A three-dimensional gravity cross-inversion method, system, storage medium and electronic device

Abstract

The present invention relates to a three-dimensional gravity cross-inversion method, system, storage medium and electronic device. The method includes: generating a conversion factor corresponding to the gravity observation field based on the type of the gravity observation field; combining the original forward formula of the three-dimensional model with the conversion factor to obtain a cross-forward formula; and performing inversion on the cross-forward formula based on the inversion formula to obtain the inversion result of the gravity observation field. By generating corresponding conversion factors for different types of gravity observation fields and using the combination of kernel functions obtained from different conversion factors for gravity cross-inversion to obtain the inversion results of corresponding types, the present invention realizes the advantage that constraint information can be flexibly added by using the three-dimensional gravity inversion method, improves the resolution of density boundary recognition, and thus enhances the accuracy of identifying different abnormal bodies.

Description

Background Art

[0002] A fault zone, also known as a "fault belt", is a zone composed of a main fault plane and the broken rock blocks on both sides, as well as several secondary faults or fracture surfaces, and is one of the important geological features. The characteristics of a fault in a three-dimensional density model are generally density boundaries, that is, regions with a relatively high rate of spatial change of density. In current research, the observational data used for calculation in conventional gravity inversion corresponds to the forward kernel function, so the inversion results are all density values. If the observational data used for inversion does not correspond to the kernel function, the inversion result will not be a density value. For example, if the observational data is the gravity horizontal gradient value and the kernel function is the kernel function corresponding to the gravity value, the inversion result will not be a density value but the horizontal gradient of the density value. Therefore, gravity inversion with different combinations of observational data and kernel functions will produce different types of inversion results, and this type of inversion is defined as three-dimensional gravity cross-inversion.

[0003] However, in current research, there are relatively few methods directly using three-dimensional inversion to study density boundaries, and there is currently a lack of technical solutions to improve the resolution of density boundary identification. Therefore, it is urgent to propose a three-dimensional gravity cross-inversion method to solve the above technical problems. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a three-dimensional gravity cross-inversion method, system, storage medium, and electronic device.

[0005] The technical solution of a three-dimensional gravity cross-inversion method of the present invention is as follows:

[0006] Generate a conversion factor corresponding to the gravity observation field based on the type of the gravity observation field;

[0007] Combine the original forward formula of the three-dimensional model with the conversion factor to obtain a cross-forward formula;

[0008] Based on the inversion formula, invert the cross-forward formula to obtain the inversion result of the gravity observation field.

[0009] The beneficial effects of a three-dimensional gravity cross-inversion method of the present invention are as follows:

[0010] The method of the present invention generates corresponding conversion factors for different types of gravity observation fields, and performs gravity cross-inversion using the kernel function combinations obtained from different conversion factors to obtain inversion results of corresponding types, realizing the advantage of being able to flexibly add constraint information using the three-dimensional gravity inversion method, improving the resolution of density boundary identification, and thus enhancing the accuracy of identifying different abnormal bodies.

[0011] On the basis of the above solution, a three-dimensional gravity cross-inversion method of the present invention can also be improved as follows.

[0012] Furthermore, the type of the gravity observation field is: vertical gravity gradient field, x-direction gravity horizontal gradient field, or y-direction gravity horizontal gradient field.

[0013] Furthermore, the conversion factor is: vertical gradient conversion factor, x-direction horizontal gradient conversion factor, or y-direction horizontal gradient conversion factor.

[0014] Furthermore, the original forward formula is: The cross forward formula is: where d represents the forward anomaly of the three-dimensional model, F represents the forward transform of the fast Fourier transform, F -1 represents the inverse transform of the fast Fourier transform, represents the forward kernel matrix of the k-th layer, k represents the k-th depth layer in the three-dimensional model, represents the model matrix of the k-th layer, p represents the total number of depth layers in the three-dimensional model, is the conversion factor, is the converted density distribution obtained by applying the conversion factor to the model matrix of the k-th layer.

[0015] Furthermore, the inversion result of the gravity observation field is: the density result corresponding to the type of the gravity observation field.

[0016] Furthermore, the density result is: the vertical gradient of the density distribution, the x-direction horizontal gradient of the density distribution, or the y-direction horizontal gradient of the density distribution.

[0017] The technical solution of a three-dimensional gravity cross-inversion system of the present invention is as follows:

[0018] It includes: a generation module, a conversion module, and an operation module;

[0019] The generation module is used for: generating the conversion factor corresponding to the gravity observation field based on the type of the gravity observation field;

[0020] The conversion module is used for: combining the original forward formula of the three-dimensional model with the conversion factor to obtain a cross forward formula;

[0021] The operation module is used for: performing inversion on the cross forward formula based on the inversion formula to obtain the inversion result of the gravity observation field.

[0022] The beneficial effects of a three-dimensional gravity cross-inversion system of the present invention are as follows:

[0023] The system of the present invention generates corresponding conversion factors for different types of gravity observation fields, and performs gravity cross-inversion using the combination of kernel functions obtained from different conversion factors to obtain the inversion results of corresponding types, realizing the advantage that constraint information can be flexibly added using the three-dimensional gravity inversion method, improving the resolution of density boundary recognition, and thus enhancing the accuracy of different anomaly body recognition.

[0024] Based on the above solution, a three-dimensional gravity cross-inversion system of the present invention can also be improved as follows.

[0025] Further, the types of the gravity observation fields are: gravity vertical gradient field, gravity horizontal gradient field in the x direction, or gravity horizontal gradient field in the y direction.

[0026] The technical solution of a storage medium of the present invention is as follows:

[0027] Instructions are stored in the storage medium, and when a computer reads the instructions, the computer is caused to execute the steps of a three-dimensional gravity cross-inversion method of the present invention.

[0028] The technical solution of an electronic device of the present invention is as follows:

[0029] It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. It is characterized in that when the processor executes the computer program, the computer is caused to execute the steps of a three-dimensional gravity cross-inversion method of the present invention. Description of the Drawings

[0030] Figure 1 It is a schematic flowchart of a three-dimensional gravity cross-inversion method according to an embodiment of the present invention;

[0031] Figure 2 It is an anomaly body model and a forward gravity field diagram in a three-dimensional gravity cross-inversion method according to an embodiment of the present invention;

[0032] Figure 3 It is a horizontal slice diagram of the gravity field focusing inversion result in a three-dimensional gravity cross-inversion method according to an embodiment of the present invention;

[0033] Figure 4 It is a horizontal slice diagram of the horizontal gradient of the gravity field focusing inversion result in a three-dimensional gravity cross-inversion method according to an embodiment of the present invention;

[0034] Figure 5 It is a horizontal slice diagram of the gravity field horizontal gradient focusing cross-inversion result in a three-dimensional gravity cross-inversion method according to an embodiment of the present invention;

[0035] Figure 6 It is a slice diagram of the inversion along line I in a three-dimensional gravity cross-inversion method according to an embodiment of the present invention;

[0036] Figure 7 It is the inversion slice diagram of line II in a three-dimensional gravity cross-inversion method according to an embodiment of the present invention;

[0037] Figure 8 It is the structural schematic diagram of a three-dimensional gravity cross-inversion system according to an embodiment of the present invention. Specific embodiments

[0038] As Figure 1 shown, a three-dimensional gravity cross-inversion method according to an embodiment of the present invention includes the following steps:

[0039] S1. Generate a conversion factor corresponding to the gravity observation field based on the type of the gravity observation field.

[0040] Among them, the gravity observation field is the observation data of three-dimensional gravity, and the type of the gravity observation field is the type of the observation data of three-dimensional gravity.

[0041] Among them, the function of the conversion factor is: when the type of the gravity observation field does not correspond to the forward kernel function, convert the result of the forward formula to obtain the conversion of the density distribution corresponding to the type of the gravity observation field.

[0042] S2. Combine the original forward formula of the three-dimensional model with the conversion factor to obtain a cross-forward formula.

[0043] Among them, the three-dimensional model is: a three-dimensional model of the underground half-space. Specifically, the x-axis and the y-axis are defined as the horizontal directions, and the z-axis is vertically downward, representing the eastward, northward, and underground depth directions respectively. Let (ξ, η, ζ) be the coordinates of any volume element dv = dξdηdζ in the abnormal body, then the gravitational potential formula dV of the mass element dm = ρ(ξ, η, ζ)dv at any point (x, y, z) in space is: In the formula: γ is the gravitational constant, and its value is 6.67×10- 11 m 3 / (kg·s 2 );r = [(x - ξ) 2 +(y - η) 2 +(z - ζ) 2 1 / 2 , r is the distance from the mass element (ξ, η, ζ) to any point (x, y, z) in space; integrate the gravitational potential formula dV over the underground half-space according to the prism to obtain the gravitational potential formula V(x, y, z): In the formula: ρ(ξ, η, ζ) represents the spatial density distribution of the underground half-space; a and b are respectively half of the lengths of the single-layer medium in the x-direction and the y-direction; L is the thickness of the underground half-space; H represents the top surface depth.

[0044] ​In the spatial domain, to illustrate the relationship between the forward calculation of a three-dimensional model and convolution, the underground three-dimensional space is divided into multiple horizontal layered media. After vertically differentiating the gravitational potential formula V(x, y, z), the forward gravity formula g(x, y, z) at a horizontal height z for a single-layer medium with a thickness of l (l < L) can be obtained: where: ρ(ξ , η) represents the lateral density distribution of the single-layer medium; a and b are respectively half of the lengths of the single-layer medium in the x-direction and y-direction. h represents the depth of the top surface, and l represents the thickness of the single-layer medium. For a single-layer medium, when the density function ρ is restricted to be invariant longitudinally and h and l are fixed, G(x - ξ, y - η) is only a function of x, ξ, y, η, and the density function ρ(ξ, η) is a function of ξ, η. According to the definition of signal convolution, g = G * ρ, where ρ is the density distribution matrix of the plate-like body in the lateral space, called the model signal; G can be regarded as the forward gravity kernel matrix of a vertical line body with a depth of h, a thickness of l, and a density of 1, and can be called the forward filter; "*" represents convolution.

[0045] In actual calculations, both the underground space and the observation points are in discrete form. The horizontal positions of the gravity measurement points and the prisms are made to correspond one by one, so as to ensure that the resolution of the model is high enough and it is convenient to design targeted calculation strategies. Convolution can be divided into linear convolution and circular convolution. Only circular convolution can use the FFT algorithm to improve the calculation speed.

[0046] Among them, the original forward formula is: where d represents the forward anomaly of the three-dimensional model, F represents the forward transform of the fast Fourier transform, F -1 represents the inverse transform of the fast Fourier transform, represents the forward kernel matrix of the k-th layer, k represents the k-th depth layer in the three-dimensional model, represents the model matrix of the k-th layer, p represents the total number of depth layers in the three-dimensional model, and "·" represents the Hadamard product, that is, the operation of multiplying the corresponding elements in matrices of the same order.

[0047] Among them, the conventional three-dimensional gravity inversion process is: for example, when the conversion factor is combined with the kernel function, where: is the converted field obtained by the gravity field through the conversion factor ; is the forward kernel function corresponding to the converted field. In the above formula, the forward kernel function corresponds to the observed field, and the inversion result is the density value. When the conversion factor is the vertical gradient filtering factor, then represents the vertical gradient of the gravity field, Denote the vertical gradient of the kernel function of the k-th layer. At this time, the above formula becomes the forward formula of the vertical gravity gradient.

[0048] S3. Based on the inversion formula, invert the cross-forward formula to obtain the inversion result of the gravity observation field.

[0049] For example, when the gravity observation field is the x-direction gravity horizontal gradient field, the inversion result is the x-direction horizontal gradient of the density distribution. Its inversion formula is:

[0050] ρ x =ρ x0 +(WG T D -1 G) -1 WG T D -1 (d - Gρ x0 );

[0051] Among them, ρ x represents the x-direction horizontal gradient of the density ρ, D -1 is the diagonal data covariance matrix, and W is the model weighting matrix.

[0052] Since the kernel function of this inversion method remains unchanged and only the observable quantity is changed, the inversion algorithm is the same as the foregoing. It should be noted that this type of inversion calculation is equivalent to the gravity gradient inversion calculation. Therefore, the value of β in the depth weighting function used for inversion calculation is 3.

[0053] Preferably, the type of the gravity observation field is: gravity field, gravity vertical gradient field, x-direction gravity horizontal gradient field or y-direction gravity horizontal gradient field.

[0054] Preferably, the conversion factor is: vertical gradient conversion factor, x-direction horizontal gradient conversion factor or y-direction horizontal gradient conversion factor.

[0055] Among them, the common expression forms of the conversion factor are:

[0056]

[0057] In the formula: u and v are the wave numbers in the x and y directions respectively; is the vertical gradient conversion factor; is the x-direction horizontal gradient conversion factor; is the y-direction horizontal gradient conversion factor; The continuation conversion factor, when h < 0, it is upward continuation, and when h > 0, it is downward continuation.

[0058] Preferably, the original forward formula is: The cross-forward formula is: Among them, d represents the forward anomaly of the 3D model, F represents the forward transform of the fast Fourier transform, and F -1 represents the inverse transform of the fast Fourier transform, represents the forward kernel matrix of the k-th layer, and k represents the k-th depth layer in the 3D model, represents the model matrix of the k-th layer, and p represents the total number of depth layers in the 3D model, is the conversion factor, is the converted density distribution obtained by applying the conversion factor to the model matrix of the k-th layer.

[0059] Specifically, when the conversion factor is combined with the density , the original forward formula is transformed into a cross-forward formula:

[0060] Preferably, the inversion result of the gravity observation field is: the density result corresponding to the type of the gravity observation field.

[0061] Preferably, the density result is: the vertical gradient of the density distribution, the horizontal gradient in the x direction of the density distribution, or the horizontal gradient in the y direction of the density distribution.

[0062] It should be noted that in order to compare the effects of gravity field inversion and gravity cross-inversion on the identification of the boundary of the abnormal body to be measured, a 3D model as shown in Figure 2 (a) is set, and the specific parameters are shown in Table 1.

[0063] Table 1 Model Parameter Table

[0064]

[0065] The underground space is divided into 101×121×15 prismatic elements of 100m×100m×100m, and the ground observation data are located at the plane center of the grid prisms, with a total of 101×121 observation data. The gravity forward field of this model ( Figure 2 (b)), the horizontal gradient in the x direction of the gravity field ( Figure 2 (c)) and the horizontal gradient in the y direction of the gravity field ( Figure 2 (d)) are used as the basic fields.

[0066] In this experiment, two methods are mainly used to obtain the boundary information of the abnormal body. The first is to perform 3D focused inversion on the gravity anomaly ( Figure 2 (b)), and then calculate the horizontal gradients in the x direction and y direction of the inversion result ( Figure 3 ); the second is to directly target the horizontal gradient anomalies in the x direction and y direction of the gravity field ( Figure 4 ); ( Figure 2 (c),Figure 2 (d)) Perform 3D cross-focusing inversion separately to obtain the horizontal gradient results in the x and y directions of the 3D density body. Finally, take the absolute value of the inversion results of the horizontal gradient anomalies in the x and y directions and then average them to obtain the horizontal gradient result of the 3D density body. Figure 5 )

[0067] According to the horizontal slice of the gravity field focusing inversion result ( Figure 3 ), it can be qualitatively seen that the deep and large anomaly body (Body No. 1) mainly appears below 500 m; the medium-depth anomaly body (Body No. 2) mainly appears in the range of 300 m - 900 m; the medium-shallow anomaly body (Body No. 3) mainly appears in the range of 100 m - 900 m; the shallow and small anomaly body (Body No. 4) mainly appears above 300 m. The gravity focusing inversion result has a high degree of coincidence with the theoretical result in terms of depth.

[0068] According to the horizontal gradient anomaly of the gravity field focusing inversion result (Method 1) ( Figure 4 , where the dashed line in Figure 6 is the position of the theoretical anomaly body) and the gravity field horizontal gradient focusing cross-inversion result (Method 2) ( Figure 5 ), it shows that in the vertical distribution, the calculation results of the two methods for depicting the 4 anomaly bodies are basically consistent with the 3D gravity focusing inversion result, and both are highly consistent with the theoretical result in terms of depth. In the horizontal distribution, for the No. 2, No. 3, and No. 4 anomaly bodies, Method 1 depicts the horizontal distribution range of the anomaly body relatively clearly, and the depiction of the bottom of the anomaly body is basically in a contracted state, while Method 2 depicts the horizontal distribution range of the anomaly body clearly, and there is no contraction in the depiction of the bottom of the anomaly body, maintaining a high degree of coincidence with the theoretical model. For the No. 1 anomaly body, the depictions of the horizontal range of the anomaly body by the two methods basically increase with the increase of depth. Method 1 depicts the horizontal distribution range of the anomaly body relatively accurately, while Method 2 has a large error in depicting the horizontal distribution range of the anomaly body.

[0069] Furthermore, vertical slices were made on the inversion results. The focusing inversion result ( Figure 6 (a), Figure 7 (a)) shows that the central position of the No. 2 anomaly body corresponds to the center of the theoretical model; the central positions of the No. 3 and No. 4 anomaly bodies are deeper than the center of the theoretical model. The reason is that the No. 3 and No. 4 anomaly bodies are relatively thin plate-like bodies, and the spectrum of their forward fields has relatively more low-frequency information, resulting in a deeper inversion result; although the No. 1 body is restricted by the inversion space, its central position is basically consistent with the center of the theoretical model.

[0070] The horizontal gradient anomaly of the gravity field focusing inversion result ( Figure 6 (b), Figure 7(b)) shows that the recognition features of the lateral boundaries of the No. 2, No. 3, and No. 4 anomalies are an inverted "eight" shape. Although the recognized positions are basically the same as the positions of the lateral boundaries of the anomalies (the dotted lines in the figure represent the boundaries of the anomalies), the tendency features of the lateral boundaries of the anomalies cannot be accurately characterized. The recognition feature of the lateral boundary of the No. 1 anomaly is a regular "eight" shape.

[0071] The horizontal gradient focusing cross-inversion result of the gravity field ( Figure 6 (c)) Figure 7 (c)) shows that the lateral boundaries of the No. 2, No. 3, and No. 4 anomalies are clearly recognized, and the anomaly centers of the inversion results have a high degree of coincidence with the boundary positions of the theoretical model. However, there are relatively large errors in the characterization of the No. 1 body, and neither the position nor the direction of the boundary can correspond.

[0072] This experiment shows that the processing results of the three methods of gravity field focusing inversion, the horizontal gradient of the gravity field focusing inversion result, and the gravity field horizontal gradient focusing inversion have their own advantages in the recognition of the boundary features of the three-dimensional model body. Generally speaking, the inversion results based on the gravity field have a better recognition effect on deep anomalies, and the inversion results based on the gravity gradient field have a better recognition effect on shallow anomalies. The reason is that the gravity gradient field is equivalent to the high-pass filter of the gravity field. Therefore, the gravity gradient field contains more high-frequency components and has a high recognition accuracy for shallow anomalies, while the gravity field contains more low-frequency components and is more accurate in the recognition of deep anomalies.

[0073] The technical solution of this embodiment realizes the advantage of flexibly adding constraint information by using the three-dimensional gravity inversion method by generating corresponding conversion factors for different types of gravity observation fields and performing gravity cross-inversion using the kernel function combinations obtained by different conversion factors, improving the resolution of density boundary recognition, and thus enhancing the accuracy of the recognition of different anomalies.

[0074] As Figure 8 shown, a three-dimensional gravity cross-inversion system 200 according to an embodiment of the present invention includes: a generation module 210, a conversion module 220, and an operation module 230;

[0075] The generation module 210 is configured to: generate a conversion factor corresponding to the gravity observation field based on the type of the gravity observation field;

[0076] The conversion module 220 is configured to: combine the original forward formula of the three-dimensional model with the conversion factor to obtain a cross-forward formula;

[0077] The operation module 230 is configured to: perform inversion on the cross-forward formula based on the inversion formula to obtain the inversion result of the gravity observation field.

[0078] Preferably, the type of the gravity observation field is: a vertical gravity gradient field, an x-direction gravity horizontal gradient field, or a y-direction gravity horizontal gradient field.

[0079] The technical solution of this embodiment generates corresponding conversion factors for different types of gravity observation fields, and uses the combination of kernel functions obtained by different conversion factors for gravity cross-inversion to obtain the inversion results of corresponding types, realizing the advantage that the three-dimensional gravity inversion method can flexibly add constraint information, improving the resolution of density boundary recognition, and thus enhancing the accuracy of identifying different abnormal bodies.

[0080] For the parameters and steps of each module in the above-mentioned three-dimensional gravity cross-inversion system 200 of this embodiment to implement corresponding functions, reference can be made to the parameters and steps in the embodiment of a three-dimensional gravity cross-inversion method in the above text, which will not be elaborated here.

[0081] A storage medium provided by an embodiment of the present invention includes: instructions are stored in the storage medium, and when a computer reads the instructions, the computer is made to execute the steps of a three-dimensional gravity cross-inversion method. Specifically, reference can be made to the parameters and steps in the embodiment of a three-dimensional gravity cross-inversion method in the above text, which will not be elaborated here.

[0082] Computer storage media such as: USB flash drives, external hard drives, etc.

[0083] An electronic device provided by an embodiment of the present invention includes a memory, a processor, and a computer program stored on the memory and executable on the processor. It is characterized in that when the processor executes the computer program, the computer is made to execute the steps of a three-dimensional gravity cross-inversion method. Specifically, reference can be made to the parameters and steps in the embodiment of a three-dimensional gravity cross-inversion method in the above text, which will not be elaborated here.

[0084] Those skilled in the art know that the present invention can be implemented as a method, a system, a storage medium, and an electronic device.

[0085] Therefore, the present invention can be embodied in the following forms, namely: it can be entirely hardware, or entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" in this document. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, which contain computer-readable program code. Any combination of one or more computer-readable media can be adopted. The computer-readable media can be computer-readable signal media or computer-readable storage media. The computer-readable storage media can be, for example, but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limitations on the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A three-dimensional gravity cross-inversion method, characterized in that Including: Generating a conversion factor corresponding to the gravity observation field based on the type of the gravity observation field; Combining the original forward modeling formula of the three-dimensional model with the conversion factor to obtain a cross forward modeling formula; Inverting the cross forward modeling formula based on the inversion formula to obtain the inversion result of the gravity observation field; The type of the gravity observation field is: gravity vertical gradient field, x-direction gravity horizontal gradient field or y-direction gravity horizontal gradient field; the conversion factor is: vertical gradient conversion factor, x-direction horizontal gradient conversion factor or y-direction horizontal gradient conversion factor; The original forward modeling formula is as follows: The cross forward modeling formula is as follows: Wherein, d represents the forward modeling anomaly of the three-dimensional model, F represents the forward transform of the fast Fourier transform, and F -1 represents the inverse transform of the fast Fourier transform, represents the forward modeling kernel matrix of the k-th layer, and k represents the k-th depth layer in the three-dimensional model, represents the model matrix of the k-th layer, and p represents the total number of depth layers in the three-dimensional model, is the conversion factor, is the converted density distribution obtained by multiplying the model matrix of the k-th layer by the conversion factor; When the gravity observation field is the x-direction gravity horizontal gradient field, the inversion result is the x-direction horizontal gradient of the density distribution, and its inversion formula is: ρ x = ρ x0 + (WG T D -1 ) -1 WG T D -1 (d - Gρ x0 ); where ρ x represents the horizontal gradient of the density ρ in the x - direction, D -1 is the diagonal data covariance matrix, and W is the model weighting matrix.

2. The three-dimensional gravity cross-inversion method according to claim 1, wherein The inversion result of the gravity observation field is: a density result corresponding to the type of the gravity observation field.

3. The three-dimensional gravity cross-inversion method according to claim 2, wherein The density result is: the vertical gradient of the density distribution, the x-direction horizontal gradient of the density distribution or the y-direction horizontal gradient of the density distribution.

4. A three-dimensional gravity cross-inversion system, characterized in that, Including: A generation module, a conversion module and a running module; The generation module is used for: generating a conversion factor corresponding to the gravity observation field based on the type of the gravity observation field; The conversion module is used for: combining the original forward modeling formula of the three-dimensional model with the conversion factor to obtain a cross forward modeling formula; The running module is used for: inverting the cross forward modeling formula based on the inversion formula to obtain the inversion result of the gravity observation field; The type of the gravity observation field is: gravity vertical gradient field, x-direction gravity horizontal gradient field or y-direction gravity horizontal gradient field; the conversion factor is: vertical gradient conversion factor, x-direction horizontal gradient conversion factor or y-direction horizontal gradient conversion factor; The original forward modeling formula is as follows: The cross forward modeling formula is as follows: Wherein, d represents the forward modeling anomaly of the 3D model, F represents the forward transform of the fast Fourier transform, and F -1 represents the inverse transform of the fast Fourier transform, represents the forward modeling kernel matrix of the k-th layer, and k represents the k-th depth layer in the 3D model, represents the model matrix of the k-th layer, and p represents the total number of depth layers in the 3D model, is the conversion factor, is the converted density distribution obtained by applying the conversion factor to the model matrix of the k-th layer; When the gravity observation field is the x-direction gravity horizontal gradient field, the inversion result is the x-direction horizontal gradient of the density distribution, and its inversion formula is: ρ x = ρ x0 + (WG T D -1 ) -1 WG T D -1 (d - Gρ x0 ); Among them, ρ x represents the horizontal gradient in the x-direction of the density ρ, D -1 is the diagonal data covariance matrix, and W is the model weighting matrix.

5. A storage medium, characterized in that, Instructions are stored in the storage medium, and when the computer reads the instructions, the computer is caused to execute the three-dimensional gravity cross-inversion method according to any one of claims 1 to 3.

6. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the computer is caused to execute the three-dimensional gravity cross-inversion method according to any one of claims 1 to 3.

Images