Based on the three magnetic anomaly components of the aeromagnetic total field anomaly, construct the magnetic gradient tensor and calculate the normalized magnetic source intensity; S3. Based on the normalized magnetic source intensity, a weighted focusing three-dimensional inversion is performed to obtain the shallow three-dimensional magnetic structure of the study area; S4. The magnetic anomaly modulus is calculated based on the three magnetic anomaly components of the aeromagnetic total field anomaly. S5. Using the shallow three-dimensional magnetic structure as a priori model, a comprehensive inversion objective function is constructed by incorporating the magnetic anomaly modulus in the form of a reference model deviation term. The comprehensive inversion objective function is solved to obtain the three-dimensional magnetic structure of the study area.

2. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 1, characterized in that, Step S1 includes: S11, For the initial aeromagnetic data of each measurement area in the study area, the low-altitude aeromagnetic data in the initial aeromagnetic data is unified to the reference altitude through the frequency domain upward extension operator to obtain the upwardly extended aeromagnetic data; S12. Based on the aeromagnetic data after upward extension of the overlapping zone in adjacent survey areas, a univariate linear regression model is constructed to solve the offset coefficient and the proportional coefficient, and the benchmark correction transformation formula between adjacent survey areas is obtained. S13, using the reference correction transformation formula, perform reference correction on the aeromagnetic data extended upward in the adjacent survey areas to obtain the reference-corrected aeromagnetic data for each survey area. S14. By interpolation, the scale of the aeromagnetic data after benchmark correction for each survey area is unified to obtain the aeromagnetic data after unified scale. S15. By using a stitching method or a hybrid method, the aeromagnetic data of each survey area at a unified scale are automatically stitched together to obtain the regional aeromagnetic data of the study area.

3. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 1, characterized in that, Step S2 includes: S21, convert the aeromagnetic total field anomaly into three magnetic anomaly components; S22, calculate the spatial partial derivatives of the three magnetic anomaly components within the frequency range, construct the second-order gradient tensor, and obtain the tensor matrix; S23, by performing eigenvalue decomposition on the tensor matrix, three eigenvalues ​​are obtained, and the normalized magnetic source intensity is calculated based on the three eigenvalues.

4. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 1, characterized in that, Step S3 includes: S31. Based on the basic theory of Tikhonov regularization inversion, a model weighting function is constructed, and the subsurface medium below the surface of the study area is divided into M uniform orthogonal subsurface grid units in three-dimensional space. The continuous magnetic susceptibility in each grid is discretized and expressed, thereby constructing a model space of length M, where M is a positive integer. S32 employs a focused support functional stabilizer and deep weighting to construct a model weighting matrix; S33, the transformation of weighted model parameters depends on the introduced model weighting matrix, and a deep weighting matrix is ​​constructed; S34. Based on the model weighting matrix and the depth weighting matrix, construct the objective function for shallow inversion, and solve the objective function for shallow inversion to obtain the shallow three-dimensional magnetic structure.

5. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 4, characterized in that, In step S34, solving the objective function of the shallow inversion to obtain the shallow three-dimensional magnetic structure includes: The objective function for the shallow inversion is iteratively optimized using the conjugate gradient method: in each iteration, the response of the shallow three-dimensional magnetic structure is calculated and continuously updated by comparing it with the observed normalized magnetic source intensity until the data fitting difference meets the set threshold, and finally the shallow three-dimensional magnetic structure is obtained.

6. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 1, characterized in that, Step S4 is achieved through the following formula: ； in, This represents the magnetic anomaly modulus. , , This indicates the components of the aeromagnetic total field anomaly in the X, Y, and Z directions.

7. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 1, characterized in that, The calculation method for the comprehensive inversion objective function in step S5 is as follows: ； ； in, Denotes the comprehensive inversion objective function, This represents the model parameters to be solved. The nonlinear forward modeling operator representing the magnetic anomaly modulus, This refers to the shallow three-dimensional magnetic structure. Represents a data weighting matrix. Represents the depth-weighted matrix. Represents the regularization factor. This represents the magnetic anomaly modulus; Indicates the first The forward modeling response of the magnetic anomaly modulus at each observation point , and These are the sensitivity matrices corresponding to the X, Y, and Z directions, respectively. , and They represent the first At each observation point, the model... The generated magnetic field components in three directions It is a positive integer.

8. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 7, characterized in that, In step S5, solving the comprehensive inversion objective function includes the following steps: The first derivative of the comprehensive inversion objective function is calculated and set to zero. The inversion is started with a preset initial value during the iteration process. The conjugate gradient method is used to solve the comprehensive inversion objective function, and finally the three-dimensional magnetic structure of the inversion model is obtained. The data weighting matrix is ​​a unit diagonal matrix, and the regularization factor is an adaptive regularization factor used for inversion calculation.

9. The comprehensive inversion method for fine analysis of the three-dimensional magnetic structure of the Earth's crust according to claim 8, characterized in that, The regularization factor is calculated using an adaptive regularization factor, including: The first iteration of the steepest descent of the conjugate gradient method in Under the circumstances, in which, This represents the initial value of the regularization factor; The initial value of the regularization parameter is determined after the first iteration according to the following formula: ； in, The regularization factor during the first iteration. This represents the current model during the first iteration. This represents the normalized magnetic source strength. This represents the model weighting matrix during the first iteration; The regularization factor is adjusted in later iterations according to the following formula: ， ； in, Indicates the first The regularization factor at the next iteration Indicates the first The regularization factor at the next iteration This represents the iteration factor.

Images