An electric source short offset transient electromagnetic quasi-three-dimensional inversion method, system, device and medium
By combining Gaussian interpolation and nonlinear conjugate gradient algorithms, the problem of three-dimensional inversion using transient electromagnetic methods at short offsets was solved, achieving efficient and accurate imaging and geological interpretation of underground electrical structures.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INSTITUTE OF GEOLOGY AND GEOPHYSICS CHINESE ACADEMY OF SCIENCES
- Filing Date
- 2025-10-10
- Publication Date
- 2026-07-03
AI Technical Summary
Traditional transient electromagnetic methods are difficult to accurately describe the complex three-dimensional electrical structure underground under short offset conditions. Two-dimensional inversion methods have significant limitations, and full three-dimensional inversion calculations are inefficient and difficult to apply in practice.
An initial 3D model is constructed using the Gaussian interpolation method, and a pseudo-3D inversion is performed using a nonlinear conjugate gradient algorithm. During the inversion process, the horizontal and vertical constraint factors are adaptively adjusted, and GPU parallel computing is used to accelerate the inversion and achieve efficient inversion.
It improves the accuracy and computational efficiency of three-dimensional imaging of underground electrical structures, enabling high-precision underground geological interpretation and resource assessment.
Smart Images
Figure CN121208950B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of geophysical exploration technology, and in particular to a method, system, equipment, and medium for short-offset transient electromagnetic pseudo-three-dimensional inversion using an electrical source. Background Technology
[0002] The Transient Electromagnetic Method (TEM), a crucial technique in geophysical exploration, reveals the electrical distribution characteristics of subsurface media by transmitting current pulses into the subsurface and observing the decaying secondary field over time after power is cut off. However, traditional TEM data inversion is mostly based on two-dimensional models, which struggle to accurately describe the complex three-dimensional electrical structure of the subsurface. This limitation is particularly pronounced under short offset conditions (i.e., short distances between the transmitter and receiver), where the spatial distribution of the electromagnetic field becomes even more complex. Furthermore, full three-dimensional inversion requires millions of grid cells, resulting in low computational efficiency and hindering practical application. Summary of the Invention
[0003] The purpose of this invention is to provide a method, system, device and medium for short offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources, aiming to solve or improve at least one of the above-mentioned technical problems.
[0004] To achieve the above objectives, the present invention provides the following solution:
[0005] A method for short-offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources includes:
[0006] Acquire multi-source data and normalize the multi-source data to obtain preprocessed data; the multi-source data includes geological information and electrical parameters;
[0007] Based on the preprocessed data, the underground space is divided and assigned values using the Gaussian interpolation method to construct an initial three-dimensional model; the construction process of the initial three-dimensional model includes kernel function optimization and multi-source data fusion.
[0008] Based on the initial 3D model, a pseudo-3D inversion is performed using a nonlinear conjugate gradient algorithm, and the values of the horizontal and vertical constraint factors are adaptively adjusted during the inversion process to obtain inversion data; the pseudo-3D inversion process includes objective function enhancement and algorithm acceleration strategies.
[0009] Three-dimensional imaging information of underground electrical structures is extracted from the inversion data; the three-dimensional imaging information is used for geological interpretation and resource assessment.
[0010] Optionally, the kernel function optimization specifically involves:
[0011] ;
[0012] in, Represents a kernel function, which measures the difference between two points in space. r i and r j The correlation, The bandwidth parameter representing the kernel function ， This represents the adaptive adjustment coefficient. Indicates the distance between measuring points.
[0013] Optionally, the multi-source data fusion specifically includes:
[0014] Integrating existing borehole resistivity ρ dill Surface geological map Construct a hybrid constraint matrix:
[0015] ;
[0016] in, Represents the kernel function. This represents the weighting coefficient, which balances the contributions of borehole and surface geological data.
[0017] Optionally, the objective function enhancement specifically includes:
[0018] ;
[0019] Among them, W d Let m be the data weight matrix, m be the model vector to be determined, and d be the weight matrix. obs For the observed data vector, d pre To predict data vectors, This is the horizontal regularization parameter. For the horizontal gradient operator, For the vertical first derivative operator, This is the regularization parameter in the vertical direction.
[0020] Optionally, the algorithm acceleration strategy specifically includes:
[0021] GPU-parallel Jacobian matrix computation is employed, and preconditioners are introduced: The iteration count is less than 50; where J is the Jacobian matrix. Let I be a small positive number, and let I be the identity matrix.
[0022] This invention also provides a short-offset transient electromagnetic pseudo-three-dimensional inversion system for electrical sources, comprising:
[0023] The data acquisition and preprocessing unit is used to acquire multi-source data and normalize the multi-source data to obtain preprocessed data; the multi-source data includes geological information and electrical parameters.
[0024] An initial 3D model construction unit is used to divide and assign values to the underground space based on the preprocessed data using the Gaussian interpolation method, thereby constructing an initial 3D model; the construction process of the initial 3D model includes kernel function optimization and multi-source data fusion.
[0025] The pseudo-3D inversion unit is used to perform pseudo-3D inversion based on the initial 3D model using a nonlinear conjugate gradient algorithm, and adaptively adjusts the values of the horizontal and vertical constraint factors during the inversion process to obtain inversion data; the pseudo-3D inversion process includes objective function enhancement and algorithm acceleration strategies;
[0026] An imaging unit is used to extract three-dimensional imaging information of underground electrical structures based on the inversion data; the three-dimensional imaging information is used for geological interpretation and resource assessment.
[0027] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the above-described method for short-offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources.
[0028] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the short-offset transient electromagnetic pseudo-three-dimensional inversion method for electrical sources as described above.
[0029] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0030] This invention discloses a method, system, device, and medium for short-offset transient electromagnetic pseudo-3D inversion of electrical sources. The method includes normalizing the multi-source data, and based on the preprocessed data, using the Gaussian interpolation method to divide and assign values to the underground space to construct an initial 3D model; based on the initial 3D model, using a nonlinear conjugate gradient algorithm to perform pseudo-3D inversion, and adaptively adjusting the values of lateral and longitudinal constraint factors during the inversion process to obtain inversion data; extracting 3D imaging information of underground electrical structures from the inversion data; and using the 3D imaging information for geological interpretation and resource assessment. This invention can establish an initial 3D model through Gaussian interpolation, perform efficient inversion calculations using a nonlinear conjugate gradient algorithm, and adaptively adjust lateral and longitudinal constraint factors during the inversion process to achieve high-precision pseudo-3D imaging of underground electrical structures. Attached Figure Description
[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0032] Figure 1 This is a schematic diagram of the short-offset transient electromagnetic pseudo-three-dimensional inversion method for electrical sources in this embodiment;
[0033] Figure 2 This is a schematic diagram of the initial 3D model constructed using Gaussian interpolation in this embodiment.
[0034] Figure 3 This is a diagram showing the pseudo-3D inversion results in this embodiment. Detailed Implementation
[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0036] The purpose of this invention is to provide a method, system, device and medium for short offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources, aiming to solve or improve at least one of the above-mentioned technical problems.
[0037] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0038] like Figures 1-3 As shown, this invention provides a method for short-offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources, comprising:
[0039] Step 100: Acquire multi-source data and normalize the multi-source data to obtain preprocessed data; the multi-source data includes geological information and electrical parameters.
[0040] Step 200: Based on the preprocessed data, the underground space is divided and assigned values using the Gaussian difference method to construct an initial three-dimensional model; the construction process of the initial three-dimensional model includes kernel function optimization and multi-source data fusion.
[0041] Step 300: Based on the initial 3D model, a pseudo-3D inversion is performed using a nonlinear conjugate gradient algorithm, and the values of the horizontal and vertical constraint factors are adaptively adjusted during the inversion process to obtain inversion data; the pseudo-3D inversion process includes objective function enhancement and algorithm acceleration strategies.
[0042] Step 400: Extract three-dimensional imaging information of underground electrical structure based on the inversion data; the three-dimensional imaging information is used for geological interpretation and resource assessment.
[0043] As a specific implementation method, the following embodiment is provided.
[0044] In this embodiment, the interpretation accuracy of transient electromagnetic data and the three-dimensional imaging capability of underground structures are improved by constructing more refined models and using more efficient inversion algorithms.
[0045] 1. Establish the initial 3D model using Gaussian interpolation.
[0046] Traditional initial model construction often employs simple uniform distribution or linear interpolation, which fails to accurately reflect the true state of complex underground electrical structures. This invention innovatively introduces the Gaussian interpolation method, utilizing known geological information and electrical parameters to perform more refined division and assignment of underground space, constructing an initial 3D model that more closely approximates reality. The Gaussian interpolation method offers advantages such as smooth transitions, high accuracy, and strong adaptability, significantly improving the accuracy and rationality of the initial model and laying a solid foundation for subsequent inversion calculations.
[0047] Kernel function optimization:
[0048] ;
[0049] in, Represents a kernel function, which measures the difference between two points in space. r i and r j The correlation, This represents the bandwidth parameter of the kernel function. The adaptive adjustment coefficient (0.3-0.7) increases with increasing formation complexity. The spacing between measuring points is typically 20-50m.
[0050] Multi-source data fusion:
[0051] By integrating existing borehole resistivity ρdill and surface geological maps, a hybrid constraint matrix is constructed:
[0052] Integrating existing borehole resistivity ρ dill Surface geological map Construct a hybrid constraint matrix:
[0053] ;
[0054] in, Represents the kernel function. This represents the weighting coefficient, which balances the contributions of borehole and surface geological data.
[0055] Thus, the correlation coefficient of the initial model can be improved from 0.35 using the traditional linear interpolation method to 0.82 using this method, as follows: Figure 1 As shown.
[0056] 2. Performing quasi-3D inversion calculations using the nonlinear conjugate gradient algorithm.
[0057] Inversion calculation is a core step in the interpretation of transient electromagnetic data. Addressing the nonlinearity, multiple solutions, and high computational cost of transient electromagnetic data under short offset conditions, this invention employs a nonlinear conjugate gradient algorithm for pseudo-3D inversion calculation. This algorithm boasts advantages such as fast convergence speed, high computational efficiency, and strong global search capability, effectively reducing computation time and resource consumption while maintaining inversion accuracy. Through iterative optimization of the nonlinear conjugate gradient algorithm, the actual underground electrical structure is gradually approximated, achieving the pseudo-3D inversion objective.
[0058] Objective function enhancement:
[0059] ;
[0060] Among them, W d Let m be the data weight matrix, m be the model vector to be determined, and d be the weight matrix. obs For the observed data vector, d pre To predict data vectors, This is the horizontal regularization parameter. For the horizontal gradient operator, For the vertical first derivative operator, This is the regularization parameter in the vertical direction.
[0061] Algorithm acceleration strategies:
[0062] GPU-parallel Jacobian matrix computation is employed, and preconditioners are introduced: The iteration count is less than 50; where J is the Jacobian matrix. Let I be a small positive number, and let I be the identity matrix.
[0063] 3. Adaptively adjust horizontal and vertical constraint factors during the inversion process.
[0064] In the inversion process, the selection of lateral and vertical constraint factors has a significant impact on the inversion results. Traditional fixed constraint factor methods are difficult to adapt to the inversion requirements under different geological conditions. This invention proposes an adaptive method for adjusting lateral and vertical constraint factors. Based on the iterative results and data fitting degree during the inversion process, the values of the constraint factors are dynamically adjusted to balance the accuracy and stability of the inversion results. By adaptively adjusting the constraint factors, it is possible to more flexibly address changes in complex underground electrical structures, improving the reliability and practicality of the inversion results. (See Table 1 for details.)
[0065] Table 1 Dynamic Adaptive Constraint Adjustment
[0066]
[0067] 4. Quantification of Uncertainty
[0068] The uncertainty of the inversion results is quantitatively assessed using posterior covariance analysis.
[0069] ;
[0070] In the formula, The residual between the measured value and the model prediction is denoted as . Let W be the posterior covariance matrix, J be the Jacobian matrix, and W be the posterior covariance matrix. d For the data weight matrix, Here, L is the regularization parameter, and L is the prior constraint matrix.
[0071] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0072] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. A method for short-offset transient electromagnetic pseudo-three-dimensional inversion of an electrical source, characterized in that, include: Acquire multi-source data and normalize the multi-source data to obtain preprocessed data; The multi-source data includes geological information and electrical parameters; Based on the preprocessed data, the underground space is divided and assigned values using the Gaussian interpolation method to construct an initial three-dimensional model; The initial 3D model construction process includes kernel function optimization and multi-source data fusion; Based on the initial three-dimensional model, a pseudo-three-dimensional inversion is performed using a nonlinear conjugate gradient algorithm, and the values of the horizontal and vertical constraint factors are adaptively adjusted during the inversion process to obtain inversion data. The pseudo-3D inversion process includes objective function enhancement and algorithm acceleration strategies; Three-dimensional imaging information of the underground electrical structure is extracted based on the inversion data; The three-dimensional imaging information is used for geological interpretation and resource assessment.
2. The method for short-offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources according to claim 1, characterized in that, The kernel function optimization specifically involves: in, Represents a kernel function, which measures the difference between two points in space. r i and r j The correlation, This represents the bandwidth parameter of the kernel function. This represents the adaptive adjustment coefficient. Indicates the distance between measuring points.
3. The method for short-offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources according to claim 1, characterized in that, The multi-source data fusion specifically refers to: Integrating existing borehole resistivity ρ dill Surface geological map Construct a hybrid constraint matrix: in, Represents the kernel function. This represents the weighting coefficient.
4. The method for short-offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources according to claim 1, characterized in that, The enhancement of the objective function specifically includes: Among them, W d Let m be the data weight matrix, m be the model vector to be determined, and d be the weight matrix. obs For the observed data vector, d pre To predict data vectors, This is the horizontal regularization parameter. For the horizontal gradient operator, For the vertical first derivative operator, This is the regularization parameter in the vertical direction.
5. The method for short-offset transient electromagnetic pseudo-three-dimensional inversion of electrical sources according to claim 1, characterized in that, The algorithm acceleration strategy is specifically as follows: GPU-parallel Jacobian matrix computation is employed, and preconditioners are introduced: The iteration count is less than 50; where J is the Jacobian matrix. Let I be a small positive number, and let I be the identity matrix.
6. A short-offset transient electromagnetic pseudo-three-dimensional inversion system with an electrical source, characterized in that, include: The data acquisition and preprocessing unit is used to acquire multi-source data and normalize the multi-source data to obtain preprocessed data. The multi-source data includes geological information and electrical parameters; The initial 3D model construction unit is used to divide and assign values to the underground space based on the preprocessed data using the Gaussian interpolation method, and construct the initial 3D model. The initial 3D model construction process includes kernel function optimization and multi-source data fusion; The pseudo-3D inversion unit is used to perform pseudo-3D inversion based on the initial 3D model using a nonlinear conjugate gradient algorithm, and adaptively adjust the values of the horizontal and vertical constraint factors during the inversion process to obtain inversion data. The pseudo-3D inversion process includes objective function enhancement and algorithm acceleration strategies; An imaging unit is used to extract three-dimensional imaging information of underground electrical structures based on the inversion data; The three-dimensional imaging information is used for geological interpretation and resource assessment.
7. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to enable the electronic device to perform the short-offset transient electromagnetic pseudo-three-dimensional inversion method for electrical sources according to any one of claims 1-5.
8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the short-offset transient electromagnetic pseudo-three-dimensional inversion method for electrical sources as described in any one of claims 1-5.