A spatially perceptive-based method for predicting formation distribution in deep-earth virtual drilling.

Abstract

This invention provides a method for predicting the formation distribution of deep earth using virtual drilling based on spatial perception. The method includes: transmitting multi-frequency detection signals into the target deep earth space and simultaneously receiving echo signals reflected from the formation; constructing a three-dimensional spatial perception field of the target deep earth space based on the amplitude and phase characteristics of the echo signals; constructing a virtual drilling model based on the three-dimensional spatial perception field, which is used to equivalently simulate the trajectory advancement and core sampling process of physical drilling; and generating and outputting the formation distribution prediction results of the target deep earth space based on the output results of the virtual drilling model. This invention aims to solve the problems of high cost, high risk, low resolution of conventional geophysical methods, and inability to equivalently measure the physical properties of core samples in traditional physical drilling in deep earth exploration, achieving non-contact, high-efficiency, and high-precision virtual core sampling and formation distribution prediction.

Description

Technical Field

[0001] This invention relates to the field of spatial computing technology, and in particular to a method for predicting the distribution of deep-earth strata through virtual drilling based on spatial perception. Background Technology

[0002] With the increasing depletion of shallow mineral resources, deep-earth resource exploration has become a strategic direction for global mining development. Deep-earth exploration usually refers to geological exploration at depths ranging from 1,000 to 5,000 meters. In this depth range, the geological conditions are complex, the temperature and pressure are high, and the lithology changes drastically. Traditional physical drilling faces prominent problems such as high cost, long cycle, and high risk. Traditional geophysical exploration methods are usually based on the propagation characteristics of macroscopic physical fields (such as seismic wave fields and electromagnetic fields). Their spatial resolution decreases sharply with increasing depth, making it difficult to accurately depict the stratigraphic interfaces and lithological distribution at depths of more than a kilometer. Especially for thin target ore bodies or complex structural areas, the prediction accuracy is difficult to meet exploration needs. "There is no existing non-contact detection method that can directly and equivalently simulate the physical drilling coring process. Although physical drilling can directly obtain rock core samples and obtain key physical property parameters such as density, elastic modulus, and porosity, it is costly and risky. Geophysical exploration, on the other hand, is less expensive, but it can only indirectly infer underground information and cannot directly provide a sequence of physical property parameters equivalent to those measured in the rock core. There is a clear information gap between the two." Therefore, there is an urgent need in this field for a spatially perceptive-based method for predicting the formation distribution of deep-earth virtual drilling to solve the above problems. Summary of the Invention

[0003] This invention provides a spatially perceptive-based method for predicting formation distribution in deep earth virtual drilling, aiming to solve the problems of high cost, high risk, low resolution of conventional geophysical methods, and inability to achieve equivalent core measurement properties in deep earth exploration. It enables non-contact, high-efficiency, and high-precision virtual coring and formation distribution prediction.

[0004] This invention provides a method for predicting formation distribution in deep earth virtual drilling based on spatial perception, comprising the following steps: Step S1: Transmit multi-frequency detection signals into the deep Earth space of the target and simultaneously receive echo signals reflected by the strata; Step S2: Construct a three-dimensional spatial sensing field of the target deep earth space based on the amplitude and phase characteristics of the echo signal; Step S3: Based on the three-dimensional spatial perception field, construct a virtual drilling model, which is used to equivalently simulate the trajectory advancement and core sampling process of physical drilling; Step S4: Based on the output of the virtual drilling model, generate and output the predicted stratigraphic distribution of the target deep space.

[0005] Compared with the prior art, the beneficial effects of this application are as follows: This invention generates virtual wells through adaptive gradient trajectory, which can automatically track the direction of the most significant changes in formation properties. The trajectory conforms to the constraints of real drilling engineering and is more in line with the actual exploration logic. This invention is the first to achieve direct equivalent simulation of the physical drilling coring process, and quickly inverts the sequence of key physical property parameters such as equivalent density, equivalent elastic modulus, and equivalent porosity that are consistent with the measured core, thus breaking down the long-standing information gap between physical drilling and geophysical exploration. This invention uses a three-dimensional spatial fusion formula to extend the one-dimensional virtual drilling trajectory into a global stratigraphic distribution model, and combines stratigraphic interface discrimination and lithological clustering rules to improve the completeness and reliability of stratigraphic prediction. This method eliminates the need for physical drilling and on-site coring, significantly reducing deep-earth exploration costs, shortening exploration cycles, and lowering operational risks. It enables large-area, high-efficiency, and high-precision rapid deep-earth exploration and mineral prospecting.

[0006] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0007] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof; in the drawings: Figure 1 This is a flowchart illustrating a spatially perceptive-based method for predicting formation distribution in deep-earth virtual drilling, as provided by this invention. Detailed Implementation

[0008] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention. Example 1

[0009] This invention provides a method for predicting formation distribution in deep earth virtual drilling based on spatial awareness. Please refer to [link to relevant documentation]. Figure 1 This includes the following steps: Step S1: Transmit multi-frequency detection signals into the deep Earth space of the target and simultaneously receive echo signals reflected by the strata; Step S2: Construct a three-dimensional spatial sensing field of the target deep earth space based on the amplitude and phase characteristics of the echo signal; Step S3: Based on the three-dimensional spatial perception field, construct a virtual drilling model. The virtual drilling model is used to equivalently simulate the trajectory advancement and core sampling process of physical drilling. Step S4: Based on the output of the virtual drilling model, generate and output the predicted stratigraphic distribution of the target deep space.

[0010] In one embodiment, in step S1, the detection signal is a broadband electromagnetic pulse signal, the number of transmitting radio points is N, N≥3, and M signal receiving points are deployed on the ground, M≥4, forming a transceiver array. After performing noise suppression, gain compensation, and direct wave removal on the echo signal at each receiving point, the peak echo amplitude corresponding to the i-th transmitting point and the j-th receiving point is extracted. and echo phase difference Where i=1,…,N, j=1,…,M; echo phase difference It is the difference between the received phase of the echo signal and the initial phase of the corresponding transmitted signal.

[0011] Specifically, noise suppression employs wavelet thresholding to decompose the echo signal into wavelets and apply soft thresholding to high-frequency coefficients, filtering out environmental electromagnetic interference and instrument noise. Gain compensation uses a time-varying gain control method, exponentially compensating the echo amplitude based on the signal propagation time to correct for geometric spread and medium absorption attenuation during electromagnetic wave propagation. Direct wave removal uses an adaptive subtraction method, calculating the theoretical arrival time of the direct wave based on the straight-line distance between the transmitter and receiver, subtracting the direct wave component from the echo signal, and retaining the ground reflection wave. Echo amplitude peak value... The echo phase is obtained by detecting the maximum amplitude value in the echo waveform. The difference between the received phase of the echo signal and the initial phase of the corresponding transmitted signal is obtained by calculating the phase difference corresponding to the peak position of the cross-correlation function between the echo waveform and the transmitted signal waveform.

[0012] In one implementation, in step S2, the three-dimensional spatial sensing field is constructed in the following manner: Using the surface exploration benchmark point of the target deep earth space as the origin of the coordinate system, and taking due north as the positive X-axis direction, due east as the positive Y-axis direction, and the vertical downward direction of the surface as the positive Z-axis direction, a three-dimensional spatial benchmark coordinate system is constructed. The target deep-earth space is discretized into a three-dimensional regular grid with equal step size. The grid step size is set according to the exploration accuracy requirements, usually taking 1 / 2 to 1 / 3 of the expected spatial resolution, for example, 5m×5m×5m. The spatial sensing intensity value at each grid point (x,y,z) is calculated. in, The average attenuation coefficient of the deep earth medium corresponding to the i-th transmission frequency point is obtained by laboratory testing of core samples from drilled wells in the target area or by statistical analysis of regional geological data. Let be the frequency of the i-th transmitting frequency point; Let be the equivalent electromagnetic wave propagation speed in the deep earth medium corresponding to the i-th transmission frequency point; The distance of the two-way propagation path is the sum of the straight-line distance from grid point (x,y,z) to the i-th transmitting point and the straight-line distance to the j-th receiving point.

[0013] Specifically, spatial perception intensity value The calculation formula is based on the ray theory of electromagnetic wave propagation, treating each transmitter-receiver pair as an independent detection channel. The energy intensity contributed by each channel at any point in space is accumulated through spherical diffusion attenuation, medium absorption attenuation, and phase coherence superposition. In the formula, The term characterizes the amplitude attenuation caused by spherical diffusion; The term characterizes the exponential decay caused by medium absorption, reflecting the dissipation characteristics of electromagnetic wave energy in deep earth media; the cosine term... Characterizing the phase coherence of waves reflects the interference effect of waves of different frequencies and propagation paths at spatial points, enabling the sensing field to reflect changes in the wave impedance of the strata.

[0014] By summing the contributions of all transmit-receive pairs, information from multiple sources and channels is fused into each grid point, constructing a complete three-dimensional spatial sensing field containing amplitude, phase, frequency, and spatial geometric location information. This field reflects both the electromagnetic property distribution of the underground medium and the spatial location information of the stratum interface.

[0015] coefficient The following method was used: If a well has been drilled in the target area, electromagnetic parameters of the drill core samples were tested in the laboratory, and the attenuation coefficient at different frequencies was measured. The average value was then taken as the mean. If no wells have been drilled, the target area is divided into several geological units based on regional geological data statistics. The empirical data of lithology-frequency attenuation relationship of each unit is consulted and values ​​are assigned according to the transmission frequency point.

[0016] Let be the equivalent electromagnetic wave propagation speed in the deep Earth medium corresponding to the i-th transmission frequency point. This speed is calculated using the following formula: in Let be the permeability of the medium, and take the permeability of free space. The equivalent dielectric constant of the medium corresponding to the i-th frequency point is obtained through dielectric constant data of known strata in the target area or laboratory tests.

[0017] Representing grid points The sum of the straight-line distance to the i-th transmitting point and the straight-line distance to the j-th receiving point, i.e. ,in Let i be the coordinates of the i-th launch point. Let j be the coordinates of the receiving point. These are the coordinates of the grid points. The echo phase difference corresponding to the i-th transmitting point and the j-th receiving point is extracted after preprocessing in step S1.

[0018] In one implementation, step S3, the process of constructing the virtual drill model, includes: Generate virtual drill trajectory curves that conform to drilling engineering specifications; Calculate the equivalent core physical response of the virtual drill along the trajectory curve; The equivalent core physical response is equivalent to the measured physical property parameters of the core obtained by physical drilling.

[0019] Specifically, the process of generating the virtual drilling trajectory curve includes: The trajectory starts at the virtual wellhead coordinates preset on the surface and ends when the preset target exploration depth (e.g., 2000m) or the spatial perception intensity signal-to-noise ratio is lower than the preset threshold (e.g., signal-to-noise ratio <3dB). The trajectory advancement direction is determined based on the gradient field of the three-dimensional spatial sensing field, which makes the virtual drill extend along the direction with the largest rate of change of spatial sensing intensity. This is because the direction with the largest rate of change of spatial sensing intensity usually corresponds to the direction with the most drastic changes in formation properties, i.e. the direction with the most exploration value. Set trajectory engineering constraints to limit the total angle change rate of the trajectory (i.e., the maximum allowable value of the change angle between adjacent step length directions) to ensure that the generated trajectory conforms to the feasibility of actual drilling engineering.

[0020] Virtual drill trajectory curve The following spatial curve differential equation is obtained by solving the fourth-order Runge-Kutta method, with the solution step size equal to the sampling interval of the virtual drill trajectory: in, The arc length parameter along the virtual drilling trajectory is defined by the trajectory starting point. =Starting from 0, it gradually increases as the trajectory extends; For the trajectory curve in arc length The spatial position vector at that location is of the form of , indicating the arc length parameter The three-dimensional spatial coordinates of the point are obtained stepwise by the fourth-order Runge-Kutta method, starting from the initial point. Initially, advance once for each step size and update the position coordinates; arc length The gradient vector of the intensity field sensed in three-dimensional space is calculated in the discrete grid using the central difference method: ,in Similarly; for those not on the grid points The location is obtained by interpolating the gradient values ​​of adjacent grid points using trilinear interpolation. The magnitude of the gradient vector is calculated using the following formula: This magnitude is used to normalize the gradient vector to a unit direction vector; The trajectory smoothing constraint function is defined as the angle of change in trajectory direction between adjacent step sizes. When the preset threshold for the rate of change of the full angle is exceeded (e.g., 30° / 100m), A value of 0 forces the trajectory to stop or change direction; otherwise, a value of 1 constrains the smoothness of the trajectory, preventing the generation of sharp curves that are impractical in engineering. When the value is 0, it means that the directional adjustment within that step size has failed, and it is necessary to backtrack or terminate trajectory generation.

[0021] By solving the above differential equation, a series of discrete points are obtained. The line connecting these points forms the virtual drilling trajectory curve. This embodiment breaks through the limitations of traditional virtual trajectories that rely on manual pre-setting, grid traversal, or simple straight lines. It uses the spatially sensed field gradient as the basis for adaptive propulsion, allowing the virtual drill to autonomously extend along the direction of the most drastic formation changes, thus improving the accuracy of the exploration probe. It introduces trajectory engineering constraints, combining mathematical optimization with drilling specifications to solve the problem of virtual trajectories being detached from engineering reality. It uses differential equations and the Runge-Kutta method to solve the problem, achieving continuous, smooth, and high-precision spatial trajectory construction, which is more in line with the actual drilling process compared to discrete step propulsion methods.

[0022] In one implementation, the process of calculating the equivalent core physical response of the virtual drill along the trajectory curve includes: The virtual drilling trajectory curve is discretized at equal arc length intervals. For each sampling point, calculate the equivalent core physical parameters. The equivalent core physical parameters include at least the equivalent density. Equivalent elastic modulus and equivalent porosity These correspond to the measured density, elastic modulus, and porosity of the rock core from physical drilling, respectively.

[0023] Specifically, the equivalent core physical parameters are obtained by solving the following set of equations: in, These represent the measured average density, average elastic modulus, and average porosity of the overlying strata in the target area, respectively. The correlation coefficient for deep earth media properties, with a value range of [value range missing]. The first directional derivative of spatial perception intensity along the trajectory direction; The second directional derivative of spatial perception intensity along the trajectory direction; It is a very small constant; The arc length on the virtual drilling trajectory is Spatial coordinates of the location These are the coordinates of the starting point of the virtual drilling trajectory. represent The spatial Euclidean distance from the spatial coordinates of the point to the starting point of the virtual drilling trajectory.

[0024] It should be noted that this embodiment uses a spatial sensing field. The changes in these parameters, instead of actual drilling and coring, allow for the derivation of equivalent core properties. Stronger means the medium is denser and reflects more light. Rapid changes along the trajectory indicate abrupt changes in stratigraphic properties. The curvature variation represents the stratigraphic structure and porosity variation; therefore, density... The elastic modulus is used to reflect the first-order change in local reflection intensity. Used to reflect spatial cumulative stiffness, formation compaction degree, and porosity. Used to reflect the undulations of the geological structure and the degree of fractures / density; For the formula Because density is positively correlated with the reflectivity of the strata and the compactness of the medium, the spatial sensing field... First derivative along the trajectory Represents: the rate of change of reflection intensity along the direction of drill bit advance; the faster the change, the more abrupt the change in formation properties, and thus the more abrupt the change in density. This is used to compress its derivative to (-1,1), ensuring that the density will not become negative or infinitely large, reflecting that the density change has a natural upper limit and will not be infinitely amplified with signal fluctuations; Positive represents the spatial perception field A stronger structure indicates a denser geological formation. rise; A negative value represents the spatial perception field. A weakening of the strata indicates that the soil is becoming more porous. decline.

[0025] For the formula The elastic modulus It reflects the stiffness, compressive strength, and compaction degree of rock. It is not determined at a single point, but rather accumulates along depth; that is, the deeper the rock, the stronger the compaction. The larger the spatial gradient, the greater the spatial gradient. The larger the value, the greater the difference in geological structure, i.e., the higher the stiffness; integral This represents the weighted accumulation of spatial sensing field inhomogeneity from the wellhead to the current point. The deeper the distance, the lower the weight (the denominator increases with depth, resulting in a decrease in weight). The greater the gradient, the greater the contribution. That is, the stiffness increases with depth, but it is much more affected by the nearby strata than by the distant ones. For the formula Since porosity is strongly correlated with formation curvature, bedding undulation, and fracture development, it can be obtained through the second derivative. It represents the curvature / convexity of the spatial sensing field along the trajectory. A large curvature indicates that the strata are curved and have many interfaces, that is, the pores are developed. A small curvature indicates that the strata are flat and dense, that is, the pores are low. Used to ensure that porosity is always between 0 and 1, tanh is used to limit the influence of curvature from overloading, a positive second derivative indicates that the formation is convex, that is, the porosity is relatively developed, and a negative second derivative indicates that the formation is concave and compacted, that is, the porosity is reduced. It should be noted that, It is obtained by calculating the arithmetic mean of the overlying formation logging data of drilled wells within the target area, and serves as the regional background reference value; the correlation coefficient... The values ​​are obtained by linear regression fitting of the core parameters measured in the drilled well with the first derivative, second derivative, and gradient value of the spatial sensing field at the corresponding location. The values ​​are all between 0 and 1.

[0026] In one implementation, step S4, the process of generating the stratigraphic distribution prediction result includes: The equivalent core physical parameters obtained from each sampling point on the trajectory curve of the virtual drilling model are interpolated in three dimensions and feature fusion is performed to generate a three-dimensional stratigraphic distribution prediction model that includes the spatial location of stratigraphic interfaces, the distribution of lithological types, and the spatial distribution of physical property parameters. Based on the model, the vertical distribution profile and planar distribution contour map of the stratigraphy at any location in the target deep space are output.

[0027] Specifically, the virtual drilling model can only obtain parameters along the trajectory line. Therefore, it is necessary to use three-dimensional spatial interpolation methods, such as Kriging interpolation and Inverse Distance Weighted Interpolation (IDW), to extend the parameters along the trajectory line to the entire three-dimensional grid space, and then perform feature fusion to obtain a global three-dimensional model. This model includes the three-dimensional spatial location of each stratum interface, the distribution range of different lithologies, and the spatial distribution of physical properties such as density, elastic modulus, and porosity. When outputting the vertical distribution profile, an arbitrary vertical section is selected, and the lithology and physical property data of all grid points on the section are extracted to draw a continuous profile. When outputting the planar contour map, a fixed depth layer is selected, and the physical property parameters on the layer are displayed using contour lines to obtain the planar distribution characteristics.

[0028] In one implementation, the three-dimensional stratigraphic distribution prediction model is generated using the following spatial fusion formula, and stratigraphic interfaces are determined based on abrupt change points in the fusion results, and lithological types are classified based on the fusion results: in, The total number of discrete sampling points for the virtual drilling trajectory; For the first Arc length coordinates of each sampling point; It is a very small constant; For the first The spatial influence radius at each sampling point is taken as 2 to 5 times the sampling interval of the virtual drill trajectory. The directional derivative of spatial perception intensity in the direction of the virtual drill trajectory method; For the formation interface discrimination step function, when in The threshold for identifying formation interfaces; These are the physical property fusion feature values ​​of the three-dimensional stratigraphic distribution model, used for stratigraphic interface extraction and lithology identification.

[0029] Specifically, the formula extends from a one-dimensional virtual drilling trajectory to a three-dimensional spatial distribution, diffuses the physical properties of each sampling point of the virtual drill along the space, and produces a significant response only at the formation interface, thereby constructing a three-dimensional formation distribution prediction model.

[0030] The formula consists of the product of three parts: physical property eigenvalues. It integrates equivalent density, elastic modulus, and porosity to form a single index reflecting the comprehensive engineering properties of rock; Gaussian kernel function The features of discrete sampling points are smoothly diffused to the surrounding space, with greater weight for closer points, thus achieving spatial continuity; step function. It is an interface selection switch, which is set to 1 only when the normal derivative exceeds the threshold (i.e., a formation interface exists), otherwise it is set to 0.

[0031] This formula ensures that only sampling points located near the stratigraphic interface participate in spatial diffusion, and that their physical properties influence the surrounding area in the form of a Gaussian kernel, while non-interface locations contribute nothing; ultimately... The high-value regions in three-dimensional space accurately correspond to the location of the formation interface, realizing the transformation from one-dimensional virtual drilling information to a three-dimensional formation interface model.

[0032] Furthermore, the process of extracting the formation interface includes: Stratigraphic interfaces are locations where physical properties change abruptly, that is... along depth Where the direction changes drastically; Take a series of samples along the virtual drilling trajectory or any vertical profile. Calculate the rate of change (gradient / difference) between adjacent points. when When examining the interface, determine that this is a geological interface; Combined with a step function If the normal derivative is large, it indicates an abrupt change in the formation reflection coefficient; if the normal derivative is small, it indicates a gradual change in the formation. Therefore, the final interface determination logic is: The spatial sensing field normal changes abruptly.

[0033] For the process of lithology identification, it is possible to utilize The range of numerical values ​​directly corresponds to lithology, namely loose sedimentary layers and topsoil layers. Smaller, mudstone, shale Medium to low, siltstone, fine sandstone Medium to dense sandstone and limestone Larger, intrusive rocks, volcanic rocks, ore bodies Significantly abnormally high; In one embodiment, for the entire three-dimensional space Perform normalization; Based on the measured core samples and well logging lithology from drilled wells in a specific area, the corresponding data for each type of lithology were statistically analyzed. Threshold range; For any grid point in space, if It was determined to be mudstone; if It is determined to be sandstone; if It was determined to be a dense rock layer or anomaly in mineralization.

[0034] It should be noted that the aforementioned parameter fitting, data regression, numerical calculation, signal preprocessing, spatial interpolation, cluster analysis, and stratigraphic interface identification, as well as other related experiments and data processing procedures, are all conventional techniques in the fields of geophysical exploration, geological engineering, and signal processing. Specifically, noise suppression, gain compensation, direct wave removal, and amplitude and phase extraction are conventional signal processing methods in the fields of ground-penetrating radar and electromagnetic exploration; physical property correlation coefficient fitting, reference parameter statistics, and threshold determination all employ commonly used methods in this field, such as linear regression, statistical averaging, and experimental calibration; and three-dimensional mesh calculation, gradient solving, numerical iteration of differential equations, spatial interpolation and fusion, and profile and contour plotting are all conventional implementation methods in the fields of numerical analysis and geological modeling.

[0035] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting formation distribution in deep earth virtual drilling based on spatial perception, characterized in that, Includes the following steps: Step S1: Transmit multi-frequency detection signals into the deep Earth space of the target and simultaneously receive echo signals reflected by the strata; Step S2: Construct a three-dimensional spatial sensing field of the target deep earth space based on the amplitude and phase characteristics of the echo signal; Step S3: Based on the three-dimensional spatial perception field, construct a virtual drilling model, which is used to equivalently simulate the trajectory advancement and core sampling process of physical drilling; Step S4: Based on the output of the virtual drilling model, generate and output the predicted stratigraphic distribution of the target deep space.

2. The method according to claim 1, characterized in that, In step S1, the detection signal is a broadband electromagnetic pulse signal, the number of transmitting radio points is N, N≥3, and M signal receiving points are deployed on the ground, M≥4, forming a transceiver array; After performing noise suppression, gain compensation, and direct wave removal on the echo signal at each receiving point, the peak echo amplitude corresponding to the i-th transmitting point and the j-th receiving point is extracted. and echo phase difference Where i=1,…,N, j=1,…,M; The echo phase difference It is the difference between the received phase of the echo signal and the initial phase of the corresponding transmitted signal.

3. The method according to claim 2, characterized in that, In step S2, the three-dimensional spatial sensing field is constructed in the following manner: Using the surface exploration benchmark point of the target deep earth space as the origin of the coordinate system, and taking due north as the positive X-axis direction, due east as the positive Y-axis direction, and the vertical downward direction of the surface as the positive Z-axis direction, a three-dimensional spatial benchmark coordinate system is constructed. The target deep-earth space is discretized into a three-dimensional regular grid with a uniform step size. The grid step size is set according to the exploration accuracy requirements, and the spatial sensing intensity value at each grid point (x, y, z) is calculated. : ， in, The average attenuation coefficient of the deep earth medium corresponding to the i-th transmission frequency point is obtained by laboratory testing of core samples from drilled wells in the target area or by statistical analysis of regional geological data. Let be the frequency of the i-th transmitting frequency point; Let be the equivalent electromagnetic wave propagation speed in the deep earth medium corresponding to the i-th transmission frequency point; The distance of the two-way propagation path is the sum of the straight-line distance from grid point (x,y,z) to the i-th transmitting point and the straight-line distance to the j-th receiving point.

4. The method according to claim 3, characterized in that, In step S3, the process of constructing the virtual drill model includes: Generate virtual drill trajectory curves that conform to drilling engineering specifications; The equivalent core physical response of the virtual drill is calculated along the trajectory curve; The equivalent core physical response is equivalent to the measured physical property parameters of the core obtained by physical drilling and coring.

5. The method according to claim 4, characterized in that, The process of generating the virtual drilling trajectory curve includes: The trajectory starts at the virtual wellhead coordinates preset on the surface and ends when the preset target exploration depth or spatial perception intensity signal-to-noise ratio is lower than a preset threshold. The trajectory advancement direction is determined based on the gradient field of the three-dimensional spatial perception field, so that the virtual drill extends along the direction with the largest rate of change of spatial perception intensity; Set trajectory engineering constraints to limit the total angle change rate of the trajectory.

6. The method according to claim 5, characterized in that, The trajectory curve of the virtual drill The following spatial curve differential equation is obtained by solving the fourth-order Runge-Kutta method, with the solution step size equal to the sampling interval of the virtual drill trajectory: ， in, The arc length parameter along the virtual drill trajectory; For the trajectory curve in arc length The spatial position vector at that location; arc length The gradient vector of the intensity field sensed in three-dimensional space; The magnitude of the gradient vector; The trajectory smoothing constraint function is defined as the angle of change in trajectory direction between adjacent step sizes. When the preset full-angle change rate threshold is exceeded, Set to 0 otherwise, and use it to constrain the smoothness of the trajectory.

7. The method according to claim 6, characterized in that, The process of calculating the equivalent core physical response of the virtual drill along the trajectory curve includes: The virtual drilling trajectory curve is discretized at equal arc length intervals. For each sampling point, the equivalent core physical parameters are calculated. These equivalent core physical parameters include at least the equivalent density. Equivalent elastic modulus and equivalent porosity These correspond to the measured density, elastic modulus, and porosity of the rock core from physical drilling, respectively.

8. The method according to claim 7, characterized in that, The equivalent core physical parameters are obtained by solving the following set of equations: ， in, These represent the measured average density, average elastic modulus, and average porosity of the overlying strata in the target area, respectively. The correlation coefficient for deep earth media properties, with a value range of [value range missing]. The first directional derivative of spatial perception intensity along the trajectory direction; The second directional derivative of spatial perception intensity along the trajectory direction; It is a very small constant; The arc length on the virtual drilling trajectory is Spatial coordinates of the location These are the coordinates of the starting point of the virtual drilling trajectory. represent The spatial Euclidean distance from the spatial coordinates of the point to the starting point of the virtual drilling trajectory.

9. The method according to claim 8, characterized in that, In step S4, the process of generating the stratigraphic distribution prediction result includes: The equivalent core physical parameters obtained at each sampling point of the trajectory curve of the virtual drilling model are subjected to three-dimensional spatial interpolation and feature fusion to generate a three-dimensional stratigraphic distribution prediction model that includes the spatial location of stratigraphic interfaces, the distribution of lithological types, and the spatial distribution of physical property parameters. Based on the model, the vertical distribution profile and planar distribution contour map of the stratigraphic system at any location in the target deep space are output.

10. The method according to claim 9, characterized in that, The three-dimensional stratigraphic distribution prediction model is generated using the following spatial fusion formula. Stratigraphic interfaces are determined based on abrupt change points in the fusion results, and lithological types are classified based on the fusion results: ， in, The total number of discrete sampling points for the virtual drilling trajectory; For the first Arc length coordinates of each sampling point; It is a very small constant; For the first The spatial influence radius at each sampling point is taken as 2 to 5 times the sampling interval of the virtual drill trajectory. The directional derivative of spatial perception intensity in the direction of the virtual drill trajectory method; For the formation interface discrimination step function, when in The threshold for identifying formation interfaces; These are the physical property fusion feature values ​​of the three-dimensional stratigraphic distribution model, used for stratigraphic interface extraction and lithology identification.

Images