A Magnetic Interface Inversion Method and Device Based on Magnetic Anomaly Modulus

Abstract

The present invention discloses a magnetic interface inversion method and device based on magnetic anomaly modulus. The method includes: constructing a plurality of magnetic interfaces, setting a reference plane corresponding to the magnetic interfaces on the surface of the earth, and the magnetic interfaces undulating up and down around the reference plane; setting a random magnetic dip angle, a random magnetic declination and a magnetization intensity for the magnetic interfaces, calculating the simulated magnetic anomaly modulus of the magnetic interfaces at each observation point on the surface of the earth, and repeating the calculation to obtain the simulated magnetic anomaly modulus of all the magnetic interfaces at each observation point on the surface of the earth; using the simulated magnetic anomaly modulus of the magnetic interfaces in a neural network for magnetic interface inversion to predict the depth distribution of the Curie surface. The present invention effectively considers the influence of remanent magnetism, improves the accuracy of the inversion result, and can obtain relatively accurate interface depth information at the same time.

Description

Technical Field

[0001] The present invention belongs to the field of geospatial technology, and more specifically, relates to a magnetic interface inversion method and device based on magnetic anomaly modulus. Background Art

[0002] Using magnetic method data inversion can delineate the depth distribution of the Curie surface, and further apply it to geothermal resource assessment, crustal structure research, geodynamic analysis, etc. The magnetization intensity in the geomagnetic field consists of induced magnetization intensity and remanent magnetization intensity. The conventional Curie surface depth inversion assumes that the parameters (intensity, inclination, declination) of the geomagnetic field in the study area are all present-day geomagnetic field parameters, that is, only induced magnetism is considered, without the influence of remanent magnetism. However, in practical applications, remanent magnetism widely exists, which leads to the unpredictability of geomagnetic field parameters and enhances the unreliability of inversion results. The existing inversion methods face the following bottleneck problems when considering remanent magnetism: the direction and intensity of remanent magnetism are both unknown, making it difficult to accurately set the geomagnetic field parameters in the inversion, which greatly affects the reliability of the inversion results.

[0003] In view of this, overcoming the technical defects of the above-mentioned existing technologies is an urgent problem to be solved in this technical field. Summary of the Invention

[0004] In view of the above defects or improvement requirements of the existing technology, the present invention provides a magnetic interface inversion method and device based on magnetic anomaly modulus, aiming to solve the technical problem of inaccurate inversion results caused by the non-consideration of remanent magnetism in the current magnetic interface inversion method.

[0005] To achieve the above object, according to one aspect of the present invention, there is provided a magnetic interface inversion method based on magnetic anomaly modulus, the method comprising:

[0006] Construct a plurality of magnetic interfaces, set a reference plane corresponding to the magnetic interfaces on the ground surface, and the magnetic interfaces undulate up and down around the reference plane;

[0007] Set a random magnetic inclination, a random magnetic declination and a magnetization intensity for the magnetic interfaces, calculate the simulated magnetic anomaly modulus of the magnetic interfaces at each observation point on the ground surface, and repeat the calculation to obtain the simulated magnetic anomaly modulus of all the magnetic interfaces at each observation point on the ground surface;

[0008] Use the simulated magnetic anomaly modulus of the magnetic interfaces in a neural network for magnetic interface inversion to predict the depth distribution of the Curie surface.

[0009] As a further improvement and supplement to the above solution, the present invention further includes the following additional technical features.

[0010] Preferably, the method for calculating the simulated magnetic anomaly modulus of the magnetic interfaces at each observation point on the ground surface includes:

[0011] The part of the magnetic interface around the reference plane is evenly divided into a number of prisms. The magnetization intensity of the prisms at the observation point is projected onto three directions, and the magnetization intensity components of the prisms in the three directions at the observation point are obtained according to the random magnetic dip angle and the random magnetic declination;

[0012] According to the magnetization intensity components, the three-component magnetic anomaly response generated by the prisms at the observation point is calculated. At the observation point, the magnetic anomaly three-components of all the prisms in the same direction are added together, and the simulated magnetic anomaly modulus of all the prisms at the observation point is calculated;

[0013] By repeating the calculation, the simulated magnetic anomaly moduli of all the prisms at each observation point are obtained, that is, the simulated magnetic anomaly moduli of the magnetic interface at each observation point are obtained. By repeating the calculation, the simulated magnetic anomaly moduli of all the magnetic interfaces at each observation point are obtained.

[0014] Preferably, the method for repeating the calculation to obtain the simulated magnetic anomaly moduli of all the magnetic interfaces at each observation point on the ground surface includes:

[0015] The simulated magnetic anomaly moduli of the magnetic interface at each observation point on the ground surface are combined into a simulated magnetic anomaly modulus matrix in the order of the observation points, that is, the simulated magnetic anomaly modulus of the magnetic interface.

[0016] Preferably, the range of the random magnetic dip angle is -90° to 90°, and the range of the random magnetic declination is -180° to 180°.

[0017] Preferably, when projecting the magnetization intensity of the prisms onto three directions, the three directions include the north-south direction, the east-west direction, and the vertical direction.

[0018] Preferably, the calculation method for obtaining the magnetization intensity components of the prisms in the three directions at a certain observation point according to the random magnetic dip angle and the random magnetic declination is:

[0019]

[0020] where (M x , M y , M z ) are the components of the magnetization intensity M in the north-south direction, the east-west direction, and the vertical direction respectively, the random magnetic dip angle is I1, and the random magnetic declination is D1.

[0021] Preferably, the method for calculating the three-component magnetic anomaly response generated by the prisms at the observation point according to the magnetization intensity components includes:

[0022] The north-south component of the magnetic anomaly generated by the prism at the observation point is H ax :

[0023]

[0024] The east-west component of the magnetic anomaly generated by the prism at the observation point is H ay :

[0025]

[0026] The vertical component of the magnetic anomaly generated by the prism at the observation point is Z a :

[0027]

[0028] Where μ0 represents the magnetic permeability of vacuum, (ξ, η, ζ) are the coordinates of any point on the prism, and the distance r between the coordinates of any point on the prism and the coordinates (x, y, z) of the observation point is r = [(ξ - x) 2 +(η - y) 2 +(ζ - z) 2 1 / 2 .

[0029] Preferably, the method for calculating the simulated magnetic anomaly modulus of all the prisms at the observation point includes:

[0030] The simulated magnetic anomaly modulus of all the prisms at the coordinates (x, y, z) of the observation point is T a (x, y, z):

[0031]

[0032] Where H' ax is the sum of the north-south components of the magnetic anomalies of all the prisms at the observation point; H' ay is the sum of the east-west components of the magnetic anomalies of all the prisms at the observation point; Z' a is the sum of the vertical components of the magnetic anomalies of all the prisms at the observation point; Ta is the simulated magnetic anomaly modulus of all the prisms at the observation point, that is, the simulated magnetic anomaly modulus of the magnetic interface at the observation point;

[0033] Calculate the simulated magnetic anomaly modulus of the magnetic interface at each observation point in turn, so as to obtain the simulated magnetic anomaly modulus of the magnetic interface.

[0034] Preferably, the method for predicting the depth distribution of the Curie surface in the neural network using the simulated magnetic anomaly modulus of the magnetic interface for magnetic interface inversion includes: ​

[0035] Input the magnetic anomaly modulus data with the centered depth value of the input label for neural network training;

[0036] All simulated magnetic anomaly modulus and real magnetic anomaly modulus data are fused in the training set of the neural network.

[0037] According to another aspect of the present invention, a magnetic interface inversion device based on magnetic anomaly modulus is provided. The device includes:

[0038] One or more processors;

[0039] A storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the magnetic interface inversion method based on magnetic anomaly modulus as described above.

[0040] Generally speaking, compared with the prior art by the above technical solutions conceived by the present invention, the following beneficial effects are obtained:

[0041] 1. Effectively consider the influence of remanent magnetism and improve the accuracy of the inversion result;

[0042] 2. Obtain relatively accurate interface depth information. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly introduced below. Obviously, the following described drawings are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0044] Figure 1 It is a flowchart of the steps of a magnetic interface inversion method based on simulated magnetic anomaly modulus provided in Embodiment 1;

[0045] Figure 2 It is a schematic diagram of the magnetic interface provided in Embodiment 1;

[0046] Figure 3 It is a schematic diagram of a prism provided in Embodiment 1;

[0047] Figure 4 It is a schematic diagram of the neural network architecture provided in Embodiment 1;

[0048] Figure 5 It is a schematic diagram of a magnetic interface inversion device based on magnetic anomaly modulus provided in Embodiment 2. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] To make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0050] Embodiment 1

[0051] Embodiment 1 provides a magnetic interface inversion method based on the simulated magnetic anomaly modulus. The method includes the following steps, as Figure 1 shown:

[0052] S101: Construct a plurality of magnetic interfaces, set a reference plane corresponding to the magnetic interfaces on the ground surface, and the magnetic interfaces undulate up and down around the reference plane.

[0053] Before constructing a plurality of magnetic interfaces, fully collect the existing Curie depth data in the study area, preprocess the data, including operations such as cropping, filling, and coordinate transformation, and analyze the data characteristics to obtain the approximate range of the interface depth parameters and the true magnetic anomaly modulus actually measured on the ground surface. If there is no available data in the study area, search for existing literature similar or close to the study area, extract the relevant Curie depth range and magnetic parameters, and use them as references to assist in the construction of the interface model. Based on the results of the analyzed data, construct and generate complex and diverse interface models.

[0054] S201: Set a random magnetic dip angle, a random magnetic declination angle, and a magnetization intensity for the magnetic interfaces, calculate the simulated magnetic anomaly modulus of the magnetic interfaces at each observation point on the ground surface, and repeat the calculation to obtain the simulated magnetic anomaly modulus of all the magnetic interfaces.

[0055] A random magnetic dip angle and a random magnetic declination angle are given to each interface to simulate the influence of remanent magnetism, and then the simulated magnetic anomaly modulus T generated by it on the ground surface is calculated a . Specifically, the construction of the magnetic interface can be as Figure 2 shown. Assume that Z = 0 is the ground surface, and there is an interface underground. The average depth of the interface is h. Take Z = h as the reference plane, and the interface undulates up and down around the reference plane. A random magnetic dip angle I1 and a random magnetic declination angle D1 are given to the interface, and it is evenly divided into a number of prisms. Then the height of each prism is its corresponding Δh, and Δh is randomly set.

[0056] At this time, the forward calculation of the magnetic interface can be regarded as the combined calculation of multiple prisms, as Figure 2As shown, X is the north-south component, Y is the east-west component, and Z is the vertical component. Assume that there is a prism underground with side lengths (a, b, Δh), the coordinates of its upper left vertex are (x0, y0, z0), the random magnetic inclination is I1, and the random magnetic declination is D1. At this time, the magnetization M of the prism can be projected onto the three directions according to the following formula, (M x , M y , M z ) are the components of the magnetization M in the north-south direction, east-west direction, and vertical direction respectively. The magnetization is set according to the geomagnetic field data provided by the International Geomagnetic Field Model, and is represented by Formula 1:

[0057]

[0058] Assume that there is an observation point grid on the earth's surface, and the number of grid points is N. The following Formula 2 can represent the magnetization of the observation point grid:

[0059]

[0060] Each row in this matrix represents the magnetization vector of an observation point, and the magnetization of the i-th point is M i =(M xi , M yi , M zi ).

[0061] If there is an observation point (x, y, z) on the earth's surface at this time, the corresponding magnetization can be brought in, and the three components H ax , H ay and z a of the magnetic anomaly response generated by the prism at the surface measurement point can be calculated according to Formulas 3, 4, and 5 respectively. μ0 represents the vacuum magnetic permeability, and (ξ, η, ζ) are the coordinates of any point on the block. Define r = [(ξ - x) 2 +(η - y) 2 +(ζ - z) 2 1 / 2 .

[0062] The north-south component of the magnetic anomaly generated by the prism at the observation point is H ax :

[0063]

[0064] The east-west component of the magnetic anomaly generated by the prism at the observation point is H ay :

[0065]

[0066] The vertical component of the magnetic anomaly generated by the prism at the observation point is Z a : ​

[0067]

[0068] After calculating the three-component magnetic anomalies obtained from the above formula, the combined responses of all prisms in each interface to the ground measurement points can be calculated respectively, which are the magnetic anomaly responses generated by the interface on the surface. Then, the superimposed simulated magnetic anomaly modulus of each measurement point is calculated using formula 6:

[0069]

[0070] where, H' ax is the sum of the north-south direction components of the magnetic anomalies of all the prisms at the observation point; H' ay is the sum of the east-west direction components of the magnetic anomalies of all the prisms at the observation point; Z' a is the sum of the vertical direction components of the magnetic anomalies of all the prisms at the observation point; Ta is the simulated magnetic anomaly modulus of the prism at the observation point.

[0071] Calculate the simulated magnetic anomaly moduli of the magnetic interface at each of the observation points in sequence, so as to obtain the simulated magnetic anomaly modulus of the magnetic interface.

[0072] Combine them in the order of the observation points into a simulated magnetic anomaly modulus matrix. The Ta calculated in this way is a matrix, corresponding to the simulated magnetic anomaly modulus of an interface. Each value in Ta corresponds to the simulated magnetic anomaly modulus measured at each measurement point. Then save this Ta as a set of simulated magnetic anomaly moduli.

[0073] S301: Input the simulated magnetic anomaly moduli of all the magnetic interfaces into the neural network for magnetic interface inversion for training, which is used to predict the depth distribution of the Curie surface.

[0074] Build a neural network that fuses the magnetic anomaly modulus and the interface depth. In the neural network, fuse all the simulated magnetic anomaly modulus data and the real magnetic anomaly modulus data to construct a training set, and integrate and analyze its distribution range characteristics and statistical characteristics, so as to identify possible singular values or data quality problems subsequently. And perform global normalization on the fused data, that is, the depth data and the modulus data respectively, so that the data range is controlled within [0,1], enabling the network to better extract data features.

[0075] Design a three-dimensional intelligent U-Net structure that can incorporate the interface depth data and the magnetic anomaly modulus data as Figure 4As shown (the black numbers represent the data volume size, and the red numbers represent the number of channels). By applying a multi-scale convolution module in the encoding and decoding parts of the network, the capture of different spatial frequencies of the modulus data is enhanced, thereby improving the response to features at different depths. The skip connection module is improved, and an attention mechanism is introduced to enhance the network model's attention to the edge features of the modulus data, enabling the model to have a more detailed boundary feature expression ability in the mapping from modulus data to depth data. There are some high-intensity peak regions in the modulus data, and the existence of these regions may cause overfitting in network training. To address this problem, an adaptive regularization process is added to the attention mechanism, that is, the strong peak regions are smoothed to limit the network model's over-reliance on specific strong signal regions. By integrating the above modules, the neural network model can further improve its ability to capture modulus data, thereby improving the prediction accuracy of the interface depth and enhancing the detailed expression of the boundary.

[0076] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, the method for calculating the magnetic anomaly modulus of the magnetic interface at each observation point on the ground surface includes:

[0077] The part of the magnetic interface around the reference plane is evenly divided into several prisms, the magnetization intensity of the prism at the observation point is projected onto three directions, and the magnetization intensity components of the prism in the three directions at the observation point are obtained according to the random magnetic dip angle and the random magnetic declination;

[0078] According to the magnetization intensity components, the three components of the magnetic anomaly response generated by the prism at the observation point are calculated. At the observation point, the magnetic anomaly three components of all the prisms in the same direction are added, and the magnetic anomaly modulus of the prism at the observation point is calculated;

[0079] Repeat the calculation to obtain the magnetic anomaly modulus of all the prisms at each observation point, that is, the magnetic anomaly modulus of the magnetic interface at each observation point. Repeat the calculation to obtain the magnetic anomaly modulus of all the magnetic interfaces at each observation point.

[0080] Assume that interface 1 is divided into 10 prisms 1, 2, 3... Calculate the three components of prism 1 at measurement point 1, calculate the three components of prism 2 at measurement point 1... Calculate the three components of 10 prisms at measurement point 1 in turn, and add the 10 H ax components, add the 10 H ay components, add the 10 Z a components. The three components after adding are the three components of interface 1. Then, through the formula on the left, the T of interface 1 at measurement point 1 can be calculated. a1, then calculate the measuring points 2, 3, 4... until the N measuring points are calculated. In this way, each measuring point has a corresponding simulated magnetic anomaly modulus T a i, and this T a i is the magnetic anomaly modulus generated by the magnetic interface at the measuring point i.

[0081] The T calculated in this way a is a matrix, corresponding to the magnetic anomaly modulus of an interface. Each value in T a corresponds to the simulated magnetic anomaly modulus measured at each measuring point. Then save this T a as a set of data.

[0082] Then calculate the T of each interface in turn a , and calculate 10,000 times to obtain 10,000 sets of data.

[0083] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, the method of repeatedly calculating the magnetic anomaly moduli of all the magnetic interfaces at each observation point on the ground includes:

[0084] Combine the magnetic anomaly moduli of all the magnetic interfaces at each observation point on the ground into a magnetic anomaly modulus matrix in the order of the observation points, that is, the magnetic anomaly modulus of the magnetic interface.

[0085] Assume that there is an observation point grid on the ground, and the number of grid points is N. The following formula 2 can represent the magnetization intensity of the observation point grid:

[0086]

[0087] Each row in this matrix represents the magnetization intensity vector of a grid point, and the magnetization intensity of the i-th point is M i =(M xi , M yi , M zi ).

[0088] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, the range of the random magnetic dip angle is -90° to 90°, and the range of the random magnetic declination is -180° to 180°.

[0089] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, project the magnetization intensity of the prism to three directions, including the north-south direction, the east-west direction, and the vertical direction.

[0090] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, the calculation method of obtaining the magnetization intensity components of the prism in three directions at a certain observation point according to the random magnetic dip angle and the random magnetic declination is:

[0091]

[0092] Among them, (M x , M y , M z ) are the components of the magnetization M in the north-south direction, east-west direction, and vertical direction respectively. The random magnetic dip angle is I1, and the random magnetic declination is D1.

[0093] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, the method for calculating the three components of the magnetic anomaly response generated by the prism at the observation point according to the magnetization components includes:

[0094] The north-south direction component of the magnetic anomaly generated by the prism at the observation point i is H ax :

[0095]

[0096] The east-west direction component of the magnetic anomaly generated by the prism at the observation point i is H ay :

[0097]

[0098] The vertical direction component of the magnetic anomaly generated by the prism at the observation point i is Z a :

[0099]

[0100] Among them, μ0 represents the magnetic permeability of vacuum, (ξ, η, ζ) are the coordinates of any point on the prism, and the distance r between the coordinates of any point on the prism and the coordinates (x, y, z) of the observation point is r = [(ξ - x) 2 + (η - y) 2 + (ζ - z) 2 1 / 2 .

[0101] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, the method for calculating the simulated magnetic anomaly modulus of all the prisms at the observation point includes:

[0102] The simulated magnetic anomaly modulus of all the prisms at the coordinates (x, y, z) of the observation point is T a (x, y, z):

[0103]

[0104] Among them, H' ax is the sum of the north-south direction components of the magnetic anomalies of all the prisms at the observation point; H'​ay is the sum of the east-west direction components of the magnetic anomalies of all the prisms at the observation point; Z' a is the sum of the vertical direction components of the magnetic anomalies of all the prisms at the observation point; Ta is the simulated magnetic anomaly modulus of all the prisms at the observation point, that is, the simulated magnetic anomaly modulus of the magnetic interface at the observation point;

[0105] Calculate the magnetic anomaly modulus of the magnetic interface at each observation point in turn, so as to obtain the magnetic anomaly modulus of the magnetic interface.

[0106] Combined with the embodiments of the present invention, there is also a preferred implementation scheme. Specifically, the method of using the magnetic anomaly modulus of the magnetic interface in a neural network for magnetic interface inversion to predict the depth distribution of the Curie surface includes:

[0107] Input the magnetic anomaly modulus data with the Curie surface depth value as the label for neural network training;

[0108] All the simulated magnetic anomaly modulus and real magnetic anomaly modulus data are fused in the training set of the neural network.

[0109] During the training process, the loss value L, accuracy A and other indicators are calculated in real time by using the following formulas 7 and 8 respectively, and the change rate L is calculated by using the following formula 9 after every 10 trainings r and A r :

[0110]

[0111]

[0112] In formula 7, n is the number of samples, that is, the number of sets of training data, y i is the true value of the i-th sample, y i is the model prediction value obtained through neural network training. In formula 8, S p is the number of correctly predicted samples. In formula 9, E represents the number of training times.

[0113] If the calculated loss change rate and accuracy change rate change by more than 10% compared with the previous calculated value, then adopt the incremental learning and reinforcement learning strategies, delete the poorly performing data, continuously integrate and input new data to optimize the training set, and at the same time adaptively adjust the network parameters to continuously improve the performance of the model obtained through neural network training. Until the change rate changes by less than 10% compared with the previous calculated value, and this state lasts for more than five times, then end the network training and save the model.

[0114] Using the method of Embodiment 1, the depth distribution of the Curie surface in a specific study area can be accurately predicted in the presence of remanent magnetism.

[0115] Embodiment 2:

[0116] Embodiment 2 provides a magnetic interface inversion device based on magnetic anomaly modulus. The device includes:

[0117] One or more processors;

[0118] A storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the magnetic interface inversion method based on magnetic anomaly modulus as described in Embodiment 1.

[0119] Figure 5 It is a schematic structural diagram of the magnetic interface inversion device based on magnetic anomaly modulus provided in Embodiment 2. Figure 5 It shows a block diagram of an exemplary magnetic interface inversion device based on magnetic anomaly modulus suitable for implementing the embodiments of the present invention. Figure 5 The shown magnetic interface inversion device based on magnetic anomaly modulus is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of the present invention.

[0120] As Figure 5 shown, the magnetic interface inversion device based on magnetic anomaly modulus is presented in the form of a general-purpose device. The components of the magnetic interface inversion device based on magnetic anomaly modulus may include but are not limited to: one or more processors or processing units, a memory, and a bus connecting different system components (including the memory and the processing unit).

[0121] The bus represents one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

[0122] The magnetic interface inversion device based on magnetic anomaly modulus typically includes a variety of computer system-readable media. These media can be any available media accessible by a device that can be modified by an intelligent logging interpretation model, including volatile and non-volatile media, removable and non-removable media.

[0123] The memory may include computer system-readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory. The magnetic interface inversion device based on magnetic anomaly modulus may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system may be used to read and write non-removable, non-volatile magnetic media (Figure 5 not shown and is typically referred to as a "hard disk drive"). Although Figure 5 not shown in, a disk drive for reading and writing to a removable non-volatile disk (such as a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (such as a CD-ROM, DVD-ROM, or other optical medium) may be provided. In these cases, each drive may be connected to the bus through one or more data medium interfaces. The memory may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0124] A program / utility with a set (at least one) of program modules may be stored, for example, in the memory. Such program modules include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment. The program modules generally perform the functions and / or methods in the embodiments described in the present invention.

[0125] The magnetic interface inversion device based on the magnetic anomaly modulus may also communicate with one or more external devices (such as a keyboard, a pointing device, a display, etc.), and may also communicate with one or more devices that enable a user to interact with the magnetic interface inversion device based on the magnetic anomaly modulus, and / or communicate with any device that enables the magnetic interface inversion device based on the magnetic anomaly modulus to communicate with one or more other devices (such as a network card, a modem, etc.). Such communication may be carried out through an input / output (I / O) interface. Also, the device for correcting the intelligent logging interpretation model may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter. As Figure 5 shown, the network adapter communicates with other modules of the magnetic interface inversion device based on the magnetic anomaly modulus through the bus. It should be understood that although not shown in the figure, other hardware and / or software modules may be used in combination with the magnetic interface inversion device based on the magnetic anomaly modulus, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0126] The processing unit executes various functional applications and data processing by running programs stored in the memory, for example, implementing the magnetic interface inversion method based on magnetic anomaly modulus provided in any embodiment of the present invention. That is: constructing a plurality of magnetic interfaces, setting a reference plane corresponding to the magnetic interfaces on the surface of the earth, and the magnetic interfaces undulating up and down around the reference plane; setting a random magnetic dip angle, a random magnetic declination and a magnetization intensity for the magnetic interfaces, calculating the simulated magnetic anomaly modulus of the magnetic interfaces at each observation point on the surface of the earth, and repeating the calculation to obtain the simulated magnetic anomaly modulus of all the magnetic interfaces at each observation point on the surface of the earth; using the simulated magnetic anomaly modulus of the magnetic interfaces in a neural network for magnetic interface inversion to predict the depth distribution of the Curie surface.

[0127] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A magnetic interface inversion method based on magnetic anomaly modulus, characterized in that: The method comprises: Constructing a plurality of magnetic interfaces, setting a reference plane corresponding to the surface of the earth and the magnetic interface, wherein the magnetic interface fluctuates up and down around the reference plane; Setting a random magnetic inclination, a random magnetic declination and a magnetization intensity for the magnetic interface, calculating a simulated magnetic anomaly modulus of the magnetic interface at each observation point on the surface, and repeatedly calculating to obtain the simulated magnetic anomaly modulus of all the magnetic interfaces at each observation point on the surface; The simulated magnetic anomaly modulus of the magnetic interface is used in a neural network for magnetic interface inversion to predict the Curie point depth distribution.

2. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 1, characterized in that: The method for calculating the simulated magnetic anomaly modulus of the magnetic interface at each observation point on the surface includes: Uniformly divide the portion of the magnetic interface surrounding the reference plane into a plurality of prisms, project the magnetization intensity of the prism at the observation point to three directions, and obtain the magnetization intensity components of the prism in the three directions at the observation point according to the random magnetic inclination and the random magnetic declination; The three components of the magnetic anomaly response generated by the prism at the observation point are calculated based on the magnetization intensity components, the three components of the magnetic anomaly of all the prisms in the same direction are added together at the observation point, and the simulated magnetic anomaly moduli of all the prisms at the observation point are calculated; Repeat the calculation to obtain the simulated magnetic anomaly modulus of all the prisms at each of the observation points, that is, to obtain the simulated magnetic anomaly modulus of the magnetic interface at each of the observation points, and repeat the calculation to obtain the simulated magnetic anomaly modulus of all the magnetic interfaces at each of the observation points.

3. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 1, characterized in that: The method of repeatedly calculating and obtaining the simulated magnetic anomaly modulus of all the magnetic interfaces at each observation point on the surface includes: The simulated magnetic anomaly moduli of the magnetic interface at each observation point on the surface are combined into a simulated magnetic anomaly modulus matrix according to the order of the observation points, that is, the simulated magnetic anomaly modulus of the magnetic interface.

4. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 1, characterized in that: The random magnetic inclination angle ranges from -90° to 90°, and the random magnetic declination angle ranges from -180° to 180°.

5. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 2, characterized in that: The magnetization intensity of the prism is projected to three directions, which include north-south direction, east-west direction and vertical direction.

6. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 2, characterized in that: The calculation method of obtaining the magnetization intensity components of the prism in three directions at a certain observation point according to the random magnetic inclination and the random magnetic declination is: Among them, (M x ,M y ,M z ) are the components of the magnetization intensity M in the north-south direction, east-west direction and vertical direction, the random magnetic inclination is I1, and the random magnetic declination is D1.

7. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 2, characterized in that: The method of calculating the three components of the magnetic anomaly response generated by the prism at the observation point based on the magnetization intensity components includes: The north-south component of the magnetic anomaly generated by the prism at the observation point is H ax : The east-west component of the magnetic anomaly generated by the prism at the observation point is H ay : The vertical component of the magnetic anomaly generated by the prism at the observation point is Z a : Wherein, μ0 represents the vacuum permeability, (ξ,η,ζ) is the coordinate of any point on the prism, and the distance between the coordinate of any point on the prism and the coordinate of the observation point (x,y,z) is r=[(ξ-x) 2 +(η-y) 2 +(ζ-z) 2 ] 1 / 2 .

8. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 2 is characterized in that: The method for calculating the simulated magnetic anomaly modulus of all the prisms at the observation point includes: The simulated magnetic anomaly modulus of all the prisms at the observation point coordinates (x, y, z) is T a (x,y,z): Among them, H' ax is the sum of the north-south components of the magnetic anomalies of all the prisms at the observation point; H' ay is the sum of the east-west components of the magnetic anomalies of all the prisms at the observation point; Z' a is the sum of the vertical direction components of the magnetic anomalies of all the prisms at the observation point; Ta is the simulated magnetic anomaly modulus of all the prisms at the observation point, that is, the simulated magnetic anomaly modulus of the magnetic interface at the observation point; The simulated magnetic anomaly modulus of the magnetic interface at each of the observation points is calculated in sequence to obtain the simulated magnetic anomaly modulus of the magnetic interface.

9. The magnetic interface inversion method based on magnetic anomaly modulus according to claim 2, characterized in that: The method of using the simulated magnetic anomaly modulus of the magnetic interface to predict Curie point depth distribution in a neural network for magnetic interface inversion comprises: The input magnetic anomaly modulus data with the Curie depth value as the label is used for neural network training; All simulated magnetic anomaly modulus and real magnetic anomaly modulus data are integrated into the training set of the neural network.

10. A magnetic interface inversion device based on magnetic anomaly modulus, characterized in that the device include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the magnetic interface inversion method based on magnetic anomaly modulus as described in any one of claims 1-9.

Images