Aeromagnetic data-based interior high-precision inversion method

Abstract

The invention provides an aeromagnetic data-based interior high-precision inversion method, and the method comprises the steps: carrying out the preprocessing of the aeromagnetic data of a target region, eliminating the noise and interference, and extracting the region magnetic anomaly information related to a deep interior; constructing a remanence intensity model based on a depth attenuation function, and obtaining the remanence intensity of each grid point of the target area according to the target area and the remanence intensity model; based on the residual magnetism intensity model, establishing a non-linear attenuation model of the residual magnetism intensity along with the depth, and obtaining the magnetic anomaly intensity generated by each depth layer in the target area on the earth surface by using the non-linear attenuation model; and inputting the regional magnetic anomaly information, the residual magnetism probability parameter, the geological constraint and the magnetic anomaly intensity into the interior inversion model to obtain an interior depth prediction value of the target region. According to the method, the influence of residual magnetism on the magnetization direction is considered, and the precision of interior inversion is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geophysical technology, and particularly relates to a high-precision inversion method of Curie surface based on aeromagnetic data. BACKGROUND

[0002] The Curie surface refers to the temperature interface at which the ferromagnetic minerals in the crust are converted into paramagnetic substances when the temperature rises to the Curie point. As the bottom interface of the magnetic layer in magnetic prospecting, the Curie surface can effectively characterize the distribution characteristics of the underground temperature field, and has important scientific significance and practical application value in the fields of geothermal energy development, oil and gas resource prediction, earthquake and volcanic disaster prevention, and primary hydrothermal mineral exploration. Through aeromagnetic data, regional magnetic anomaly information caused by the magnetic basement can be extracted, and the depth distribution of the Curie surface can be accurately calculated by using the inversion method, thereby providing a key basis for related geological research and resource exploration.

[0003] The existing Curie surface inversion method is usually based on the assumption that there is no remanence in the study area, that is, the effective magnetization inclination, declination and other parameters are directly set to the parameter values of the present geomagnetic field in the inversion process. However, in areas with complex geological history and significant remanence influence, this assumption will ignore the significant contribution of remanence to the magnetization direction, resulting in a large deviation of the Curie surface inversion result, thereby reducing the inversion accuracy and reliability, and being difficult to meet the needs of high-precision resource exploration and geological research.

[0004] In view of the above problems, the present application provides a high-precision inversion method of Curie surface based on aeromagnetic data suitable for the case where remanence exists. SUMMARY

[0005] The present application provides a high-precision inversion method of Curie surface based on aeromagnetic data, which mainly aims to consider the influence of remanence on the inversion of Curie surface and effectively improve the inversion accuracy of Curie surface.

[0006] In a first aspect, the present application provides a high-precision inversion method of Curie surface based on aeromagnetic data, comprising:

[0007] Pretreating the aeromagnetic data of the target area to eliminate noise and interference, and extracting regional magnetic anomaly information related to the deep Curie surface;

[0008] Constructing a remanence intensity model based on a depth decay function, and obtaining the remanence intensity of each grid point in the target area according to the target area and the remanence intensity model;

[0009] Based on the remanence intensity model, a nonlinear decay model of remanence intensity with depth is established, and the magnetic anomaly intensity generated at the surface by each depth layer in the target area is obtained by using the nonlinear decay model;

[0010] input the regional magnetic anomaly information, the remanent magnetization probability parameter, the geological constraint and the magnetic anomaly strength into the crustal depth inversion model to obtain a crustal depth prediction value of the target region, wherein the remanent magnetization probability parameter is a mean value and a variance of a total distribution field composed of remanent magnetization strengths of each grid point in the target region;

[0011] The crustal depth inversion model extracts a magnetic anomaly spatial pattern by using a residual convolution module, and fuses the remanent magnetization probability parameter and the geological constraint by using a parallel full connection branch, and the geological constraint includes spatial distribution of different rock types, fault zone distribution, fault position and fracture zone range.

[0012] Further, the crustal depth inversion model is obtained by training through the following steps:

[0013] Based on the remanent magnetization strength model, random sampling of remanent magnetization strength and direction combination generates magnetic anomaly strength simulation data;

[0014] The magnetic anomaly strength simulation data and the collected magnetic anomaly strength real data are added to a training data set, and the crustal depth inversion model is trained by using the training data set;

[0015] The crustal depth prediction value output by the crustal depth inversion model is converted into a theoretical magnetic anomaly by using a differentiable physical layer, a joint loss function is constructed with a crustal depth measured value, a magnetic anomaly-depth physical relationship is forced to be satisfied, and a trained crustal depth inversion model is obtained.

[0016] Further, the calculation formula of the joint loss function is as follows:

[0017] L=L data +λL laplace ;

[0018]

[0019]

[0020] Wherein, L represents the joint loss function, λ represents a weight parameter, N is the number of measuring points, represents a crustal depth prediction value of the i-th measuring point, represents a crustal depth theoretical value of the i-th measuring point, M represents a total number of magnetic potential points participating in the calculation of the joint loss function, represents a direction derivative, i.e. gradient, U j represents a magnetic potential of the j-th point, i and j are both positive integers.

[0021] Further, the aeromagnetic data of the target region is preprocessed to eliminate noise and interference, and regional magnetic anomaly information related to deep crustal depth is extracted, and the steps include:

[0022] Using geomagnetic diurnal variation correction and altitude correction to weaken external magnetic field interference of the aeromagnetic data;

[0023] Wavelet transform and principal component analysis are used to eliminate high-frequency noise in the aeromagnetic data while retaining low-frequency magnetic anomaly components dominated by deep Curie surface areas, ultimately obtaining the regional magnetic anomaly information.

[0024] Furthermore, the calculation formula of the residual magnetism intensity model is as follows:

[0025] R(z i )=R0·f(z i );

[0026]

[0027] Among them, R(z i ) represents the depth value z i The residual magnetic intensity at z i Represents the depth value of the i-th group, R0 represents the residual magnetic intensity at z = 0 on the surface, α is the attenuation coefficient, β is the nonlinear index, and i is a positive integer. This formula represents the underground depth z i The residual magnetic intensity R(z i ) can be obtained by multiplying the remanent magnetic intensity R0 at the surface by the depth attenuation function.

[0028] Furthermore, the calculation formula of the nonlinear attenuation model is as follows:

[0029]

[0030] Among them, ΔB(z i ) represents the depth value z i The ratio of the magnetic anomaly intensity generated at the surface to the remanent magnetic intensity, z i represents the depth value of the i-th group, k is the proportional coefficient, R(z i ) represents the depth value z i The residual magnetic intensity at the depth z is i The ratio of the surface magnetic anomaly intensity to the remanent magnetic intensity ΔB(z i ), the proportional coefficient k, the residual magnetic intensity R(z i ) and depth z i The square of ΔB(z i ) and the residual magnetic intensity R(z i ) is proportional to the depth z i It is inversely proportional to the square of , representing the quantitative influence of depth and remanent magnetization intensity on the surface magnetic anomaly intensity and remanent magnetization intensity ratio.

[0031] Further, the regional magnetic anomaly information, the residual magnetization probability parameter, the geological constraint and the magnetic anomaly intensity are input into the intracrustal inversion model to obtain an intracrustal depth prediction value of the target region, and the step comprises:

[0032] The regional magnetic anomaly information and the residual magnetization probability parameter are taken as additional feature vectors.

[0033] The geological constraint is converted into a two-dimensional network matrix.

[0034] The additional feature vectors, the two-dimensional network matrix and the magnetic anomaly intensity are input into the intracrustal inversion model to obtain the intracrustal depth prediction value.

[0035] In a second aspect, an embodiment of the present application provides an intracrustal high-precision inversion system based on aeromagnetic data, comprising:

[0036] A preprocessing module is configured to preprocess aeromagnetic data of a target region, eliminate noise and interference, and extract regional magnetic anomaly information related to deep intracrustal.

[0037] A residual magnetization module is configured to construct a residual magnetization intensity model based on a depth attenuation function, and obtain residual magnetization intensity of each grid point in the target region according to the target region and the residual magnetization intensity model.

[0038] A magnetic anomaly module is configured to establish a nonlinear attenuation model of residual magnetization intensity with depth based on the residual magnetization intensity model, and obtain magnetic anomaly intensity generated by each depth layer in the target region on the ground surface by using the nonlinear attenuation model.

[0039] An inversion module is configured to input the regional magnetic anomaly information, the residual magnetization probability parameter, the geological constraint and the magnetic anomaly intensity into an intracrustal inversion model to obtain an intracrustal depth prediction value of the target region, wherein the residual magnetization probability parameter is a mean value and a variance of a total distribution field composed of residual magnetization intensity of each grid point in the target region.

[0040] The intracrustal inversion model uses a residual convolution module to extract a magnetic anomaly spatial pattern, and a parallel full connection branch to fuse the residual magnetization probability parameter and the geological constraint, and the geological constraint includes spatial distribution of different rock types, distribution of fault zones, fault positions and ranges of fracture zones.

[0041] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned intracrustal high-precision inversion method based on aeromagnetic data when executing the computer program.

[0042] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program is executed by a processor to implement the steps of the above-mentioned high-precision inversion method of intracrustal Curie surface based on aeromagnetic data.

[0043] The high-precision inversion method of intracrustal Curie surface based on aeromagnetic data provided by the present application solves the signal mixing problem caused by the superposition of remanence and induced magnetism through preprocessing in the presence of remanence; and introduces random remanence disturbance with spatial correlation, establishes a nonlinear decay model of remanence strength with depth based on depth decay function, simulates the coupling effect of remanence at different depth horizons on the surface magnetic anomaly signal, and solves the problem that the effective magnetization inclination / declination estimation deviation is caused by ignoring the remanence component in the traditional method; in addition, the above steps can be used to construct a magnetic anomaly synthetic data set containing remanence-induced magnetic mixed effect, effectively enhance the generalization ability of the intracrustal inversion model to complex remanence interference mode, and through the space-depth double constraint mechanism, the remanence distortion which is difficult to separate in the traditional method is quantized into learnable physical features, providing the intracrustal inversion model with a magnetic field input closer to the real geological conditions, and improving the inversion precision of the intracrustal Curie surface. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 A flowchart of the high-precision inversion method of intracrustal Curie surface based on aeromagnetic data provided by an embodiment of the present application is provided.

[0045] Figure 2 A structural schematic diagram of the high-precision inversion system of intracrustal Curie surface based on aeromagnetic data provided by an embodiment of the present application is provided.

[0046] Figure 3 A structural schematic diagram of the computer device provided by an embodiment of the present application is provided.

[0047] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0048] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference signs represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as a limitation of the present application.

[0049] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings of the embodiments of the present application, so that those skilled in the art can better understand the solutions of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0050] In the embodiments of the present application, at least one refers to one or more; multiple refers to two or more than two. In the description of the present application, the terms "first", "second", "third" and the like are only used for distinguishing the purposes of description, and cannot be understood as indicating or implying relative importance, nor can be understood as indicating or implying order. In addition, the terms "first", "second" are only used for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more than two, unless otherwise specifically limited.

[0051] In the description of the present application, the terms "including", "containing", "having" and their variants mean "including but not limited to", unless otherwise specifically emphasized.

[0052] Figure 1 A flow chart of a high-precision inversion method for the earth's interior based on aeromagnetic data is provided in the embodiments of the present application, as shown in Figure 1 The method comprises the following steps:

[0053] S10, pre-processing the aeromagnetic data of the target area to eliminate noise and interference, and extracting regional magnetic anomaly information related to the deep earth's interior;

[0054] The step S10 comprises:

[0055] S11, using geomagnetic diurnal variation correction and height correction to weaken the external magnetic field interference of the aeromagnetic data;

[0056] S12, using wavelet transform and principal component analysis to eliminate high-frequency noise of the aeromagnetic data, while retaining low-frequency magnetic anomaly components dominated by the deep earth's interior, and finally obtaining the regional magnetic anomaly information.

[0057] In the presence of residual magnetism, the aeromagnetic data preprocessing step needs to focus on solving the signal mixing problem caused by the superposition of residual magnetism and magnetic induction. First, the geomagnetic diurnal variation correction and height correction are used to weaken the external magnetic field interference in the aeromagnetic data. Then, the wavelet transform and principal component analysis are used to eliminate the high-frequency noise such as aircraft interference and surface magnetic body noise, while retaining the low-frequency magnetic anomaly components dominated by the deep subsurface, ensuring that the regional magnetic anomaly information obtained can reflect the shape of the deep subsurface and retain the contribution characteristics of the residual magnetism effect, providing a high-quality data set for the training of the subsequent subsurface inversion model.

[0058] S20, constructing a residual magnetism intensity model based on a depth attenuation function, and obtaining residual magnetism intensities of each grid point in the target region according to the target region and the residual magnetism intensity model;

[0059] In the traditional method, it is assumed that the magnetization direction is determined only by the present geomagnetic field, and the residual magnetism component is ignored, which will cause the effective magnetization inclination / declination estimation deviation. In the embodiment of the present application, a random residual magnetism disturbance with spatial correlation is introduced in the magnetic layer, and a nonlinear decay model of residual magnetism intensity with depth is established based on the depth attenuation function, so as to simulate the coupling effect of residual magnetism at different depth horizons on the surface magnetic anomaly signal.

[0060] The specific inputs for constructing a random residual magnetism intensity model with spatial correlation in the study area are as follows:

[0061] The study area is divided into a spatial grid, where the grid coordinates are (x i ,y j ), the number of grid points is MxN, and the point distance is Δx and Δy; and the spatial correlation parameters are set, where the horizontal correlation length is L x and L y , by setting the horizontal correlation length, the range of spatial correlation can be controlled; and the statistical parameters are set, where the disturbance mean is μ and the variance is σ 2 .

[0062] The output of the residual magnetism intensity model is as follows: the residual magnetism disturbance value ΔM(x i ,y j ) of each grid point, which satisfies the random distribution of spatial correlation.

[0063] The calculation steps of the residual magnetism intensity model are as follows:

[0064] 1. Generate white noise, generate white noise field W(x i ,y j ) ~ N(μ,σ 2 ) obeying normal distribution, without spatial correlation.

[0065] 2. Introduce spatial correlation:

[0066] First, calculate the covariance of any two points

[0067] wherein denotes the covariance of two points, σ 2 is the variance, reflecting the overall fluctuation degree; d x =|x1-x2|, d y =|y1-y2| is the distance of two points in x, y direction, L x , L y is the characteristic length, determining the influence degree of distance on covariance. The formula shows that the covariance of two points exponentially decays with the increase of their distance in x, y direction, that is, the points closer in space have stronger correlation, which reflects the influence of spatial position on the correlation of magnetic characteristics, and the magnetic characteristics of points closer are more similar. For each grid point (x i ,y j ), the residual magnetization disturbance value can be obtained by weighted average of adjacent points:

[0068]

[0069] Finally, the calculation formula of the obtained residual magnetization intensity model is as follows:

[0070] R(z i )=R0·f(z i );

[0071]

[0072] wherein, R(z i ) represents the residual magnetization intensity at the depth value z i , z i represents the i-th group of depth values, R0 represents the residual magnetization intensity at the ground surface z=0, α is the attenuation coefficient, β is the nonlinear index, and i is a positive integer. The formula shows that the residual magnetization intensity R(z i at the underground depth z i can be obtained by multiplying the residual magnetization intensity R0 at the ground surface and the depth attenuation function.

[0073] S30, based on the residual magnetization intensity model, a nonlinear attenuation model of residual magnetization intensity with depth is established, and the magnetic anomaly intensity generated by each depth layer in the target area at the ground surface is obtained by using the nonlinear attenuation model;

[0074] In the embodiment of the application, the nonlinear attenuation model of residual magnetization intensity with depth based on the residual magnetization intensity model of depth attenuation function is specifically input as follows:

[0075] Depth horizon: a group of depth values z1, z2, …, z N , wherein z=0 is the ground surface, and the direction downward is positive;

[0076] Surface remanence intensity: R0, remanence intensity at the surface z = 0;

[0077] Attenuation function parameter: nonlinear attenuation function Wherein, a is the attenuation coefficient, and β is the nonlinear index.

[0078] The magnetic anomaly intensity generated by each depth layer at the surface is proportional to the remanence intensity and inversely proportional to the square of the depth:

[0079]

[0080] Wherein, ΔB(z i ) represents the ratio of the magnetic anomaly intensity generated at the surface to the remanence intensity at the depth value z i , z i represents the i-th set of depth values, k is the proportional coefficient, and R(z i ) represents the remanence intensity at the depth value z i . The formula shows that the ratio ΔB(z i ) of the corresponding surface magnetic anomaly intensity to the remanence intensity at the depth z i is determined by the proportional coefficient k, the remanence intensity R(z i ) at the depth, and the square of the depth z i , that is, ΔB(z i ) is proportional to the remanence intensity R(z i ) and inversely proportional to the square of the depth z i , reflecting the quantitative influence relationship of the depth and the remanence intensity on the ratio of the surface magnetic anomaly intensity to the remanence intensity.

[0081] The above steps can be used to construct a magnetic anomaly synthesis data set containing remanence-magnetism mixed effect, effectively enhancing the generalization ability of the neural network to complex remanence interference mode. The design quantifies the remanence distortion which is difficult to separate in the traditional method into a learnable physical feature through a space-depth double constraint mechanism, and provides the neural network with a magnetic field input closer to the real geological conditions.

[0082] S40, input the regional magnetic anomaly information, the remanence probability parameter, the geological constraint and the magnetic anomaly intensity into the inside inversion model to obtain the inside depth prediction value of the target region, wherein the remanence probability parameter is the mean and variance of the total distribution field composed of the remanence intensity of each grid point in the target region;

[0083] The inside inversion model uses a residual convolution module to extract the magnetic anomaly spatial pattern, and a parallel full connection branch to fuse the remanence probability parameter and the geological constraint, wherein the geological constraint includes the spatial distribution of different rock types, the distribution of fault zones, the fault position and the range of the fracture zone.

[0084] wherein, the step S40 comprises:

[0085] S41, taking the regional magnetic anomaly information and the residual magnetism probability parameter as an additional feature vector;

[0086] For the problem of nonlinear mapping caused by residual magnetism and magnetic coupling, a hybrid neural network architecture embedded with physical information is designed as the inversion model. In addition to the regional magnetic anomaly information, the residual magnetism probability parameter is introduced as an additional feature vector at the input end. The residual magnetism distribution field is mainly composed of the total distribution field of the residual magnetism intensity R(z i ) of each grid point generated by the above steps. The residual magnetism probability parameter is a global statistical quantity (mean, variance) of the distribution field, which is used for the input of the residual magnetism average intensity and deviation degree in the subsequent inversion model.

[0087] S42, converting the geological constraints into a two-dimensional network matrix;

[0088] The residual convolution module is used in the inversion model to extract the spatial pattern of the magnetic anomaly, and the residual magnetism probability parameter (residual magnetism mean, variance) and the geological constraint are fused in parallel through the full connection branch. The geological constraint refers to the spatial distribution of different rock types (such as granite, basalt), and the fault zone distribution refers to the position of the fault and the range of the fracture zone, which reflects the influence of the geological structure on the residual magnetism characteristics.

[0089] Before inputting the inversion model, the above-mentioned geological constraints are converted into a two-dimensional network matrix as a spatial mask, for example, the fault constraint has a fault zone of 1 and no fault zone of 2, and the rock constraint has granite of 1 and basalt of 2. The converted mask matrix is input as a separate channel together with the regional magnetic anomaly information, for example, the regional magnetic anomaly information is channel one and the mask matrix is channel two.

[0090] S43, inputting the additional feature vector, the two-dimensional network matrix and the magnetic anomaly intensity into the inversion model to obtain the inversion depth prediction value.

[0091] The above-mentioned additional feature vector, two-dimensional network matrix and magnetic anomaly intensity are input into the inversion model to obtain the inversion depth prediction value. Through the differentiable physical layer (Laplace equation, etc.), the predicted depth is converted into a theoretical magnetic anomaly, and a joint loss function is constructed with the measured data to force the network to satisfy the magnetic anomaly-depth physical relationship, thereby improving the accuracy of the inversion depth prediction value.

[0092] The inversion model in the embodiment is a neural network model, which needs to be trained before application. Specifically, the inversion model is trained by the following steps:

[0093] Based on the residual intensity model, random sampling residual intensity, direction combination generates magnetic anomaly intensity simulation data;

[0094] The magnetic anomaly intensity simulation data and the collected magnetic isomerization intensity real data are added to a training data set, and the intracrustal inversion model is trained by using the training data set;

[0095] The intracrustal depth prediction value output by the intracrustal inversion model is converted into a theoretical magnetic anomaly through a differentiable physical layer, and a joint loss function is constructed with the intracrustal depth measured value, so that the magnetic anomaly-depth physical relationship is forced to be satisfied, and the trained intracrustal inversion model is obtained.

[0096] In the embodiment of the application, the magnetic anomaly intensity simulation data generated by the residual intensity model and the measured and collected magnetic isomerization intensity real data are used to form a training data set, so that the generalization problem caused by the spatial heterogeneity of residual parameters is solved; and the Monte Carlo Dropout strategy is introduced in the training, and the standard deviation of the intracrustal depth prediction value is calculated through multiple random forward propagation during reasoning, so as to quantify the influence of the uncertainty of the residual parameters on the inversion result. At the same time, an adaptive weighted loss function is used to apply higher weight to the measured data area, so as to balance the theoretical completeness of the synthetic data and the local reality of the measured data, thereby improving the training accuracy of the intracrustal inversion model.

[0097] (1) First, calculate the ordinary loss function:

[0098]

[0099] Among them, L data is the ordinary loss value (MSE), represents the intracrustal depth prediction value of the i-th measuring point, represents the intracrustal depth theoretical value of the i-th measuring point, N is the number of measuring points, and i is a positive integer.

[0100] (2) Then, the physical information constraint loss is used:

[0101] The predicted value is forced to satisfy the Laplace equation At this time, the sum of squares of the Laplace equation residuals of all grid points in the calculation area is calculated:

[0102]

[0103] Among them, Llaplace represents the physical information constraint loss, M represents the total number of magnetic potential points participating in the calculation of the joint loss function, represents the directional derivative, i.e. gradient, U j represents the magnetic potential of the j-th point, and j is a positive integer.

[0104] (3) combine the above (1) and (2) to obtain a joint loss function.

[0105] Combine the two, drive the network to fit the data and meet the physical law L=L data +λL laplace , λ is a weight parameter, used to adjust the strength of the physical constraint. This design deeply couples the residual magnetism model with data-driven learning, breaking through the limitation of traditional pure data-driven models lacking physical interpretation.

[0106] The high-precision inversion method for the Curie point based on the aeromagnetic data provided by the application solves the signal mixing problem caused by the superposition of residual magnetism and magnetic induction in the presence of residual magnetism through preprocessing; and introduces a random residual magnetism disturbance with spatial correlation, establishes a nonlinear decay model of residual magnetism strength with depth based on a deep decay function, simulates the coupling effect of residual magnetism at different depth horizons on the surface magnetic anomaly signal, solves the problem that the effective magnetization inclination / declination estimation deviation is caused by ignoring the residual magnetism component in the traditional method; in addition, the above steps can be used to construct a magnetic anomaly synthetic data set containing residual magnetism-magnetic induction mixed effects, effectively enhance the generalization ability of the Curie point inversion model to complex residual magnetism interference patterns, and through the space-depth double constraint mechanism, the residual magnetism distortion which is difficult to separate in the traditional method is quantized into learnable physical features, providing a more realistic geological condition magnetic field input for the Curie point inversion model, and improving the Curie point inversion precision.

[0107] Figure 2 The structure diagram of the high-precision inversion system for the Curie point based on the aeromagnetic data provided by the embodiment of the application is as shown in Figure 2 The system comprises a preprocessing module 210, a residual magnetism module 220, a magnetic anomaly module 230 and an inversion module 240, wherein:

[0108] The preprocessing module 210 is used for preprocessing the aeromagnetic data of the target area, eliminating noise and interference, and extracting regional magnetic anomaly information related to the deep Curie point;

[0109] The residual magnetism module 220 is used for constructing a residual magnetism strength model based on a depth decay function, and obtaining the residual magnetism strength of each grid point in the target area according to the target area and the residual magnetism strength model;

[0110] The magnetic anomaly module 230 is used for establishing a nonlinear decay model of residual magnetism strength with depth based on the residual magnetism strength model, and obtaining the magnetic anomaly strength generated by each depth layer in the target area on the ground surface by using the nonlinear decay model;

[0111] an inversion module 240 for inputting the regional magnetic anomaly information, the remanent magnetization probability parameters, geological constraints, and the magnetic anomaly intensity into a Curie point inversion model to obtain a predicted Curie point depth of the target area, wherein the remanent magnetization probability parameters are the mean and variance of a total distribution field composed of the remanent magnetization intensity of each grid point in the target area;

[0112] The Curie point inversion model uses a residual convolution module to extract the spatial pattern of magnetic anomalies, and a parallel fully connected branch to fuse the remanent magnetization probability parameters and geological constraints. The geological constraints include the spatial distribution of different rock types, the distribution of fault zones, and the location and extent of fracture zones.

[0113] 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. For details, please refer to the above method embodiment. This system embodiment will not go into details.

[0114] Each module in the aforementioned high-precision Curie inversion system based on aeromagnetic data can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0115] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a computer storage medium and an internal memory. The computer storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the computer storage medium. The database of the computer device is used to store data generated or obtained during the execution of a high-precision Curie point inversion method based on aeromagnetic data, such as aeromagnetic data and a residual magnetic intensity model. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a high-precision Curie point inversion method based on aeromagnetic data is implemented.

[0116] 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, the processor implementing the steps of the method for high-precision inversion of the interior of the earth based on aeromagnetic data in one of the above embodiments when executing the computer program. Alternatively, the processor implements the functions of each module / unit in the embodiment of the system for high-precision inversion of the interior of the earth based on aeromagnetic data when executing the computer program.

[0117] In one embodiment, a computer storage medium is provided, the computer storage medium storing a computer program, the computer program being executable on a processor to implement the steps of the method for high-precision inversion of the interior of the earth based on aeromagnetic data in one of the above embodiments. Alternatively, the computer program is executable on the processor to implement the functions of each module / unit in the above embodiment of the system for high-precision inversion of the interior of the earth based on aeromagnetic data.

[0118] A person of ordinary skill in the art can understand that all or part of the processes in the above embodiments can be completed by a computer program instructing related hardware, and the computer program can be stored in a non-volatile computer readable storage medium and can include the processes of the above embodiments when executed.

[0119] A person of ordinary skill in the art can understand that all or part of the processes in the above embodiments can be completed by a computer program instructing related hardware, and the computer program can be stored in a non-volatile computer readable storage medium and can include the processes of the above embodiments when executed.

[0120] The above-described embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A high-precision Curie point inversion method based on aeromagnetic data, characterized in that: include: Preprocess the aeromagnetic data of the target area to eliminate noise and interference and extract regional magnetic anomaly information related to deep Curie surface areas; Constructing a remanent magnetization intensity model based on a depth attenuation function, and obtaining the remanent magnetization intensity of each grid point in the target area according to the target area and the remanent magnetization intensity model; Based on the residual magnetic intensity model, a nonlinear attenuation model of the residual magnetic intensity with depth is established, and the nonlinear attenuation model is used to obtain the magnetic anomaly intensity generated on the surface of each depth layer in the target area; Inputting the regional magnetic anomaly information, the remanent magnetization probability parameters, geological constraints, and the magnetic anomaly intensity into a Curie point inversion model to obtain a predicted Curie point depth of the target area, wherein the remanent magnetization probability parameters are the mean and variance of the total distribution field, and the total distribution field is composed of the remanent magnetization intensity of each grid point in the target area; The Curie point inversion model uses a residual convolution module to extract the spatial pattern of magnetic anomalies, and a parallel fully connected branch to fuse the remanent magnetization probability parameters and geological constraints. The geological constraints include the spatial distribution of different rock types, the distribution of fault zones, and the location and extent of fracture zones.

2. The high-precision Curie point inversion method based on aeromagnetic data according to claim 1, characterized in that: The Curie point inversion model is trained by the following steps: Based on the residual magnetic intensity model, randomly sampling residual magnetic intensity and direction combinations to generate magnetic anomaly intensity simulation data; adding the simulated magnetic anomaly intensity data and the collected real magnetic anomaly intensity data into a training data set, and using the training data set to train the Curie point inversion model; The Curie point depth prediction value output by the Curie point inversion model is converted into a theoretical magnetic anomaly through a differentiable physical layer, and a joint loss function is constructed with the measured Curie point depth value to enforce the magnetic anomaly-depth physical relationship, thereby obtaining a trained Curie point inversion model.

3. The high-precision Curie point inversion method based on aeromagnetic data according to claim 2, characterized in that: The calculation formula of the joint loss function is as follows: L=L data +λL laplace ; Wherein, L represents the joint loss function, λ represents the weight parameter, and N is the number of measurement points. represents the predicted Curie depth of the i-th measuring point, represents the theoretical value of the Curie depth of the i-th measuring point, M represents the total number of magnetic potential points involved in the calculation of the joint loss function, Indicates the directional derivative, i.e., the gradient, U j represents the magnetic potential at point j, where i and j are both positive integers.

4. The high-precision Curie point inversion method based on aeromagnetic data according to claim 1, characterized in that: The aeromagnetic data of the target area is preprocessed to eliminate noise and interference and extract regional magnetic anomaly information related to the deep Curie surface, the steps comprising: Using geomagnetic diurnal variation correction and altitude correction to weaken external magnetic field interference of the aeromagnetic data; Wavelet transform and principal component analysis are used to eliminate high-frequency noise in the aeromagnetic data while retaining low-frequency magnetic anomaly components dominated by deep Curie surface areas, ultimately obtaining the regional magnetic anomaly information.

5. The high-precision Curie point inversion method based on aeromagnetic data according to claim 1, characterized in that: The calculation formula of the residual magnetism intensity model is as follows: R(z i )=R0·f(z i ); Among them, R(z i ) represents the depth value z i The residual magnetic intensity at z i represents the i-th group of depth values, R0 represents the residual magnetic intensity at z = 0 on the surface, α is the attenuation coefficient, β is the nonlinear exponent, and i is a positive integer.

6. The high-precision Curie point inversion method based on aeromagnetic data according to claim 1, characterized in that: The calculation formula of the nonlinear attenuation model is as follows: Among them, ΔB(z i ) represents the depth value z i The ratio of the magnetic anomaly intensity generated at the surface to the remanent magnetic intensity, z i represents the depth value of the i-th group, k is the proportional coefficient, R(z i ) represents the depth value z i The residual magnetic intensity at , i is a positive integer.

7. The high-precision Curie point inversion method based on aeromagnetic data according to claim 1, characterized in that: The step of inputting the regional magnetic anomaly information, the remanent magnetization probability parameter, the geological constraint and the magnetic anomaly intensity into a Curie point inversion model to obtain a predicted Curie point depth value of the target area comprises: Taking the regional magnetic anomaly information and the remanent magnetization probability parameter as additional feature vectors; converting the geological constraints into a two-dimensional network matrix; The additional eigenvector, the two-dimensional network matrix and the magnetic anomaly intensity are input into the Curie point inversion model to obtain the Curie point depth prediction value.

8. A high-precision Curie point inversion system based on aeromagnetic data, characterized in that: include: The preprocessing module is used to preprocess the aeromagnetic data of the target area, eliminate noise and interference, and extract regional magnetic anomaly information related to deep Curie surface areas; A residual magnetization module is used to construct a residual magnetization intensity model based on a depth attenuation function, and obtain the residual magnetization intensity of each grid point in the target area according to the target area and the residual magnetization intensity model; A magnetic anomaly module is used to establish a nonlinear attenuation model of residual magnetic intensity with depth based on the residual magnetic intensity model, and to obtain the magnetic anomaly intensity generated on the surface of each depth layer in the target area using the nonlinear attenuation model; an inversion module, configured to input the regional magnetic anomaly information, the remanent magnetization probability parameters, geological constraints, and the magnetic anomaly intensity into a Curie point inversion model to obtain a predicted Curie point depth of the target area, wherein the remanent magnetization probability parameters are the mean and variance of a total distribution field, and the total distribution field is composed of the remanent magnetization intensity of each grid point in the target area; The Curie point inversion model uses a residual convolution module to extract the spatial pattern of magnetic anomalies, and a parallel fully connected branch to fuse the remanent magnetization probability parameters and geological constraints. The geological constraints include the spatial distribution of different rock types, the distribution of fault zones, and the location and extent of fracture zones.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the Curie point high-precision inversion method based on aeromagnetic data as claimed in any one of claims 1 to 7 are implemented.

10. A computer storage medium storing a computer program, wherein: When the computer program is executed by a processor, the steps of the Curie point high-precision inversion method based on aeromagnetic data as claimed in any one of claims 1 to 7 are implemented.

Images