A data processing method and system for improving the accuracy of magnetotelluric exploration

Abstract

This invention relates to the field of data processing, specifically to a data processing method and system for improving the accuracy of magnetotelluric exploration. The method includes the following steps: collecting electromagnetic observation data and formation logging data from the target area; establishing apparent resistivity curves for each measuring point in the target area under different frequency conditions; determining the resistivity value range corresponding to each lithology; and generating a lithology frequency band template library; analyzing the apparent resistivity curves based on frequency variations to obtain noise-sensitive frequency bands and constructing a spectrum perturbation table; analyzing the spectrum perturbation table to generate a frequency weight mapping matrix; analyzing the apparent resistivity curves using the frequency weight mapping matrix to generate fitted curves and a curve segment structure table; analyzing the fitted curves and curve segment structure table using the lithology frequency band template library to generate a curve segment interpretation structure set; and analyzing adjacent curve segments using the curve segment interpretation structure set to construct an electrical geological boundary matrix. This invention can improve the accuracy of lithology identification.

Description

Technical Field

[0001] This invention relates to the field of data processing, and specifically to a data processing method and system for improving the accuracy of magnetotelluric exploration. Background Technology

[0002] Magnetotelluric (MT) is a non-source electromagnetic detection method widely used in oil and gas, shale gas, geothermal, and deep structural exploration. It boasts advantages such as large detection depth, wide applicability to various terrains, and strong penetration capability through high-resistivity overburden. With the increasing demand for deep resource exploration and the intensified development of shale gas and deep oil and gas resources, the accuracy and interpretation capability of MMT data have become key factors affecting exploration success.

[0003] In existing technologies, in multi-source noise interference scenarios, especially in mountainous areas, industrial and mining areas, and areas with dense traffic interference sources, strong disturbances are prone to occur in high-frequency or specific frequency bands of the apparent resistivity curve. Conventional filtering strategies cannot effectively isolate frequency domain anomalies, causing abnormal frequency bands to mislead structure identification during inversion. In areas with complex lithological variations, the differences in resistivity response of different lithologies are not reflected in the early stages of data processing, resulting in data interpretation relying on human experience and poor stability. This limits the application of magnetotelluric methods in refined resource assessment, which is a problem we need to solve. Summary of the Invention

[0004] The purpose of this invention is to address the problems existing in the background technology by proposing a data processing method and system to improve the accuracy of magnetotelluric exploration.

[0005] The technical solution of this invention: A data processing method for improving the accuracy of magnetotelluric exploration, comprising the following steps: S1. Collect electromagnetic observation data and formation logging data of the target area, and establish the apparent resistivity curves of each measuring point in the target area under different frequency conditions; determine the resistivity range of each lithology based on the formation logging data of each lithology; and generate a lithology frequency band template library. S2. Analyze the apparent resistivity curve based on frequency changes to obtain the noise-sensitive frequency band and construct a spectrum perturbation table; analyze the spectrum perturbation table to generate a frequency weighting mapping matrix; S3. Combine the frequency weight mapping matrix to identify the structural segments of the apparent resistivity curve and perform fitting and lithology matching to generate the fitted curve and curve segment structure table; combine the lithology frequency band template library to analyze the fitted curve and curve segment structure table to generate the curve segment interpretation structure set. S4. Combine the curve segment interpretation structure set to analyze adjacent curve segments and construct the electrical geological boundary matrix.

[0006] Preferably, the process of collecting electromagnetic observation data and formation logging data of the target area, establishing apparent resistivity curves for each measuring point in the target area under different frequency conditions, determining the resistivity range corresponding to each lithology based on the formation logging data of each lithology, and generating a lithology frequency band template library includes: Electromagnetic observation data includes electric field observation data and magnetic field observation data; formation logging data includes formation resistivity, porosity, water saturation, and formation water resistivity; frequency domain transformation is performed on the collected electric field and magnetic field observation data to obtain the apparent resistivity and phase angle of each measuring point at multiple observation frequencies; based on the apparent resistivity obtained at multiple observation frequencies, apparent resistivity curves of each measuring point in the target area are plotted, and phase curves are constructed based on the phase angles, generating apparent resistivity curves and phase response curves of each measuring point in the target area under different frequency conditions; the apparent resistivity and phase angles are processed to construct an apparent resistivity phase sequence; lithology is classified according to the formation logging data, and the resistivity value range of each lithology is determined based on the porosity, water saturation, and formation water resistivity; The rate of change, direction of change, and phase inflection points of the apparent resistivity phase sequence in different frequency bands are statistically analyzed to extract the electrical variation characteristics within the frequency bands. The electrical variation characteristics are then analyzed in relation to the resistivity range corresponding to each lithology to determine the apparent resistivity variation characteristics, phase variation characteristics, and waveform characteristic parameters of each lithology under different frequency conditions. Based on the corresponding lithology category identifier and frequency interval identifier, lithology frequency band templates are constructed, and multiple lithology frequency band templates are combined to generate a lithology frequency band template library.

[0007] Preferably, the process of analyzing the apparent resistivity curve based on frequency changes to obtain the noise-sensitive frequency band and constructing a spectrum perturbation table includes: The preprocessed apparent resistivity curve is processed by Fast Fourier Transform to extract frequency domain feature values; based on the frequency sequence curve, the perturbation rate is calculated at each frequency point f, and the signal-to-noise ratio value at the current frequency point f is obtained, and the noise perturbation coefficient corresponding to the frequency point is calculated. The noise disturbance coefficient corresponding to each frequency point is compared with the preset noise disturbance threshold. If the noise disturbance coefficient is greater than the preset noise disturbance threshold, the corresponding frequency point is classified as a noise-sensitive frequency band. A disturbance mark status is added to the noise-sensitive frequency band, and the disturbance mark status is recorded in the spectrum disturbance table. The spectrum disturbance table includes the frequency point number, disturbance rate, signal-to-noise ratio, noise disturbance coefficient, and disturbance mark status.

[0008] Preferably, the process of analyzing the spectral perturbation table and generating the frequency weight mapping matrix is ​​as follows: For each frequency point in the spectrum disturbance table, calculate the frequency point weighting coefficient based on the corresponding noise disturbance coefficient; according to the frequency point weighting coefficient, divide the frequency point into weight level segments, including high weight segment, medium weight segment and low weight segment, and combine each weight segment in frequency order to form a frequency weight mapping matrix; associate the frequency weight mapping matrix with the corresponding apparent resistivity and phase angle.

[0009] Preferably, the process of identifying structural segments of the apparent resistivity curve by combining the frequency weight mapping matrix, fitting and matching the segments with lithology, and generating the fitted curve and the curve segment structure table includes: The first and second derivatives of the apparent resistivity curve after weighting by the frequency weight mapping matrix are calculated to obtain the curve rate of change sequence and the curve curvature sequence. Based on the sign change of the first derivative and the local extrema of the second derivative, the structural inflection point of the curve is determined, and the apparent resistivity curve is divided into multiple curve segments with each inflection point as the boundary. For each curve segment, the frequency range, apparent resistivity value sequence, and phase value sequence within the curve segment are extracted to construct the initial fitting dataset for the curve segment. In each curve segment, a weighted fitting objective function is established based on the weighting coefficients. The weighted fitting objective function is input into the fitting algorithm to solve for the fitting coefficients, thereby obtaining the fitting curve for each curve segment. The fitting residual, fitting curve shape, apparent resistivity jump amplitude, and phase change trend of each curve segment are used as explanatory features to generate a curve segment structure table.

[0010] Preferably, the process of analyzing the fitted curves and curve segment structure tables in conjunction with the lithological frequency band template library to generate a set of curve segment interpretation structures includes: The fitting parameters, interpretation features, and apparent resistivity variation features of each curve segment are input into the lithology frequency band template library. The similarity between the fitted curve of the curve segment and each lithology frequency band template in the template library is calculated. The best matching lithology template is selected according to the similarity value, and the lithology label corresponding to the best matching lithology template is assigned to the corresponding curve segment to obtain the curve segment lithology matching factor. The lithology matching factor, fitting residual, apparent resistivity abrupt change position, and phase abrupt change position are combined to generate the curve segment interpretation structure set.

[0011] Preferably, the process of analyzing adjacent curve segments and constructing an electrical geological boundary matrix by combining the curve segment interpretation structure set includes: Cluster analysis is performed on the curve segment interpretation structure set to identify apparent resistivity abrupt change points, phase jump points, and lithological matching factor variation amplitudes between adjacent curve segments. The analysis is then performed to generate first-level and second-level boundary candidates. The boundary positions between adjacent curve segments that meet the first-level or second-level boundary candidate conditions are denoted as the electrical abrupt change point set. Based on the set of electrical abrupt change points and the corresponding lithological change categories, a two-dimensional boundary label matrix is ​​constructed. Based on the boundary label matrix, the two-dimensional boundary label matrix is ​​smoothed by interpolation, and the missing boundary points are supplemented by the third-order spline interpolation method to generate a complete electrical geological boundary matrix.

[0012] This invention also discloses a data processing system for improving the accuracy of magnetotelluric exploration, comprising a management center, wherein the management center is communicatively connected to a data acquisition module, a data analysis module, a data matching module, and a data processing module. The data acquisition module is used to collect electromagnetic observation data and formation logging data of the target area, establish the apparent resistivity curves of each measuring point in the target area under different frequency conditions; determine the resistivity range of each lithology based on the formation logging data of each lithology; and generate a lithology frequency band template library. The data analysis module is used to analyze the apparent resistivity curve based on frequency changes, obtain the noise-sensitive frequency band, and construct a spectrum perturbation table; it then analyzes the spectrum perturbation table to generate a frequency weighting mapping matrix. The data matching module is used to identify and fit the apparent resistivity curve by combining the frequency weight mapping matrix, and generate the fitted curve and curve segment structure table. The fitted curve and curve segment structure table are analyzed by combining the lithology frequency band template library to generate the curve segment interpretation structure set. The data processing module is used to analyze adjacent curve segments by combining the curve segment interpretation structure set and construct the electrical geological boundary matrix.

[0013] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects: by constructing an apparent resistivity phase sequence and generating a lithological frequency band template library based on formation logging data, the lithological response constraint of the apparent resistivity curve is realized, so that the subsequent curve interpretation has a parameterized lithological reference basis, avoiding the multiple solutions generated by the apparent resistivity curve under the condition of no lithological boundary constraint; by constructing a spectrum perturbation table and generating a frequency weight mapping matrix, and combining the weight information to identify the structural segments of the curve and match the lithology, dynamic weighting processing based on frequency perturbation characteristics is realized, so that a one-to-one interpretation link is formed between the curve segment boundary, the fitted curve and the lithological category, providing quantifiable electrical abrupt change information support for constructing the electrical geological boundary matrix. Attached Figure Description

[0014] Figure 1 This is a flowchart of one embodiment of the present invention. Detailed Implementation

[0015] Example 1, as Figure 1 As shown, the present invention proposes a data processing method to improve the accuracy of magnetotelluric exploration, comprising the following steps: S1. Collect electromagnetic observation data and formation logging data of the target area, and establish the apparent resistivity curves of each measuring point in the target area under different frequency conditions; determine the resistivity range of each lithology based on the formation logging data of each lithology; and generate a lithology frequency band template library. S2. Analyze the apparent resistivity curve based on frequency changes to obtain the noise-sensitive frequency band and construct a spectrum perturbation table; analyze the spectrum perturbation table to generate a frequency weighting mapping matrix; S3. Combine the frequency weight mapping matrix to identify the structural segments of the apparent resistivity curve and perform fitting and lithology matching to generate the fitted curve and curve segment structure table; combine the lithology frequency band template library to analyze the fitted curve and curve segment structure table to generate the curve segment interpretation structure set. S4. Combine the curve segment interpretation structure set to analyze adjacent curve segments and construct the electrical geological boundary matrix.

[0016] It should be further explained that, in the specific implementation process, the following steps are taken: collecting electromagnetic observation data and formation logging data of the target area; establishing apparent resistivity curves for each measuring point in the target area under different frequency conditions; determining the resistivity range for each lithology based on the formation logging data; and generating a lithology frequency band template library. The electromagnetic observation data includes electric field observation data and magnetic field observation data; the electric field observation data includes electric field amplitude and phase information at multiple measuring points at different frequencies; the magnetic field observation data includes magnetic field amplitude and phase information at corresponding frequencies. The formation logging data refers to multiple sets of logging parameter data obtained based on boreholes or logging profiles already deployed in the target area, including but not limited to formation resistivity, porosity, water saturation, and formation water resistivity; the formation resistivity is used to reflect the lithological correspondence with the electromagnetic observation data, and the porosity, water saturation, and formation water resistivity are used to invert the true formation resistivity according to Archie's law; Frequency domain transformation was performed on the collected electric field and magnetic field observation data to obtain the apparent resistivity of each measuring point at multiple observation frequencies. With phase angle According to the formula for calculating apparent resistivity and phase in magnetotelluric sounding: ; ; in, This refers to the vacuum permeability; This refers to angular frequency; This refers to electric field observation data; This refers to magnetic field observation data; Apparent resistivity obtained at multiple observation frequencies Plot the apparent resistivity curves at each measuring point in the target area, and based on the phase angle... Construct phase curves to generate apparent resistivity curves and phase response curves for each measurement point in the target area under different frequency conditions; The apparent resistivity and phase angle are processed to construct an apparent resistivity phase sequence; Based on formation logging data, lithology is classified, and according to Archie's law, the resistivity range of each lithology is determined based on its porosity, water saturation, and formation water resistivity. ; in, This refers to the formation resistivity of various lithologies; It refers to the rock structure coefficient, which represents the complexity of the rock's pore structure and is determined by actual measurements; This refers to the resistivity of formation water; "Porosity" refers to porosity; "m" refers to the porosity index, which indicates the influence of the rock skeleton structure on current conduction. This refers to water saturation; n refers to the saturation index. The rate of change, direction of change, and phase inflection points of the apparent resistivity phase sequence in different frequency bands are statistically analyzed to extract the electrical variation characteristics within the frequency bands. The electrical variation characteristics are compared with the resistivity value range corresponding to each lithology in a frequency band-by-frequency analysis to determine the apparent resistivity variation characteristics, phase variation characteristics, and waveform characteristic parameters of each lithology under different frequency conditions. Based on the corresponding lithology category identifier and frequency interval identifier, lithology frequency band templates are constructed, and multiple lithology frequency band templates are combined to generate a lithology frequency band template library.

[0017] It should be further explained that, in the specific implementation process, the apparent resistivity curve is analyzed based on frequency changes to obtain the noise-sensitive frequency band, and a spectrum perturbation table is constructed; the process of analyzing the spectrum perturbation table and generating the frequency weighting mapping matrix is ​​as follows: The pre-processed apparent resistivity curve Perform Fast Fourier Transform (FFT) processing to extract frequency domain feature values; based on the frequency sequence curve, calculate the perturbation rate D(f) at each frequency point f, where the perturbation rate is defined as: ; in, and These are the apparent resistivity values ​​of the adjacent frequencies to the current frequency. Let f be the apparent resistivity at the current frequency; and obtain the signal-to-noise ratio (SNR) (f) at the current frequency, then calculate the noise disturbance coefficient N (f) corresponding to the frequency, using the following formula: ; The noise disturbance coefficient corresponding to each frequency point is compared with the preset noise disturbance threshold. Compare, if N(f) is greater than If the corresponding frequency point is classified as a noise-sensitive frequency band, a disturbance mark status is added to the noise-sensitive frequency band, and the disturbance mark status is recorded in the spectrum disturbance table, which includes the frequency point number, disturbance rate D(f), signal-to-noise ratio SNR(f), noise disturbance coefficient N(f), and disturbance mark status.

[0018] For each frequency point in the spectral perturbation table, calculate the frequency point weighting coefficient based on the corresponding noise perturbation coefficient N(f). Frequency weighting coefficient The calculation method is as follows: ; in, This represents the upper limit of the noise disturbance coefficient amplitude. The minimum weighting coefficient threshold is used; based on the frequency point weighting coefficient, the frequency points are divided into weight level segments, including high weight segment, medium weight segment and low weight segment, and each weight segment is combined into a frequency weight mapping matrix in frequency order; the frequency weight mapping matrix is ​​associated with the corresponding apparent resistivity and phase angle. After the weight assignment is completed, a low-order fitting model is used for noise-sensitive frequency bands based on the weight distribution results, and a high-order fitting model is used for non-noise-sensitive frequency bands. The fitting model is used to constrain the data interpretation process in the curve structure segmentation step and to ensure the continuity and segmentability of each frequency band curve segment.

[0019] It should be further explained that, in the specific implementation process, the apparent resistivity curve is segmented and identified using the frequency weight mapping matrix, and then fitted and matched with lithology to generate the fitted curve and the curve segment structure table; the fitted curve and the curve segment structure table are analyzed using the lithology frequency band template library to generate the curve segment interpretation structure set. The first and second derivatives of the apparent resistivity curve after weighting by the frequency weight mapping matrix are calculated to obtain the curve rate of change sequence and the curve curvature sequence. Based on the sign change of the first derivative and the local extrema of the second derivative, the structural inflection points of the curve are determined, and the apparent resistivity curve is divided into multiple curve segments using each inflection point as a boundary. For each curve segment, the frequency range, apparent resistivity value sequence, and phase value sequence are extracted to construct an initial fitting dataset for the curve segment. Within each curve segment, weighting coefficients are used to... Establish a weighted fitting objective function and select a fitting formula according to the preset fitting model: ; The weighted fitting objective function is input into the fitting algorithm to solve for the fitting coefficients A, B, and C, resulting in the fitted curve for each curve segment. The fitting residual, fitted curve shape, apparent resistivity jump amplitude, and phase change trend of each curve segment are used as explanatory features to generate a curve segment structure table. The curve segment structure table includes the curve segment number, frequency range, derivative rate of change, curvature information, fitting parameters, residual value, and curve change trend marker.

[0020] Input the fitting parameters, interpretation features, and apparent resistivity variation features of each curve segment into the lithology frequency band template library; The similarity between the fitted curve based on the curve segment and the template library for each lithological frequency band is calculated. The similarity is calculated using vector cosine similarity or correlation coefficient. ; in, This is the fitted value sequence for the curve segment. This is the reference response sequence for the corresponding lithological template; Based on the similarity value, the best matching lithology template is selected, and the lithology label corresponding to the best matching lithology template is assigned to the corresponding curve segment to obtain the curve segment lithology matching factor. The lithological matching factor, fitting residual, apparent resistivity abrupt change location, and phase abrupt change location are combined to generate a curve segment interpretation structure set, which includes curve segment number, lithological matching factor, suspected geological boundary location, apparent resistivity change category, and phase change category. Specifically, the apparent resistivity change categories include abrupt increases, abrupt decreases, and gradual changes; the phase change categories include abrupt increases, abrupt decreases, and gradual changes.

[0021] It should be further explained that, in the specific implementation process, the process of analyzing adjacent curve segments in conjunction with the curve segment interpretation structure set to construct the electrical geological boundary matrix is ​​as follows: Cluster analysis was performed on the interpreted structure set of curve segments to identify abrupt changes in apparent resistivity, phase jumps, and variations in lithological matching factors between adjacent curve segments. Suspected geological interfaces were then identified based on the following combination rules: When the apparent resistivity fitting residual of adjacent curve segments exceeds the set threshold, and the apparent resistivity change amplitude is greater than the set apparent resistivity change amplitude threshold, it is determined to be a first-level boundary candidate. When the change value of the phase slope is greater than the set phase slope change threshold, and the decrease value of the lithology matching factor is greater than the lithology matching factor change threshold, it is determined to be a secondary boundary candidate; The boundary positions between adjacent curve segments that satisfy the first-level or second-level boundary candidate conditions are denoted as the electrical abrupt change point set; Based on the set of electrical abrupt change points and the corresponding lithological change categories, a two-dimensional boundary label matrix B(i,j) is constructed. Each row of the matrix represents the survey line number i, and each column represents the survey point number j. B(i,j)=1: indicates the existence of a first-order apparent resistivity boundary; B(i,j)=2: indicates the existence of a phase abrupt change boundary; B(i,j)=3: indicates an uncertain lithological boundary; B(i,j)=0: indicates no boundary. Based on the boundary label matrix B(i,j) and the frequency weight mapping matrix, the two-dimensional boundary label matrix is ​​smoothed by interpolation, and the missing boundary points are filled by the third-order spline interpolation method to generate the complete electrical geological boundary matrix M(i,j).

[0022] Example 2: The data processing system for improving the accuracy of magnetotelluric exploration proposed in this invention is applied to the data processing method for improving the accuracy of magnetotelluric exploration described in Example 1. Specifically, it includes a management center, which is communicatively connected to a data acquisition module, a data analysis module, a data matching module, and a data processing module. The data acquisition module is used to collect electromagnetic observation data and formation logging data of the target area, establish the apparent resistivity curves of each measuring point in the target area under different frequency conditions; determine the resistivity range of each lithology based on the formation logging data of each lithology; and generate a lithology frequency band template library. The data analysis module is used to analyze the apparent resistivity curve based on frequency changes, obtain the noise-sensitive frequency band, and construct a spectrum perturbation table; it then analyzes the spectrum perturbation table to generate a frequency weighting mapping matrix. The data matching module is used to identify and fit the apparent resistivity curve by combining the frequency weight mapping matrix, and generate the fitted curve and curve segment structure table. The fitted curve and curve segment structure table are analyzed by combining the lithology frequency band template library to generate the curve segment interpretation structure set. The data processing module is used to analyze adjacent curve segments by combining the curve segment interpretation structure set and construct the electrical geological boundary matrix.

[0023] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A data processing method for improving the accuracy of magnetotelluric exploration, characterized in that, Includes the following steps: S1. Collect electromagnetic observation data and formation logging data of the target area, and establish the apparent resistivity curves of each measuring point in the target area under different frequency conditions; determine the resistivity range of each lithology based on the formation logging data of each lithology; and generate a lithology frequency band template library. S2. Analyze the apparent resistivity curve based on frequency changes to obtain the noise-sensitive frequency band and construct a spectrum perturbation table; analyze the spectrum perturbation table to generate a frequency weighting mapping matrix; S3. Combine the frequency weight mapping matrix to identify the structural segments of the apparent resistivity curve and perform fitting and lithology matching to generate the fitted curve and curve segment structure table; combine the lithology frequency band template library to analyze the fitted curve and curve segment structure table to generate the curve segment interpretation structure set. S4. Combine the curve segment interpretation structure set to analyze adjacent curve segments and construct the electrical geological boundary matrix.

2. The data processing method for improving the accuracy of magnetotelluric exploration according to claim 1, characterized in that, Electromagnetic observation data and formation logging data of the target area were collected, and apparent resistivity curves of each measuring point in the target area under different frequency conditions were established; based on the formation logging data of each lithology, the resistivity range corresponding to each lithology was determined. The process of generating a lithology frequency band template library includes: Electromagnetic observation data includes electric field observation data and magnetic field observation data; formation logging data includes formation resistivity, porosity, water saturation, and formation water resistivity; frequency domain transformation is performed on the collected electric field and magnetic field observation data to obtain the apparent resistivity and phase angle of each measuring point at multiple observation frequencies; based on the apparent resistivity obtained at multiple observation frequencies, apparent resistivity curves of each measuring point in the target area are plotted, and phase curves are constructed based on the phase angles, generating apparent resistivity curves and phase response curves of each measuring point in the target area under different frequency conditions; the apparent resistivity and phase angles are processed to construct an apparent resistivity phase sequence; lithology is classified according to the formation logging data, and the resistivity value range of each lithology is determined based on the porosity, water saturation, and formation water resistivity; The rate of change, direction of change, and phase inflection points of the apparent resistivity phase sequence in different frequency bands are statistically analyzed to extract the electrical variation characteristics within the frequency bands. The electrical variation characteristics are then analyzed in relation to the resistivity range corresponding to each lithology to determine the apparent resistivity variation characteristics, phase variation characteristics, and waveform characteristic parameters of each lithology under different frequency conditions. Based on the corresponding lithology category identifier and frequency interval identifier, lithology frequency band templates are constructed, and multiple lithology frequency band templates are combined to generate a lithology frequency band template library.

3. The data processing method for improving the accuracy of magnetotelluric exploration according to claim 2, characterized in that, The process of analyzing the apparent resistivity curve based on frequency changes to obtain the noise-sensitive frequency band and constructing a spectrum perturbation table includes: The preprocessed apparent resistivity curve is processed by Fast Fourier Transform to extract frequency domain feature values; based on the frequency sequence curve, the perturbation rate is calculated at each frequency point f, and the signal-to-noise ratio value at the current frequency point f is obtained, and the noise perturbation coefficient corresponding to the frequency point is calculated. The noise disturbance coefficient corresponding to each frequency point is compared with the preset noise disturbance threshold. If the noise disturbance coefficient is greater than the preset noise disturbance threshold, the corresponding frequency point is classified as a noise-sensitive frequency band. A disturbance mark status is added to the noise-sensitive frequency band, and the disturbance mark status is recorded in the spectrum disturbance table. The spectrum disturbance table includes the frequency point number, disturbance rate, signal-to-noise ratio, noise disturbance coefficient, and disturbance mark status.

4. The data processing method for improving the accuracy of magnetotelluric exploration according to claim 3, characterized in that, The process of analyzing the spectral perturbation table and generating the frequency weight mapping matrix is ​​as follows: For each frequency point in the spectrum disturbance table, calculate the frequency point weighting coefficient based on the corresponding noise disturbance coefficient; according to the frequency point weighting coefficient, divide the frequency point into weight level segments, including high weight segment, medium weight segment and low weight segment, and combine each weight segment in frequency order to form a frequency weight mapping matrix; associate the frequency weight mapping matrix with the corresponding apparent resistivity and phase angle.

5. A data processing method for improving the accuracy of magnetotelluric exploration according to claim 4, characterized in that, The process of identifying structural segments of the apparent resistivity curve by combining the frequency weighting mapping matrix, fitting and matching the segments with lithology, and generating the fitted curve and the curve segment structure table includes: The first and second derivatives of the apparent resistivity curve after weighting by the frequency weight mapping matrix are calculated to obtain the curve rate of change sequence and the curve curvature sequence. Based on the sign change of the first derivative and the local extrema of the second derivative, the structural inflection point of the curve is determined, and the apparent resistivity curve is divided into multiple curve segments with each inflection point as the boundary. For each curve segment, the frequency range, apparent resistivity value sequence, and phase value sequence within the curve segment are extracted to construct the initial fitting dataset for the curve segment. In each curve segment, a weighted fitting objective function is established based on the weighting coefficients. The weighted fitting objective function is input into the fitting algorithm to solve for the fitting coefficients, thereby obtaining the fitting curve for each curve segment. The fitting residual, fitting curve shape, apparent resistivity jump amplitude, and phase change trend of each curve segment are used as explanatory features to generate a curve segment structure table.

6. The data processing method for improving the accuracy of magnetotelluric exploration according to claim 5, characterized in that, The process of analyzing the fitted curves and curve segment structure tables using a lithological frequency band template library to generate a set of curve segment interpretation structures includes: The fitting parameters, interpretation features, and apparent resistivity variation features of each curve segment are input into the lithology frequency band template library. The similarity between the fitted curve of the curve segment and each lithology frequency band template in the template library is calculated. The best matching lithology template is selected according to the similarity value, and the lithology label corresponding to the best matching lithology template is assigned to the corresponding curve segment to obtain the curve segment lithology matching factor. The lithology matching factor, fitting residual, apparent resistivity abrupt change position, and phase abrupt change position are combined to generate the curve segment interpretation structure set.

7. A data processing method for improving the accuracy of magnetotelluric exploration according to claim 6, characterized in that, The process of constructing the electrical geological boundary matrix by analyzing adjacent curve segments using the curve segment interpretation structure set includes: Cluster analysis is performed on the curve segment interpretation structure set to identify apparent resistivity abrupt change points, phase jump points, and lithological matching factor variation amplitudes between adjacent curve segments. The analysis is then performed to generate first-level and second-level boundary candidates. The boundary positions between adjacent curve segments that meet the first-level or second-level boundary candidate conditions are denoted as the electrical abrupt change point set. Based on the set of electrical abrupt change points and the corresponding lithological change categories, a two-dimensional boundary label matrix is ​​constructed. Based on the boundary label matrix, the two-dimensional boundary label matrix is ​​smoothed by interpolation, and the missing boundary points are supplemented by the third-order spline interpolation method to generate a complete electrical geological boundary matrix.

8. A data processing system for improving the accuracy of magnetotelluric exploration, specifically applied to the data processing method for improving the accuracy of magnetotelluric exploration as described in any one of claims 1 to 7, comprising a management center, characterized in that, The management center's communication connections include a data acquisition module, a data analysis module, a data matching module, and a data processing module. The data acquisition module is used to collect electromagnetic observation data and formation logging data of the target area, establish the apparent resistivity curves of each measuring point in the target area under different frequency conditions; determine the resistivity range of each lithology based on the formation logging data of each lithology; and generate a lithology frequency band template library. The data analysis module is used to analyze the apparent resistivity curve based on frequency changes, obtain the noise-sensitive frequency band, and construct a spectrum perturbation table; it then analyzes the spectrum perturbation table to generate a frequency weighting mapping matrix. The data matching module is used to identify and fit the apparent resistivity curve by combining the frequency weight mapping matrix, and generate the fitted curve and curve segment structure table. The fitted curve and curve segment structure table are analyzed by combining the lithology frequency band template library to generate the curve segment interpretation structure set. The data processing module is used to analyze adjacent curve segments by combining the curve segment interpretation structure set and construct the electrical geological boundary matrix.

Images