Controllable source electromagnetic method time-frequency domain three-dimensional forward modeling and feature analysis method

Abstract

The invention relates to the technical field of exploration geophysics. The time-frequency domain three-dimensional forward modeling and feature analysis method based on the controllable source electromagnetic method comprises the steps that electrical parameter measurement processing is conducted on a target area rock ore specimen, and an electrical parameter set is obtained; based on the electrical parameter set, performing electrical category division processing through clustering analysis to generate an electrical classification knowledge base; acquiring an electromagnetic emission signal; performing time domain forward modeling calculation processing to obtain transient response; based on the electromagnetic emission signal and the electrical property classification knowledge base, performing frequency domain forward modeling calculation processing to obtain a specified frequency band response; performing parameter iterative optimization processing to generate a three-dimensional electrical parameter model; performing abnormal region extraction processing based on the three-dimensional electrical parameter model to obtain a target abnormal region; and carrying out boundary delineation processing on the target body to obtain a spatial position of the target body so as to achieve the technical effects of improving the characterization precision of the electrical parameters, reducing the reconstruction distortion of the time-frequency response and enhancing the geological interpretability of an inversion result.

Description

Technical Field

[0001] The present invention relates to the field of exploration geophysical technology, and in particular to a controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method. Background Art

[0002] In the current field of deep mineral resource exploration, wide-area electromagnetic (WAM) has become a core technology for locating concealed ore bodies due to its advantages such as deep detection depth and high resolution. Three-dimensional forward modeling and feature analysis in the time-frequency domain are key to achieving high-precision geological target identification, directly impacting exploration efficiency and resource assessment accuracy.

[0003] However, the relevant technology has the following problems: the traditional electrical parameter mean classification method cannot capture the nonlinear change law of the physical properties of rocks and minerals, resulting in a systematic offset in the basis of forward modeling; the isolated execution of time domain and frequency domain forward calculations causes distortion in the reconstruction of the full time-frequency response, resulting in reduced reliability of feature extraction; the separation of physical property constraints and time-frequency characteristics makes the inversion results lose geological interpretability. Summary of the Invention

[0004] Based on this, it is necessary to provide a three-dimensional forward modeling and characteristic analysis method of controlled source electromagnetic method in the time-frequency domain to address the above technical problems, so as to achieve the technical effects of improving the accuracy of electrical parameter characterization, reducing the distortion of time-frequency response reconstruction, and enhancing the geological interpretability of inversion results.

[0005] In a first aspect, the present application provides a controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method, the method comprising:

[0006] Conduct electrical parameter measurement and processing on rock and mineral samples in the target area to obtain an electrical parameter set; based on the electrical parameter set, perform electrical classification processing through cluster analysis to generate an electrical classification knowledge base;

[0007] Acquire electromagnetic emission signals; perform time-domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain transient responses; perform frequency-domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain responses in designated frequency bands; perform joint extrapolation of the transient responses and responses in designated frequency bands based on orthogonal basis functions to generate a full time-frequency domain response dataset;

[0008] Perform feature analysis on the full time-frequency domain response data set to generate time-frequency feature anomaly results;

[0009] Obtain data fitting items and model constraints; construct physical property constraints based on the electrical property classification knowledge base to obtain physical property constraints; combine physical property constraints, data fitting items, and model constraints to generate a multi-constraint objective function; perform parameter iterative optimization based on the multi-constraint objective function and time-frequency characteristic anomaly results to generate a three-dimensional electrical parameter model;

[0010] Obtain geological data; extract and process the abnormal area based on the three-dimensional electrical parameter model to obtain the target abnormal area; based on the spatial relationship between the target abnormal area and the geological data, delineate the target body boundary to obtain the target body's spatial position.

[0011] Furthermore, based on the electrical parameter set, electrical classification is performed through cluster analysis to generate an electrical classification knowledge base, including:

[0012] Extracting complex resistivity parameters from the electrical parameter set to generate a complex resistivity parameter sample set;

[0013] The complex resistivity parameter sample set is processed by fuzzy clustering algorithm to obtain electrical property categories and determine the cluster center.

[0014] The electrical property categories and cluster centers are stored as an electrical property classification knowledge base.

[0015] Furthermore, the complex resistivity parameter sample set is processed by fuzzy clustering algorithm to divide the electrical property categories and determine the cluster centers, including:

[0016] Perform membership initialization operation on the complex resistivity parameter sample set to generate an initial membership matrix;

[0017] Use the following formula to calculate the cluster center position based on the initial membership matrix to generate the updated cluster center:

[0018]

[0019] Among them, v i Indicates the updated position of the i-th cluster center, u ij represents the membership of the jth data point to the i-th cluster center, m represents the fuzzy factor, x j represents the feature vector of the jth data point, and n represents the total number of data points in the dataset;

[0020] Recalculate the membership matrix based on the updated cluster centers to generate an optimized membership matrix;

[0021] Perform clustering stability verification on the optimized membership matrix. When the convergence condition is met, output the electrical category and cluster center. Otherwise, return to perform cluster center iteration processing.

[0022] Furthermore, based on the electromagnetic emission signal and the electrical classification knowledge base, frequency domain forward modeling is performed to obtain the response of the specified frequency band, including:

[0023] Based on the electromagnetic emission signal and electrical property classification knowledge base, the frequency domain electric field control equation including the Cole-Cole model is constructed;

[0024] The frequency domain electric field governing equations are solved by the vector finite element method to generate the frequency domain electromagnetic response distribution;

[0025] Extract the response components in the specified frequency band from the frequency domain electromagnetic response distribution to generate the specified frequency band response.

[0026] Furthermore, based on the orthogonal basis functions, the transient response and the specified frequency band response are jointly extrapolated to generate a full time-frequency domain response dataset, including:

[0027] Perform time series sampling on transient responses to generate time domain response vectors; perform frequency point sampling on specified frequency band responses to generate frequency domain response vectors;

[0028] The time-frequency conversion relationship between the time domain response vector and the frequency domain response vector is constructed through orthogonal basis functions;

[0029] Perform coefficient calculation operations on the time-frequency conversion relationship based on the least squares fitting criterion to generate extrapolation coefficients;

[0030] The extrapolation coefficients and orthogonal basis functions are used to perform response data extrapolation calculations to generate a full time-frequency domain response data set.

[0031] Furthermore, based on the three-dimensional electrical parameter model, abnormal area extraction processing is performed to obtain the target abnormal area, including:

[0032] Performing feature recognition operations on the resistivity parameter distribution data and the polarizability parameter distribution data in the three-dimensional electrical parameter model to identify resistivity abnormality feature areas and polarizability abnormality feature areas;

[0033] The following formula is used to perform spatial overlay analysis on the resistivity anomaly characteristic area and the polarizability anomaly characteristic area to extract the resistivity-polarizability composite anomaly area:

[0034]

[0035] Among them, S overlap represents the spatial overlap, k represents the total number of discrete grid points, H i represents the volume of the i-th grid cell, C(x i ,y i ,z i ) indicates whether the i-th grid point belongs to the composite anomaly area;

[0036] Perform spatial continuity analysis on the resistivity-polarizability composite anomaly area to generate the target anomaly area.

[0037] Furthermore, based on the spatial relationship between the target anomaly area and the geological data, the boundary of the target body is delineated to obtain the spatial position of the target body, including:

[0038] Perform fault structure identification operations and lithologic boundary identification operations on geological data to generate fault structure elements and lithologic boundary elements;

[0039] Construct a geological constraint model based on the spatial topological relationship between the target abnormal area and the fault structural elements and lithologic boundary elements;

[0040] Perform boundary morphology optimization on the target anomaly area according to the geological constraint model to generate the target body boundary;

[0041] The boundary of the target body is converted into three-dimensional coordinate data to generate the spatial position of the target body.

[0042] In a second aspect, the present application also provides a controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis system, the system comprising:

[0043] The electrical knowledge base construction module is used to measure and process the electrical parameters of rock and mineral samples in the target area to obtain an electrical parameter set; based on the electrical parameter set, electrical classification is performed through cluster analysis to generate an electrical classification knowledge base;

[0044] The time-frequency response reconstruction module is used to obtain electromagnetic emission signals; based on the electromagnetic emission signals and the electrical property classification knowledge base, it performs time-domain forward modeling to obtain transient responses; based on the electromagnetic emission signals and the electrical property classification knowledge base, it performs frequency-domain forward modeling to obtain the responses in the specified frequency band; based on the orthogonal basis functions, it performs joint extrapolation of the transient responses and the responses in the specified frequency band to generate a full time-frequency domain response dataset;

[0045] The time-frequency feature extraction module is used to perform feature analysis on the full time-frequency domain response data set and generate time-frequency feature anomaly results;

[0046] The multi-constraint inversion module is used to obtain data fitting terms and model constraints. Based on the electrical classification knowledge base, physical property constraints are constructed to obtain physical property constraints. Physical constraints, data fitting terms, and model constraints are combined to generate a multi-constraint objective function. Based on the multi-constraint objective function and time-frequency characteristic anomaly results, parameter iterative optimization is performed to generate a three-dimensional electrical parameter model.

[0047] The geological target positioning module is used to obtain geological data; based on the three-dimensional electrical parameter model, it extracts the abnormal area and obtains the target abnormal area; based on the spatial relationship between the target abnormal area and the geological data, it delineates the target body boundary and obtains the target body's spatial position.

[0048] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any method in the first aspect of the present application are implemented.

[0049] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any method in the first aspect of the present application when the computer program is executed by a processor.

[0050] The present application provides a three-dimensional forward modeling and feature analysis method in the time-frequency domain of a controlled source electromagnetic method, including: measuring and processing the electrical parameters of rock and mineral samples in the target area to obtain an electrical parameter set; based on the electrical parameter set, performing electrical classification processing through cluster analysis to generate an electrical classification knowledge base; acquiring an electromagnetic emission signal; performing time-domain forward modeling calculation processing based on the electromagnetic emission signal and the electrical classification knowledge base to obtain a transient response; performing frequency-domain forward modeling calculation processing based on the electromagnetic emission signal and the electrical classification knowledge base to obtain a specified frequency band response; performing joint extrapolation processing on the transient response and the specified frequency band response based on an orthogonal basis function to generate a full time-frequency domain response data set; performing feature analysis processing on the full time-frequency domain response data set to generate a time-frequency feature set. The method comprises the following steps: obtaining anomaly results; obtaining data fitting items and model constraint items; constructing physical property constraint items based on the electrical classification knowledge base to obtain physical property constraint items; combining physical constraint items, data fitting items and model constraint items to generate a multi-constraint objective function; performing iterative parameter optimization based on the multi-constraint objective function and time-frequency characteristic anomaly results to generate a three-dimensional electrical parameter model; obtaining geological data; extracting anomaly areas based on the three-dimensional electrical parameter model to obtain a target anomaly area; delineating the target body boundary based on the spatial relationship between the target anomaly area and the geological data to obtain the spatial position of the target body, so as to achieve the technical effects of improving the accuracy of electrical parameter characterization, reducing the distortion of time-frequency response reconstruction, and enhancing the geological interpretability of the inversion results. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 Flowchart of a controlled source electromagnetic method three-dimensional forward modeling and characteristic analysis method in the time-frequency domain in one embodiment of the present invention;

[0053] Figure 2 A flowchart of generating a full time-frequency domain response data set by performing joint extrapolation processing on transient response and specified frequency band response based on orthogonal basis functions in one embodiment of the present invention;

[0054] Figure 3The structure diagram of the controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis system in one embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the above-mentioned purposes, features and advantages of the present application more clearly understood, the specific implementation methods of the present application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to fully understand the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar improvements without violating the connotation of the application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0056] like Figure 1 As shown, the present application provides a controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method, the method comprising:

[0057] S101: Conduct electrical parameter measurement on rock and mineral samples in the target area to obtain an electrical parameter set; based on the electrical parameter set, perform electrical classification through cluster analysis to generate an electrical classification knowledge base.

[0058] Specifically, electrical parameters of rock and mineral samples within the target area are measured. These parameters include resistivity, polarizability, and other indicators related to the electromagnetic properties of rock. Measuring equipment and standardized experimental procedures ensure the accuracy and reliability of the acquired data. Based on the collected electrical parameter sets, cluster analysis is used to classify the electrical properties into categories.

[0059] Cluster analysis is a statistical method that divides data in a dataset into different categories or clusters. Its purpose is to make data objects within the same category have high similarity, while data objects between different categories have greater differences. In the above steps, the data points in the electrical parameter set are clustered based on the similarity of their electrical characteristics, thereby identifying rock and mineral categories with different electrical properties. This process helps to gain a deeper understanding of the electrical structure distribution of rocks and minerals in the target area, providing basic data support for further geological modeling and resource assessment.

[0060] The electrical property categories and their associated characteristic information derived from the cluster analysis are then integrated and stored to create an electrical classification knowledge base. This knowledge base not only contains the electrical parameter ranges and characteristic descriptions of various rocks and minerals, but also includes relevant geological background information, providing key physical property basis for subsequent wide-field electromagnetic forward modeling and inversion calculations.

[0061] S102: Acquire electromagnetic emission signals; perform time domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain transient responses; perform frequency domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain responses in designated frequency bands; perform joint extrapolation of the transient responses and responses in designated frequency bands based on orthogonal basis functions to generate a full time and frequency domain response dataset.

[0062] Specifically, the electromagnetic emission signal is obtained. This step is the basic data input for the subsequent electromagnetic response calculation. Afterwards, the electromagnetic emission signal obtained is used in combination with the previously established electrical classification knowledge base to carry out time domain forward modeling. In the time domain forward modeling process, the propagation and response of the electromagnetic field in the target area are simulated by numerical simulation methods to obtain transient response results, which reflect the changing characteristics of the electromagnetic field in the time domain. Based on the electromagnetic emission signal and the electrical classification knowledge base, the frequency domain forward modeling is also performed. The frequency domain forward modeling is performed by converting the time domain signal to the frequency domain, or directly performing numerical simulation in the frequency domain to obtain the electromagnetic response within the specified frequency band, that is, the specified frequency band response. This step helps to analyze the characteristics of the electromagnetic field at different frequencies.

[0063] After obtaining the transient response and the response in the specified frequency band, the response information in the two different domains is integrated and jointly extrapolated based on the orthogonal basis function. The orthogonal basis function, as a mathematical tool, is used to establish the connection between the time domain response and the frequency domain response. By finding the common characteristics and change patterns between the two responses, the extrapolation from the known time-frequency response information to the unknown response is achieved. This method generates a full time-frequency domain response dataset, which integrates the electromagnetic response information in the time and frequency domains. It can more comprehensively reflect the electromagnetic characteristics of the target area in the full time-frequency domain, providing a more complete and accurate data foundation for further feature analysis and inversion calculations.

[0064] S103: Perform feature analysis on the full time-frequency domain response data set to generate a time-frequency feature anomaly result.

[0065] Specifically, feature analysis is performed on the full time-frequency domain response dataset. This involves extracting key feature information from both the time and frequency domains, such as rise time, peak time, and decay constant in the time domain, and center frequency, bandwidth, amplitude spectrum, and phase spectrum in the frequency domain. After feature extraction, these features are fused to construct a comprehensive feature vector, thereby more comprehensively characterizing the time-frequency characteristics of the electromagnetic response.

[0066] An anomaly detection algorithm then analyzes the integrated feature vector to identify data points or segments that differ significantly from the background electromagnetic response. These deviations from the normal pattern constitute time-frequency anomalies. The resulting time-frequency anomaly output provides a key basis for subsequent geological target identification and location, enabling more accurate detection of underground geological structures and potential mineral resources.

[0067] S104: Acquire data fitting items and model constraint items; construct physical property constraint items based on the electrical property classification knowledge base to obtain physical property constraint items; combine the physical property constraint items, data fitting items, and model constraint items to generate a multi-constraint objective function; perform parameter iterative optimization based on the multi-constraint objective function and time-frequency characteristic anomaly results to generate a three-dimensional electrical parameter model.

[0068] Specifically, after obtaining the data fitting items and model constraint items, the physical property constraint items are constructed and processed in combination with the electrical property classification knowledge base. This includes establishing physical property constraint conditions that are consistent with the actual geological situation based on the characteristics of different electrical property categories in the electrical property classification knowledge base, and converting the above conditions into physical property constraint items to reflect the differences and distribution patterns of the electrical parameters of rocks and minerals, thereby enhancing the geological rationality of the inversion results. After completing the construction of the physical property constraint items, they are organically combined with the data fitting items and model constraint items to generate a multi-constraint objective function. Among them, the data fitting items are mainly used to ensure the consistency of the inversion results with the measured data; the model constraint items focus on ensuring the smoothness and stability of the model to avoid problems such as overfitting.

[0069] When constructing a multi-constraint objective function, the weights of each constraint are appropriately balanced to ensure that each constraint is effective during the inversion process. Based on this multi-constraint objective function and combined with the results of time-frequency characteristic anomalies, an iterative parameter optimization process is performed. Through continuous adjustment of model parameters, the objective function gradually converges, generating a three-dimensional electrical parameter model that more accurately reflects the electrical characteristics of the underground geological structure, providing more precise geological information for deep mineral resource exploration.

[0070] S105: Acquire geological data; extract the abnormal area based on the three-dimensional electrical parameter model to obtain the target abnormal area; and delineate the target body boundary based on the spatial relationship between the target abnormal area and the geological data to obtain the spatial position of the target body.

[0071] Specifically, after acquiring geological data, the system uses a three-dimensional electrical parameter model as a basis for extracting and processing anomaly regions. This involves performing feature recognition on the resistivity and polarizability parameter distribution data within the three-dimensional electrical parameter model. During this recognition process, image processing techniques and pattern recognition algorithms are used to identify regions with characteristic resistivity and polarizability anomalies. These regions are then spatially overlaid and analyzed to extract composite resistivity-polarizability anomaly regions. Based on this, spatial continuity analysis is further performed on the composite anomaly regions to generate target anomaly regions.

[0072] Based on the spatial relationship between the target anomaly area and the geological data, the target volume boundary is delineated. This involves detailed fault structure and lithologic boundary identification within the geological data, generating fault structural elements and lithologic boundary elements. Furthermore, a geological constraint model is constructed based on the spatial topological relationship between the target anomaly area, the fault structural elements, and the lithologic boundary elements. Using the geological constraint model, the target anomaly area boundary morphology is optimized to generate the target volume boundary. The target volume boundary is then converted into three-dimensional coordinate data to determine the target volume's spatial location. This process fully integrates geological information with the electrical parameter model, completing the complete process from anomaly area extraction to target volume positioning, providing more accurate spatial location information for subsequent geological exploration and resource assessment.

[0073] An embodiment of the present application provides a three-dimensional forward modeling and feature analysis method in the time-frequency domain of a controlled source electromagnetic method, comprising: measuring and processing electrical parameters of rock and mineral samples in a target area to obtain an electrical parameter set; based on the electrical parameter set, performing electrical classification processing through cluster analysis to generate an electrical classification knowledge base; acquiring an electromagnetic emission signal; performing time-domain forward modeling calculation processing based on the electromagnetic emission signal and the electrical classification knowledge base to obtain a transient response; performing frequency-domain forward modeling calculation processing based on the electromagnetic emission signal and the electrical classification knowledge base to obtain a specified frequency band response; performing joint extrapolation processing on the transient response and the specified frequency band response based on an orthogonal basis function to generate a full time-frequency domain response data set; performing feature analysis processing on the full time-frequency domain response data set to generate a time-frequency domain response data set. Frequency characteristic anomaly results; obtain data fitting items and model constraint items; based on the electrical classification knowledge base, perform physical property constraint item construction processing to obtain physical property constraint items; combine physical constraint items, data fitting items and model constraint items to generate a multi-constraint objective function; based on the multi-constraint objective function and time-frequency characteristic anomaly results, perform parameter iterative optimization processing to generate a three-dimensional electrical parameter model; obtain geological data; based on the three-dimensional electrical parameter model, perform anomaly area extraction processing to obtain the target anomaly area; based on the spatial relationship between the target anomaly area and the geological data, perform target body boundary delineation processing to obtain the target body spatial position, so as to achieve the technical effects of improving the electrical parameter characterization accuracy, reducing the time-frequency response reconstruction distortion, and enhancing the geological interpretability of the inversion results.

[0074] In one embodiment, the Cole-Cole model is used to fit the electrical characteristics of the earth medium and construct a geophysical parameter model. Electromagnetic response characteristics are obtained in both high-frequency and low-frequency bands through three-dimensional forward modeling in the time-frequency domain. In the high-frequency band, time-domain forward modeling is used to simulate the transient response of the electromagnetic field and capture the rapidly changing characteristics of the high-frequency electromagnetic signal. In the low-frequency band, frequency-domain forward modeling is used to obtain the low-frequency electromagnetic response and analyze its stability. Laguerre polynomials are used to fit the electromagnetic responses in these two frequency bands, and Fourier transform technology is then used to convert and fuse the time-domain and frequency-domain response data. By jointly calculating the Laguerre polynomial coefficients, a coordinated analysis of the high-frequency response in the time domain and the low-frequency response in the frequency domain is achieved, thereby constructing a relatively complete time-frequency forward response characteristic. This process not only enhances the detection capability of underground geological structures but also provides more accurate and comprehensive basic data support for subsequent electromagnetic data inversion and geological target identification.

[0075] Furthermore, based on the electrical parameter set, electrical classification is performed through cluster analysis to generate an electrical classification knowledge base, including:

[0076] Extracting complex resistivity parameters from the electrical parameter set to generate a complex resistivity parameter sample set;

[0077] The complex resistivity parameter sample set is processed by fuzzy clustering algorithm to obtain electrical property categories and determine the cluster center.

[0078] The electrical property categories and cluster centers are stored as an electrical property classification knowledge base.

[0079] Specifically, complex resistivity parameters are extracted from the electrical parameter set to generate a complex resistivity parameter sample set. Complex resistivity parameters are important indicators for characterizing the electrical characteristics of rocks and minerals. They integrate information on resistivity and polarizability and can more comprehensively reflect the electromagnetic properties of rocks and minerals. Afterwards, the generated complex resistivity parameter sample set is processed by a fuzzy clustering algorithm. The fuzzy clustering algorithm is a soft computing technology that allows data points to belong to multiple clusters but have different degrees of membership, which gives it an advantage in processing geological data with fuzzy boundaries such as the electrical properties of rocks and minerals. Through this algorithm, the data points in the complex resistivity parameter sample set are divided into different electrical categories, and the cluster center of each category is determined. The cluster center represents the typical electrical characteristics of the data points in that category.

[0080] The resulting electrical property categories and their corresponding cluster centers are then systematically stored and organized to construct an electrical property classification knowledge base. This knowledge base not only stores the electrical property characteristics of each category but also links them to the geological context, providing important physical property evidence for subsequent electromagnetic forward modeling and geological modeling.

[0081] Furthermore, the complex resistivity parameter sample set is processed by fuzzy clustering algorithm to divide the electrical property categories and determine the cluster centers, including:

[0082] Perform membership initialization operation on the complex resistivity parameter sample set to generate an initial membership matrix;

[0083] Use the following formula to calculate the cluster center position based on the initial membership matrix to generate the updated cluster center:

[0084]

[0085] Among them, v i Indicates the updated position of the i-th cluster center, u ij represents the membership of the jth data point to the i-th cluster center, m represents the fuzzy factor, x j represents the feature vector of the jth data point, and n represents the total number of data points in the dataset;

[0086] Recalculate the membership matrix based on the updated cluster centers to generate an optimized membership matrix;

[0087] Perform clustering stability verification on the optimized membership matrix. When the convergence condition is met, output the electrical category and cluster center. Otherwise, return to perform cluster center iteration processing.

[0088] Specifically, a membership initialization operation is performed on the complex resistivity parameter sample set to generate an initial membership matrix. This initial membership matrix fuzzily describes the degree of association between each data point and each potential cluster center, providing a starting point for subsequent clustering iterations. Using the update rules of the fuzzy clustering algorithm, the new positions of the cluster centers are calculated based on the initial membership matrix, generating updated cluster centers. This process involves comprehensively determining the movement direction and distance of the cluster centers in the feature space based on each data point's membership to the cluster center, the data point's eigenvector, and the fuzzy factor, thereby ensuring that the cluster centers better represent the data characteristics of the corresponding category.

[0089] Based on the updated cluster centers, the membership of each data point to each cluster center is recalculated to generate an optimized membership matrix. This process reflects the changes in the relationship between the data points and the cluster centers, further refining the correlation between the data points and the cluster centers. The optimized membership matrix is ​​then subjected to cluster stability verification to determine whether the clustering results meet the convergence criteria. If the convergence criteria are met, the final electrical classification results and the corresponding cluster centers are output. Otherwise, the cluster center iteration process is repeated based on the optimized membership matrix until convergence criteria are met.

[0090] Furthermore, based on the electromagnetic emission signal and the electrical classification knowledge base, frequency domain forward modeling is performed to obtain the response of the specified frequency band, including:

[0091] Based on the electromagnetic emission signal and electrical property classification knowledge base, the frequency domain electric field control equation including the Cole-Cole model is constructed;

[0092] The frequency domain electric field governing equations are solved by the vector finite element method to generate the frequency domain electromagnetic response distribution;

[0093] Extract the response components in the specified frequency band from the frequency domain electromagnetic response distribution to generate the specified frequency band response.

[0094] Specifically, a frequency domain electric field control equation containing the Cole-Cole model is constructed. This step includes combining electrical information such as complex resistivity parameters in the electrical classification knowledge base with electromagnetic emission signals, and using the Cole-Cole model to describe the dispersion characteristics of the medium, thereby establishing a mathematical model that can more accurately reflect the electromagnetic response of the underground medium and generating the frequency domain electric field control equation. Afterwards, the frequency domain electric field control equation is solved by the vector finite element method. As an efficient numerical simulation method, the vector finite element method can more accurately calculate the distribution of the electromagnetic field in complex geological structures, and then generate the frequency domain electromagnetic response distribution of the entire calculation area. This step not only needs to consider the propagation characteristics of the electromagnetic field in space, but also needs to be combined with the electrical parameters of the geological body to ensure the accuracy of the simulation results.

[0095] Next, the response components within a specified frequency band are extracted from the resulting frequency-domain electromagnetic response distribution to generate the designated frequency band response. This step requires selecting an appropriate frequency band based on the specific exploration objective and electromagnetic response characteristics. Appropriate signal processing techniques are then employed to separate the response information within the targeted frequency band from the broadband electromagnetic response, yielding response data that characterizes the electromagnetic characteristics of the subsurface geological structure within the specified frequency band. This process provides critical foundational data for subsequent electromagnetic data inversion and geological target identification.

[0096] like Figure 2 As shown in the figure, based on the orthogonal basis functions, the transient response and the specified frequency band response are jointly extrapolated to generate a full time-frequency domain response data set, including:

[0097] S201: Performing a time series sampling operation on the transient response to generate a time domain response vector; performing a frequency point sampling operation on the specified frequency band response to generate a frequency domain response vector;

[0098] S202: constructing a time-frequency conversion relationship between the time domain response vector and the frequency domain response vector through orthogonal basis functions;

[0099] S203: performing coefficient calculation operations on the time-frequency conversion relationship based on the least squares fitting criterion to generate extrapolation coefficients;

[0100] S204: Performing a response data extrapolation calculation operation using the extrapolation coefficients and the orthogonal basis functions to generate a full time-frequency domain response data set.

[0101] Specifically, step S201 involves performing a time series sampling operation on the transient response, discretizing the continuous time domain signal and generating a time domain response vector consisting of response values ​​at multiple time points. Simultaneously, a frequency point sampling operation is performed on the response of a specified frequency band, selecting key frequency points within the band and generating a frequency domain response vector containing the response amplitude and phase information for each frequency point. These steps lay the data foundation for the subsequent joint extrapolation processing.

[0102] Step S202 includes constructing a time-frequency conversion relationship between the time-domain response vector and the frequency-domain response vector using orthogonal basis functions. By projecting the time-domain and frequency-domain response vectors into the orthogonal space formed by the basis functions, the intrinsic connection between the two is found. This conversion relationship can reveal the common characteristics of the signal in the time and frequency domains and provide a mathematical framework for extrapolating the response data.

[0103] Step S203 includes performing coefficient calculation operations on the established time-frequency conversion relationship based on the least squares fitting criterion. The least squares fitting criterion solves the optimal extrapolation coefficients by minimizing the sum of squared residuals between the observed values ​​and the model predicted values. The above coefficients reflect the weights of the time domain and frequency domain response data in the orthogonal basis function space and are key parameters for achieving response data extrapolation. Through the above process, it can be ensured that the extrapolation results are as consistent as possible with the known time-frequency response data.

[0104] Step S204 involves performing an extrapolation calculation on the response data using the extrapolation coefficients and orthogonal basis functions. The extrapolation coefficients are multiplied by the orthogonal basis functions and summed to generate a full time-frequency domain response dataset. This step fuses the response information from the time and frequency domains, expanding the time-frequency coverage of the original data and generating a full time-frequency domain response dataset containing richer information. This dataset not only includes the observed time-frequency response information but also obtains a wider range of time-frequency response characteristics through extrapolation, providing more comprehensive data support for subsequent geological target identification and analysis.

[0105] Furthermore, based on the three-dimensional electrical parameter model, abnormal area extraction processing is performed to obtain the target abnormal area, including:

[0106] Performing feature recognition operations on the resistivity parameter distribution data and the polarizability parameter distribution data in the three-dimensional electrical parameter model to identify resistivity abnormality feature areas and polarizability abnormality feature areas;

[0107] The following formula is used to perform spatial overlay analysis on the resistivity anomaly characteristic area and the polarizability anomaly characteristic area to extract the resistivity-polarizability composite anomaly area:

[0108]

[0109] Among them, S overlap represents the spatial overlap, k represents the total number of discrete grid points, H i represents the volume of the i-th grid cell, C(x i ,y i ,z i ) indicates whether the i-th grid point belongs to the composite anomaly area;

[0110] Perform spatial continuity analysis on the resistivity-polarizability composite anomaly area to generate the target anomaly area.

[0111] Specifically, feature recognition operations are performed on the resistivity parameter distribution data and polarizability parameter distribution data in the three-dimensional electrical parameter model. By analyzing the distribution characteristics of resistivity and polarizability, image processing technology and pattern recognition algorithms are used to identify areas with abnormal resistivity characteristics and areas with abnormal polarizability characteristics. Areas with abnormal resistivity characteristics are areas where the resistivity values ​​deviate significantly from the background value, which may indicate the presence of different rock types or geological structures. Areas with abnormal polarizability characteristics are areas with significantly abnormal polarizability values, which are usually related to physical properties such as the mineralization or hydraulic properties of the rock.

[0112] Next, a spatial overlay analysis is performed. The resistivity and polarizability anomaly characteristic areas are spatially overlaid. Using spatial analysis techniques and combined with geological background knowledge, the resistivity-polarizability composite anomaly area is extracted. This process considers the spatial distribution, morphological characteristics, and interrelationships of the anomaly areas to determine the scope and characteristics of the composite anomaly area.

[0113] Perform spatial continuity analysis on the resistivity-polarizability composite anomaly area. Spatial continuity analysis aims to assess the spatial coherence and integrity of the composite anomaly area. By analyzing the anomaly area's boundaries, internal structure, and relationships with surrounding geological bodies, a target anomaly area is generated. This step helps eliminate sporadic, discontinuous anomalies and retain geologically significant continuous anomaly areas, providing more accurate information about the anomaly area for subsequent target body boundary delineation and geological interpretation.

[0114] Furthermore, based on the spatial relationship between the target anomaly area and the geological data, the boundary of the target body is delineated to obtain the spatial position of the target body, including:

[0115] Perform fault structure identification operations and lithologic boundary identification operations on geological data to generate fault structure elements and lithologic boundary elements;

[0116] Construct a geological constraint model based on the spatial topological relationship between the target abnormal area and the fault structural elements and lithologic boundary elements;

[0117] Perform boundary morphology optimization on the target anomaly area according to the geological constraint model to generate the target body boundary;

[0118] The boundary of the target body is converted into three-dimensional coordinate data to generate the spatial position of the target body.

[0119] Specifically, the geological data is processed for fault structure identification and lithologic boundary identification. Fault structure identification aims to determine the location, strike, and displacement characteristics of underground faults. This information is crucial for understanding the evolution of geological structures and controlling the distribution of ore bodies. Lithologic boundary identification aims to clarify the contact relationships between different rock types. Through these steps, fault structure elements and lithologic boundary elements are generated, which serve as the basis for subsequent analysis.

[0120] Next, a geological constraint model is constructed based on the spatial topological relationships between the target anomaly area and the fault structural elements and lithologic boundary elements. Spatial topological relationship analysis includes studying the relative position, intersection, and inclusion relationships between the target anomaly area and the identified faults and lithologic boundaries. By quantifying these spatial relationships and incorporating them into the geological constraint model, the model's geological rationality and constraints can be enhanced.

[0121] Next, the target anomaly area undergoes boundary morphology optimization based on the geological constraint model. Boundary morphology optimization involves adjusting and refining the target anomaly area's boundaries, guided by geological constraints, to ensure they more closely align with geological reality. This step includes removing noise from the boundaries, smoothing boundary curves, and correcting for any inappropriate intersections between the boundaries and geological structures, ultimately improving the accuracy of the target anomaly area's boundaries.

[0122] The optimized target volume boundary is converted into 3D coordinate data to generate the target volume's spatial position. This 3D coordinate data conversion requires expanding the 2D boundary information into 3D space. This conversion process yields a relatively precise 3D position of the target volume, providing critical spatial information support for subsequent geological modeling, resource estimation, and mining planning.

[0123] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0124] In one embodiment, Figure 3 As shown, the present application also provides a controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis system 300, which includes:

[0125] The electrical knowledge base construction module 301 is used to measure and process the electrical parameters of rock and mineral samples in the target area to obtain an electrical parameter set; based on the electrical parameter set, electrical classification is performed through cluster analysis to generate an electrical classification knowledge base;

[0126] The time-frequency response reconstruction module 302 is used to obtain electromagnetic emission signals; perform time-domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain transient responses; perform frequency-domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain responses in designated frequency bands; and perform joint extrapolation of the transient responses and responses in designated frequency bands based on orthogonal basis functions to generate a full time-frequency domain response dataset.

[0127] The time-frequency feature extraction module 303 is used to perform feature analysis on the full time-frequency domain response data set to generate a time-frequency feature anomaly result;

[0128] The multi-constraint inversion module 304 is used to obtain data fitting terms and model constraints; construct physical property constraints based on the electrical property classification knowledge base to obtain physical property constraints; combine the physical property constraints, data fitting terms, and model constraints to generate a multi-constraint objective function; and perform parameter iterative optimization based on the multi-constraint objective function and time-frequency characteristic anomaly results to generate a three-dimensional electrical parameter model.

[0129] The geological target positioning module 305 is used to obtain geological data; based on the three-dimensional electrical parameter model, perform abnormal area extraction processing to obtain the target abnormal area; based on the spatial relationship between the target abnormal area and the geological data, perform target body boundary delineation processing to obtain the target body spatial position.

[0130] Specifically, the electrical property knowledge base construction module 301 measures the electrical parameters of rock and mineral samples in the target area, obtains an electrical parameter set, and uses cluster analysis to classify the electrical properties into categories, generating an electrical property classification knowledge base to provide basic data support for subsequent forward modeling. The time-frequency response reconstruction module 302 acquires electromagnetic emission signals and, in combination with the electrical property classification knowledge base, performs forward modeling in both the time and frequency domains to obtain transient responses and responses in designated frequency bands. This module uses orthogonal basis functions to jointly extrapolate these two responses, generating a more comprehensive full-time and frequency domain response dataset.

[0131] The time-frequency feature extraction module 303 performs in-depth feature analysis on the full time-frequency domain response dataset, extracting time-frequency feature anomalies to provide key information for inversion calculations. The multi-constrained inversion module 304 integrates data fitting terms, model constraints, and physical property constraints constructed based on the electrical classification knowledge base to generate a multi-constrained objective function. Using parameter iterative optimization techniques, it generates a three-dimensional electrical parameter model. The geological target positioning module 305 combines geological data with the three-dimensional electrical parameter model to determine the spatial position of the target through operations such as anomaly area extraction and target boundary delineation, thereby achieving relatively accurate positioning of the geological target.

[0132] The electrical knowledge base construction module 301 is further used for:

[0133] Extracting complex resistivity parameters from the electrical parameter set to generate a complex resistivity parameter sample set;

[0134] The complex resistivity parameter sample set is processed by fuzzy clustering algorithm to obtain electrical property categories and determine the cluster center.

[0135] The electrical property categories and cluster centers are stored as an electrical property classification knowledge base.

[0136] The electrical knowledge base construction module 301 is further used for:

[0137] Perform membership initialization operation on the complex resistivity parameter sample set to generate an initial membership matrix;

[0138] Use the following formula to calculate the cluster center position based on the initial membership matrix to generate the updated cluster center:

[0139]

[0140] Among them, v i Indicates the updated position of the i-th cluster center, u ij represents the membership of the jth data point to the i-th cluster center, m represents the fuzzy factor, x j represents the feature vector of the jth data point, and n represents the total number of data points in the dataset;

[0141] Recalculate the membership matrix based on the updated cluster centers to generate an optimized membership matrix;

[0142] Perform clustering stability verification on the optimized membership matrix. When the convergence condition is met, output the electrical category and cluster center. Otherwise, return to perform cluster center iteration processing.

[0143] The time-frequency response reconstruction module 302 is further configured to:

[0144] Based on the electromagnetic emission signal and electrical property classification knowledge base, the frequency domain electric field control equation including the Cole-Cole model is constructed;

[0145] The frequency domain electric field governing equations are solved by the vector finite element method to generate the frequency domain electromagnetic response distribution;

[0146] Extract the response components in the specified frequency band from the frequency domain electromagnetic response distribution to generate the specified frequency band response.

[0147] The time-frequency response reconstruction module 302 is further configured to:

[0148] Perform time series sampling on transient responses to generate time domain response vectors; perform frequency point sampling on specified frequency band responses to generate frequency domain response vectors;

[0149] The time-frequency conversion relationship between the time domain response vector and the frequency domain response vector is constructed through orthogonal basis functions;

[0150] Perform coefficient calculation operations on the time-frequency conversion relationship based on the least squares fitting criterion to generate extrapolation coefficients;

[0151] The extrapolation coefficients and orthogonal basis functions are used to perform response data extrapolation calculations to generate a full time-frequency domain response data set.

[0152] The geological target positioning module 305 is also used to:

[0153] Performing feature recognition operations on the resistivity parameter distribution data and the polarizability parameter distribution data in the three-dimensional electrical parameter model to identify resistivity abnormality feature areas and polarizability abnormality feature areas;

[0154] The following formula is used to perform spatial overlay analysis on the resistivity anomaly characteristic area and the polarizability anomaly characteristic area to extract the resistivity-polarizability composite anomaly area:

[0155]

[0156] Among them, S overlap represents the spatial overlap, k represents the total number of discrete grid points, H i represents the volume of the i-th grid cell, C(x i ,y i,z i ) indicates whether the i-th grid point belongs to the composite anomaly area;

[0157] Perform spatial continuity analysis on the resistivity-polarizability composite anomaly area to generate the target anomaly area.

[0158] The geological target positioning module 305 is also used to:

[0159] Perform fault structure identification operations and lithologic boundary identification operations on geological data to generate fault structure elements and lithologic boundary elements;

[0160] Construct a geological constraint model based on the spatial topological relationship between the target abnormal area and the fault structural elements and lithologic boundary elements;

[0161] Perform boundary morphology optimization on the target anomaly area according to the geological constraint model to generate the target body boundary;

[0162] The boundary of the target body is converted into three-dimensional coordinate data to generate the spatial position of the target body.

[0163] In one embodiment, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0164] In one embodiment, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps in the above-mentioned method embodiments when the computer program is executed by a processor.

[0165] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0166] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. The three-dimensional forward modeling and characteristic analysis method of controlled source electromagnetic method in time and frequency domain is characterized by: The method comprises: Conducting electrical parameter measurement on rock and mineral samples in the target area to obtain an electrical parameter set; performing electrical classification based on the electrical parameter set through cluster analysis to generate an electrical classification knowledge base; Acquire an electromagnetic emission signal; perform a time-domain forward modeling process based on the electromagnetic emission signal and the electrical property classification knowledge base to obtain a transient response; perform a frequency-domain forward modeling process based on the electromagnetic emission signal and the electrical property classification knowledge base to obtain a specified frequency band response; and perform a joint extrapolation process on the transient response and the specified frequency band response based on an orthogonal basis function to generate a full time-frequency domain response data set; Performing feature analysis on the full time-frequency domain response data set to generate a time-frequency feature anomaly result; Acquire data fitting items and model constraint items; perform physical property constraint item construction processing based on the electrical property classification knowledge base to obtain physical property constraint items; combine the physical property constraint items, the data fitting items, and the model constraint items to generate a multi-constraint objective function; perform parameter iterative optimization processing based on the multi-constraint objective function and the time-frequency feature anomaly results to generate a three-dimensional electrical parameter model; Acquire geological data; extract and process the abnormal area based on the three-dimensional electrical parameter model to obtain the target abnormal area; and delineate the target body boundary based on the spatial relationship between the target abnormal area and the geological data to obtain the spatial position of the target body.

2. The controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to claim 1 is characterized in that: The electrical classification process is performed based on the electrical parameter set through cluster analysis to generate an electrical classification knowledge base, including: extracting complex resistivity parameters from the electrical parameter set to generate a complex resistivity parameter sample set; Performing fuzzy clustering algorithm processing on the complex resistivity parameter sample set to obtain electrical property categories and determine cluster centers; The electrical property categories and the cluster centers are stored as the electrical property classification knowledge base.

3. The controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to claim 2 is characterized in that: The performing of fuzzy clustering algorithm processing on the complex resistivity parameter sample set to obtain electrical property categories and determine cluster centers includes: performing a membership initialization operation on the complex resistivity parameter sample set to generate an initial membership matrix; The following formula is used to calculate the cluster center position based on the initial membership matrix to generate the updated cluster center: Among them, v i Indicates the updated position of the i-th cluster center, u ij represents the membership of the jth data point to the i-th cluster center, m represents the fuzzy factor, x j represents the feature vector of the jth data point, and n represents the total number of data points in the dataset; Recalculating the membership matrix based on the updated cluster centers to generate an optimized membership matrix; A clustering stability verification operation is performed on the optimized membership matrix, and when a convergence condition is met, the electrical property category and the cluster center are output; otherwise, the process returns to perform cluster center iteration processing.

4. The controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to claim 1 is characterized in that: The performing frequency domain forward calculation processing based on the electromagnetic emission signal and the electrical property classification knowledge base to obtain a specified frequency band response includes: Constructing a frequency-domain electric field control equation including a Cole-Cole model based on the electromagnetic emission signal and the electrical property classification knowledge base; Solving the frequency domain electric field control equation by a vector finite element method to generate a frequency domain electromagnetic response distribution; A response component within a specified frequency band is extracted from the frequency-domain electromagnetic response distribution to generate the specified frequency band response.

5. The controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to claim 1 is characterized in that: The step of performing a joint extrapolation process on the transient response and the specified frequency band response based on the orthogonal basis function to generate a full time-frequency domain response data set includes: Performing a time series sampling operation on the transient response to generate a time domain response vector; performing a frequency point sampling operation on the specified frequency band response to generate a frequency domain response vector; Constructing a time-frequency conversion relationship between the time domain response vector and the frequency domain response vector through orthogonal basis functions; Performing a coefficient calculation operation on the time-frequency conversion relationship based on a least squares fitting criterion to generate an extrapolation coefficient; The extrapolation coefficients and the orthogonal basis functions are used to perform a response data extrapolation calculation operation to generate the full time-frequency domain response data set.

6. The controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to claim 1 is characterized in that: The abnormal region extraction process is performed based on the three-dimensional electrical parameter model to obtain a target abnormal region, including: Performing feature recognition on the resistivity parameter distribution data and the polarizability parameter distribution data in the three-dimensional electrical parameter model to identify resistivity abnormality feature areas and polarizability abnormality feature areas; The following formula is used to perform spatial overlay analysis on the resistivity anomaly characteristic area and the polarizability anomaly characteristic area to extract the resistivity-polarizability composite anomaly area: Among them, S overlap represents the spatial overlap, k represents the total number of discrete grid points, H i represents the volume of the i-th grid cell, C(x i ,y i ,z i ) indicates whether the i-th grid point belongs to the composite anomaly area; A spatial continuity analysis operation is performed on the resistivity-polarizability composite abnormal area to generate the target abnormal area.

7. The controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to claim 1 is characterized in that: The target body boundary delineation process is performed based on the spatial relationship between the target abnormal area and the geological data to obtain the target body spatial position, including: Performing fault structure identification operations and lithologic boundary identification operations on the geological data to generate fault structure elements and lithologic boundary elements; Constructing a geological constraint model based on the spatial topological relationship between the target abnormal area, the fault structural elements, and the lithologic boundary elements; Performing boundary morphology optimization on the target abnormal area according to the geological constraint model to generate a target body boundary; The boundary of the target body is converted into three-dimensional coordinate data to generate the spatial position of the target body.

8. The controlled source electromagnetic method three-dimensional time-frequency domain forward modeling and characteristic analysis system is characterized by: The system comprises: An electrical knowledge base construction module is used to measure and process the electrical parameters of rock and mineral samples in the target area to obtain an electrical parameter set; based on the electrical parameter set, electrical classification is performed through cluster analysis to generate an electrical classification knowledge base; a time-frequency response reconstruction module for acquiring electromagnetic emission signals; performing time-domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain transient responses; performing frequency-domain forward modeling based on the electromagnetic emission signals and the electrical property classification knowledge base to obtain responses in designated frequency bands; and performing joint extrapolation processing on the transient responses and the responses in designated frequency bands based on orthogonal basis functions to generate a full time-frequency domain response dataset; A time-frequency feature extraction module is used to perform feature analysis on the full time-frequency domain response data set to generate a time-frequency feature anomaly result; A multi-constraint inversion module is configured to obtain data fitting items and model constraint items; construct physical property constraint items based on the electrical property classification knowledge base to obtain physical property constraint items; combine the physical property constraint items, the data fitting items, and the model constraint items to generate a multi-constraint objective function; and perform parameter iterative optimization based on the multi-constraint objective function and the time-frequency characteristic anomaly results to generate a three-dimensional electrical parameter model; The geological target positioning module is used to obtain geological data; based on the three-dimensional electrical parameter model, the abnormal area is extracted to obtain the target abnormal area; based on the spatial relationship between the target abnormal area and the geological data, the target body boundary is delineated to obtain the target body spatial position.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the controlled source electromagnetic method time-frequency domain three-dimensional forward modeling and characteristic analysis method according to any one of claims 1 to 7 are implemented.

Images