Geological structure automatic identification method and system

Abstract

The invention provides a geological structure automatic identification method, and belongs to the technical field of geological structure automatic identification, and the method comprises the following steps: obtaining historical geological structure data, and carrying out the preprocessing of the collected data; for different types of geologic structure data, multiple geologic structure analysis functions are established, the types of the multiple geologic structure analysis functions correspond to the type of the geologic structure data, and the multiple geologic structure analysis functions are used for calculating different types of geologic structure features; and establishing a geologic structure identification model for combining the various geologic structure analysis functions and carrying out comprehensive calculation on the geologic structure characteristics. According to the geological structure automatic identification method and system provided by the invention, the collected different types of geological data can be comprehensively analyzed, and a more accurate identification result can be obtained.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of automatic identification of geological structure, and more particularly, relates to a method and system for automatically identifying geological structure BACKGROUND

[0002] Geological structure includes geological phenomena and features such as strata, faults, fractures, folds, and magmatic rocks. Taking fractures as an example, the shape, distribution, and width of fractures are of great significance in studying groundwater flow, seismic activity, and mineral resource distribution. In the process of geological exploration and engineering construction, accurate identification and detection of the above-mentioned geological structure are the basis for ensuring engineering safety, developing reasonable exploitation plans, and predicting disasters.

[0003] However, traditional geological structure identification methods usually rely on manual observation or simple image processing techniques, which are not only inefficient and time-consuming, but also susceptible to human factors, resulting in low accuracy of the identification results. Therefore, the prior art proposes an automatic identification method for geological structure. Specifically, automatic identification of geological structure refers to the process of using computer technology and geological knowledge to automatically identify geological structure. Automatic identification of geological structure can effectively improve the work efficiency in the fields of geological exploration, geological disaster assessment, and mineral resource assessment. Deep learning algorithms, especially convolutional neural networks (CNN), are often used for automatic identification. They can learn and extract complex geological structure features from a large amount of geological image data, avoiding the limitations of traditional methods and meeting the needs of modern geological exploration for efficient and accurate detection technology.

[0004] However, the automatic identification of geological structure in the above-mentioned manner usually focuses on identifying data in a certain aspect of the geological structure. It is difficult to combine different types of geological data for comprehensive analysis, resulting in low accuracy of the identification results of the traditional method. Therefore, in order to improve the accuracy of geological structure identification, an automatic identification method that can combine different types of geological data for comprehensive analysis is needed. The present application solves this technical problem. SUMMARY

[0005] The present application provides a method and system for automatically identifying geological structure, which can perform comprehensive analysis on different types of geological data collected to obtain more accurate identification results.

[0006] A method for automatically identifying geological structure, comprising the following steps:

[0007] Obtain historical geological structure data and preprocess the collected data;

[0008] For different types of geological structure data, a plurality of geological structure analysis functions are established, the types of the plurality of geological structure analysis functions correspond to the types of the geological structure data, and are used to calculate different types of geological structure characteristics;

[0009] A geological structure identification model is established, which is used to combine the plurality of geological structure analysis functions to comprehensively calculate the geological structure characteristics;

[0010] The preprocessed historical geological structure data is input into the geological structure identification model to calculate the characteristic value of the geological structure;

[0011] The calculated characteristic value of the geological structure is fitted with the actual true characteristic value, and the geological structure identification model is adjusted according to the fitting result to reduce the calculation error;

[0012] Current geological structure data is collected and input into the adjusted geological structure identification model, and the current geological structure is identified by the adjusted geological structure identification model to obtain the characteristic value of the current geological structure;

[0013] According to the characteristic value of the current geological structure, the type of the current geological structure is determined.

[0014] Further, the geological structure data includes a set of seismic data D, a set of geological exploration data E, an evolution time T of a geological process, a set of physical parameters P of geological materials, a set of parameters M of crustal movement, and a set of parameters S of stress distribution.

[0015] Further, the geological structure identification model is:

[0016] ;

[0017] Wherein G represents the characteristic value of the geological structure; represents a seismic data analysis function; represents a geological exploration data analysis function; represents a time factor analysis function; represents a geological material physical parameter analysis function; represents a crustal movement analysis function; represents a stress distribution analysis function; - represents a weight value corresponding to the geological structure data; - represents a power exponent parameter corresponding to the geological structure data.

[0018] Further, the specific calculation formula of the seismic data analysis function is as follows:

[0019] ;

[0020] wherein, is a scaling factor for seismic data; is a maximum value of data in a seismic data set; is a minimum value of data in a seismic data set; is an average value of data in a seismic data set; is a standard deviation of data in a seismic data set.

[0021] Further, the specific calculation formula of the geological exploration data analysis function is as follows:

[0022] ;

[0023] wherein, is a scaling factor for geological exploration data, is a sum of data in a geological exploration data set, is a median of data in a geological exploration data set, is a maximum value of data in a geological exploration data set, is a minimum value of data in a geological exploration data set.

[0024] Further, the specific calculation formula of the time factor analysis function is as follows:

[0025] ;

[0026] wherein, is a scaling factor for time factor.

[0027] Further, the specific calculation formula of the geological material physical parameter analysis function is as follows:

[0028] ;

[0029] wherein, is a weight of geological density, is a geological density, is a weight of geological elastic modulus, is a geological elastic modulus, is a weight of geological viscosity, is a geological viscosity.

[0030] Further, the specific calculation formula of the crust movement analysis function is as follows:

[0031] ;

[0032] is a weight of crust movement rate, a weight of a crust movement rate, a weight of a seismic slip rate, a seismic slip rate.

[0033] Further, a specific calculation formula of the stress distribution analysis function is as follows:

[0034] ;

[0035] a weight of a crust stress field, a measured value of the crust stress field, a weight of a tension / compression stress, a measured value of the tension / compression stress.

[0036] The application further discloses a geological structure automatic identification system based on the geological structure automatic identification method.

[0037] The data acquisition module is used to acquire geological structure data, wherein the geological structure data comprises a seismic data set, a geological exploration data set, an evolution time of a geological process, a physical parameter set of a geological material, a parameter set of a crust movement and a parameter set of a stress distribution.

[0038] The model setting module is used to set a geological structure identification model, input the geological structure data into the geological structure identification model, calculate eigenvalues of a geological structure through the geological structure identification model, and the geological structure identification model is composed of a plurality of different types of analysis functions, and the analysis functions are used to analyze the geological structure data.

[0039] The geological structure identification module is used to fit the eigenvalues of the historical geological structure with real eigenvalues of the corresponding geological structure, adjust the geological structure identification model according to a fitting result, reduce a calculation error of the geological structure identification model, and identify a current geological structure through the adjusted geological structure identification model.

[0040] The application has the following technical effects:

[0041] (1) The application can acquire the seismic data set, the geological exploration data set, the evolution time of the geological process, the physical parameter set of the geological material, the parameter set of the crust movement and the parameter set of the stress distribution, and comprehensively calculate the above-mentioned multiple types of data, so that the comprehensive analysis of the geological structure can be realized, the eigenvalues of the corresponding geological structure are obtained, the type of the corresponding geological structure can be judged according to the size of the eigenvalues, the accuracy of the automatic identification is improved, and the error caused by the singularity of the data is reduced.

[0042] (2) In this scheme, the feature values ​​calculated by the geological structure identification model can be fitted according to the real feature values ​​of the geological structure, thereby adjusting the parameter values ​​in the geological structure identification model, thereby optimizing the geological structure identification model, reducing the calculation error, making the subsequent calculation results closer to the real value, and further improving the accuracy of automatic identification of geological structures. Attached Figure Description

[0043] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0045] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indicators will also change accordingly.

[0046] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the use of "and / or" or "and / or" throughout the text includes three parallel solutions. For example, "A and / or B" includes solution A, solution B, or a solution where both A and B are satisfied simultaneously. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0047] See Figure 1 An automatic geological structure identification method includes the following steps:

[0048] Historical geological structure data is acquired, and the collected data is preprocessed.

[0049] For different types of geological structure data, multiple geological structure analysis functions are established. The types of these geological structure analysis functions correspond to the types of geological structure data and are used to calculate the characteristics of different types of geological structures.

[0050] A geological structure identification model is established to combine various geological structure analysis functions to comprehensively calculate geological structure characteristics;

[0051] The preprocessed historical geological structure data is input into the geological structure identification model to calculate the feature values ​​of the geological structure;

[0052] The calculated feature values ​​of the geological structure are fitted with the actual feature values, and the geological structure identification model is adjusted according to the fitting results to reduce calculation errors.

[0053] Collect current geological structure data and input the current geological structure data into the adjusted geological structure recognition model. The adjusted geological structure recognition model is used to identify the current geological structure and obtain the feature value of the current geological structure.

[0054] The type of the current geological structure is determined based on the magnitude of its characteristic values.

[0055] The steps 101-103 in the attached diagram represent each step of the work process.

[0056] Preferably, historical data can be obtained through the following methods:

[0057] (1) Methods for acquiring earthquake datasets:

[0058] Seismic wave monitoring stations / networks (such as the China Earthquake Networks Center and the USGS): These stations continuously monitor seismic waves (P-waves, S-waves, etc.) using seismographs (such as short-period seismographs and broadband seismographs) installed underground.

[0059] Seismic exploration (active source): The seismic source is artificially detonated, and the reflected and refracted waves are recorded using detectors to obtain the velocity model of the underground structure.

[0060] Instruments and equipment: three-component seismograph, source excitation device, detector array.

[0061] (2) Methods for obtaining geological exploration datasets:

[0062] Geophysical exploration: gravity exploration (by measuring gravity anomalies); magnetic exploration (measuring magnetic field distribution); electrical exploration (such as direct current method and induced polarization method); seismic exploration (see above for details).

[0063] Geological drilling: drilling earth cores and analyzing the properties of strata at different depths through experiments.

[0064] Ground-penetrating radar (GPR): High-frequency electromagnetic waves propagate and reflect underground to obtain information about near-surface structures.

[0065] Geological surveying: Conducting on-site investigations and mapping of geomorphic structures such as outcrops, faults, and folds.

[0066] (3) Methods for obtaining the evolution time of geological processes:

[0067] Dating techniques include: radioisotope dating (such as K-Ar method, U-Pb method, C14 method); thermoluminescence dating and optically stimulated luminescence dating (applicable to sediments).

[0068] Sequence stratigraphy analysis: comparing stratigraphic correlations in different regions and inferring geological history by combining fossil records.

[0069] Remote sensing image analysis: monitoring landform evolution and analyzing the trend of fault zone activity.

[0070] (4) Methods for obtaining the set of physical parameters of geological materials (including density, elastic modulus, viscosity, etc.):

[0071] Laboratory rock property testing: Compression tests, triaxial shear tests, creep tests, etc., are performed on rock samples obtained from drilling.

[0072] Field testing techniques: seismic wave velocity testing (indirectly inferring elastic modulus); density meter, porosity testing, etc.

[0073] Numerical inversion (such as seismic inversion): deducing material parameters from known wave velocity distribution.

[0074] (5) Methods for obtaining crustal movement parameter sets (such as crustal movement rate, slip rate, etc.):

[0075] GPS Measurement (GNSS): Precision satellite positioning for monitoring horizontal and vertical displacement of the Earth's crust.

[0076] InSAR (Synthetic Aperture Radar Interferometry): Detects minute surface deformations using satellite radar images.

[0077] Fault slip observation: Estimating slip rate through slip traces and fault activity records.

[0078] Strain gauges and crustal deformation meters: Sensors are installed on the ground and underground for long-term monitoring.

[0079] (6) Methods for obtaining the set of stress distribution parameters:

[0080] Hydraulic fracturing test: Water is injected into the borehole to apply pressure, and the direction of crack propagation and stress magnitude are observed.

[0081] Earthquake focal mechanism solution: The underground stress state is inferred by inferring the direction of the initial motion of the P-wave.

[0082] Laboratory rock mechanics experiment: Controlling the loading direction and rate to deduce the triaxial stress state.

[0083] Analysis of the sliding direction at the fault plane: Inferring the direction and magnitude of the principal axes of tensile and compressive stresses.

[0084] The above methods can be used to obtain different types of geological structure data. The methods for collecting the geological structure data are all based on existing technologies, and the specific principles and processes will not be detailed here.

[0085] Preferably, since the collected data may contain errors, performing calculations on data with large errors can easily lead to low accuracy of the calculation results. Therefore, this embodiment provides a verification device for data acquisition, the components of which are as follows:

[0086] The data acquisition interface module includes a seismic signal interface (compatible with three-component seismographs and geophones), a geophysical exploration interface (compatible with gravimeters), a GNSS / InSAR data interface (compatible with GNSS receivers), a rock mass parameter interface (compatible with triaxial compressors), and a stress monitoring interface (compatible with hydraulic fracturing systems). This enables unified access to different types of data and facilitates analog-to-digital conversion of raw data.

[0087] Preprocessing module: Includes low-noise amplifier, filter, and data buffer chip, used to perform preliminary noise reduction on the acquired data and buffer the acquired data to avoid data loss, thereby realizing preprocessing.

[0088] Verification module: Includes an embedded chip (using STM32H7+FPGA heterogeneous computing). The embedded chip verifies the data according to the verification algorithm to determine the accuracy of the data and distinguish between valid data (data with high accuracy) and suspected error data (data with low accuracy).

[0089] Communication module: Includes wireless communication module and local storage hard drive, used to upload valid data to the cloud, and can store suspected error data through local storage hard drive for subsequent manual review. After review, if the suspected error data is actually valid data, the data is uploaded to the cloud and used to calculate together with the previous valid data, so as to obtain more accurate calculation results.

[0090] The verification algorithm specifically includes four layers of verification methods:

[0091] (1) Threshold verification

[0092] For seismic data: Verify the reasonable range of P-wave travel time, such as 0.5s-5s; the specific threshold can be adapted to the depth of the strata. Also verify the amplitude fluctuation range, such as ≤5m / s², to avoid extreme values ​​caused by equipment malfunction. Furthermore, verify the reasonable range of the P-wave / S-wave velocity ratio, such as 1.73-2.0, to ensure the data conforms to the wave velocity characteristics of the rock mass. By comparing the acquired data with preset thresholds, the accuracy of the data is determined; if it falls within the preset threshold range, the data is considered accurate.

[0093] For geological exploration data: it is necessary to verify the range of gravity anomalies, magnetic field strength fluctuations, and core porosity ranges. The single-point values ​​and fluctuation amplitudes of adjacent points of the above data should be double-verified. When the gravity or porosity exceeds the physically reasonable range, or the single-point sudden change of magnetic field strength is >50nT (the preset threshold) and there is no adjacent point to corroborate it, the data is judged as suspected error data.

[0094] For crustal movement data: it is necessary to verify the horizontal movement rate range (preset to 0-10 mm / yr), GNSS data positioning error (≤5 mm), and the ratio of slip rate to crustal movement rate (≤0.5). When the data falls within the above ranges, it is considered valid data.

[0095] For stress distribution data: the range of in-situ stress values ​​(0-50 MPa, which is the maximum compressive strength range of conventional rock mass), and the difference between tensile and compressive stress (≤30 MPa). When the collected data exceeds the above range, it indicates that the data exceeds the bearing capacity of conventional rock mass, and is therefore judged as suspected error data.

[0096] (2) Geological correlation verification

[0097] Stratigraphic correlation verification: For example, in the Longmenshan Fault Zone in Sichuan (a high-incidence earthquake zone), the P-wave travel time of earthquake data should show the pattern of "large fluctuations near the fault zone and small fluctuations in stable strata". If there are violent wave velocity fluctuations in stable strata in this type of geological area, the data is judged as suspected error data.

[0098] Parameter correlation verification: For example, when the rock mass density is >2.8g / cm³ (dense igneous rock), its elastic modulus should be >35GPa. If contradictory data such as "density 2.9g / cm³ but elastic modulus <20GPa" are collected, they are marked as suspected error data.

[0099] Time correlation verification: For geological evolution time data, combined with regional geological maps (e.g., the strata age of the North China Plain is mostly >20 Ma), if data of "ancient strata with T < 5 Ma" are collected, they are judged as suspected error data.

[0100] (3) Cross-validation of data consistency

[0101] Since this solution uses multi-source data fusion to improve the accuracy of the calculation results, the verification method in this solution also adopts this core logic. Cross-verification of different types of data is implemented at the acquisition end, which can more effectively eliminate errors in single types of data. The specific scenario is as follows:

[0102] Cross-validate seismic data (D) with rock mass physical parameters (P): If the seismic wave velocity indicates that the rock mass is dense, but the collected rock mass density is <2.5g / cm³, then it is determined that at least one of the two types of data has an error, and the two types of data are marked as suspected error data, which need to be manually verified later.

[0103] Cross-reference between crustal movement data (M) and stress data (S): If the crustal movement rate is >3 mm / yr (active area), but the stress data shows that the in-situ stress is <10 MPa (low stress state), and there is no evidence of regional tectonic relaxation events, then it is determined to be an error in the stress data.

[0104] Cross-reference between exploration data (E) and rock mass parameters (P): If gravity anomalies indicate the presence of a high-density body, but the rock mass density data is <2.6 g / cm³, then it is determined that there is an error in the exploration data or density test.

[0105] (4) Lightweight machine learning-assisted verification

[0106] The verification process incorporates a lightweight CNN + decision tree-based verification model. The model's parameters are embedded in a chip and are used to perform verification in the following ways:

[0107] The model is trained using historical valid data (i.e., different types of data whose accuracy has been verified) in this scheme to learn the feature distribution of valid data;

[0108] When new data is collected, the model extracts data features, including the peak shape of seismic data and the distribution of outliers in exploration data, and determines whether the data features are similar to the data features of valid data, thereby outputting the probability that the data is valid.

[0109] In this embodiment, data with a probability ≥ 85% is considered qualified data (accurate data), data with a probability between 50% and 85% is considered suspected error data (marked), and data with a probability < 50% is considered high error data (which needs to be removed).

[0110] The verification process for the above-mentioned verification devices is as follows:

[0111] The collected data is transmitted to the data acquisition interface to achieve analog-to-digital conversion. Then, the preprocessing module performs filtering, amplification, noise removal and other processing on the converted data. The four layers of verification are performed in sequence: threshold verification, geological correlation verification, cross-validation, and machine learning verification. Each layer of verification outputs the corresponding verification result.

[0112] If all layers of validation pass, the data is considered valid and will be uploaded to the cloud for further calculation. If one or two layers of validation fail, the data is considered to be suspected of being erroneous and needs to be marked and reviewed by staff. If three or four layers of validation fail, the data is considered to be high-error data and will be removed.

[0113] By employing the above methods, errors can be controlled at the source of the entire geological identification process, i.e., verification is completed at the data acquisition end, preventing data with large errors from entering subsequent calculations and affecting the accuracy of the results. Furthermore, the verification device can be integrated into existing geological exploration equipment through the data acquisition interface, eliminating the need to rebuild the acquisition system.

[0114] Furthermore, the geological structure data includes a set of seismic data D, a set of geological exploration data E, the evolution time of geological processes T, a set of physical parameters of geological materials P, a set of parameters of crustal movement M, and a set of parameters of stress distribution S.

[0115] Furthermore, the geological structure identification model is as follows:

[0116] ;

[0117] Wherein, G represents the characteristic value of the geological structure; Represents earthquake data analysis functions; Functions representing geological exploration data analysis; This represents a function for analyzing time factors. Functions representing the analysis of physical parameters of geological materials; Represents the function for analyzing crustal movement; Represents the stress distribution analysis function; - This represents the weight value of the corresponding geological structure data; - This represents the power-law parameter of the corresponding geological structure data, used to adjust the intensity of the nonlinear characteristics of different parameters.

[0118] in, For earthquake data weights, Weighting of geological exploration data Weighted by time factor, Weights for physical parameters of geological materials As the weight of crustal movement, This represents the stress distribution weight.

[0119] Specifically, each item in the above geological structure identification model represents the following aspects:

[0120] This represents the impact of seismic activity on the stability of underground structures, measuring the amplitude of fluctuations and the degree of energy concentration.

[0121] : Represents anomaly areas revealed by geological exploration data, such as cavities, faults, and water-rich areas;

[0122] : Represents the impact of "geological maturity" caused by the structural evolution history on the current state;

[0123] : Represents the influence of geotechnical material properties (stiffness, plasticity, rheology) on stability;

[0124] : Represents the periodic disturbances and deformation trends brought about by crustal movements;

[0125] : Represents the intensity of the effect of the current stress environment (tension / compression) on structural risk.

[0126] The aforementioned characteristic value G is a comprehensive quantitative index of the geological structure's "type attribute + stability state," with a value range of [0,1]. When the value of G is closer to 1, it indicates that the geological structure is more unstable and the type is more complex, such as the presence of active faults. When the value of G is closer to 0, it indicates that the current geological structure is more stable and the type is more uniform, such as ancient faults. When the value of G is between 0.3 and 0.7, it indicates that the geological structure has semi-active faults or folds containing fissures.

[0127] Optionally, the above calculation formula is applicable to complex geological regions to accommodate nonlinear relationships between parameters. For simple geological regions, such as plains, a linear model can be used for calculation, as follows:

[0128] ;

[0129] This allows for simplified calculation methods to obtain more accurate results, effectively improving calculation efficiency while maintaining high accuracy.

[0130] Furthermore, the specific calculation formula for the seismic data analysis function is as follows:

[0131] ;

[0132] in, The scaling factor represents the seismic data; The maximum value of the data in the earthquake dataset; The minimum value of the data in the earthquake dataset; This represents the average value of the data in the earthquake dataset. This represents the standard deviation of the data in the earthquake dataset.

[0133] The formula `max(D)-min(D)` quantifies the fluctuation range; larger fluctuations indicate more significant differences in underground structures. For example, active faults can cause seismic wave reflection disturbances, leading to a dramatic increase in the fluctuation range. `mean(D)+std(D)` eliminates interference from data mean shift and dispersion. For instance, a large overall stratum depth in the region can lead to a higher mean, or random noise can cause large dispersion, ensuring that the fluctuation range calculation focuses on differences in geological structure rather than the overall background. The scaling factor `A∈[0.1,0.5]` maps the calculation results to the [0,1] interval, achieving dimensional consistency with other data types and facilitating weighted fusion of multi-source data.

[0134] Existing technologies typically employ either the mean or the standard deviation separately, both using a single calculation method. However, the function in this scheme combines the mean and dispersion to eliminate background interference and facilitates the unification of dimensions to [0,1], which is beneficial for the weighted fusion of multi-source data.

[0135] Furthermore, the specific calculation formula for the geological exploration data analysis function is as follows:

[0136] ;

[0137] in, The scaling factor for geological exploration data. It is the sum of data in the geological exploration dataset. The median of the data in the geological exploration dataset. The maximum value of the data in the geological exploration dataset. It represents the minimum value of the data in the geological exploration dataset.

[0138] Existing technologies typically use only sum(E), which is prone to missing underground cavities. In contrast, this solution uses sum(E)-median(E), which is more effective in locating local anomalies (such as cavities and veins) than existing technologies. This overcomes the problem of difficulty in distinguishing local structures in existing technologies.

[0139] Furthermore, the specific calculation formula for the time factor analysis function is as follows:

[0140] ;

[0141] in, The scaling factor represents the time factor. The above calculation method conforms to the laws of geological evolution, that is, the stability of ancient structures increases at a slower rate. Linear calculations can easily lead to ancient faults being misidentified as active faults. Therefore, the calculation method in this scheme can avoid the above misidentification.

[0142] Furthermore, the specific calculation formula for the physical parameter analysis function of the geological materials is as follows:

[0143] ;

[0144] in, As the weight of geological density, For geological density, As the weight of the geological elastic modulus, For geological elastic modulus, As the weight of geological viscosity, This refers to geological viscosity. Therefore, by comprehensively calculating geological density, geological elastic modulus, and geological viscosity, the accuracy of the calculation results is improved.

[0145] Furthermore, the specific calculation formula for the crustal movement analysis function is as follows:

[0146] ;

[0147] The weighting of crustal movement rate, The rate of crustal movement, The weights for seismic slip velocity, This represents the seismic slip rate. It can enhance the contribution of sliding rate to structural risk (the risk increases dramatically when the sliding rate is >2 mm / yr), thereby avoiding the underestimation of fault risk in linear calculations and helping to improve the accuracy of calculation results.

[0148] Furthermore, the specific calculation formula for the stress distribution analysis function is as follows:

[0149] ;

[0150] The weights of the geostress field are given. This represents the measured value of the geostress field. As the weight of tensile / compressive stress, These are the measured values ​​of tensile / compressive stress.

[0151] The following examples, using different application scenarios, illustrate the relationship between G-values ​​and geological structures:

[0152] (1) Fault-like structures

[0153] When the value of G is in the range (0.6, 1.0), it indicates that the geological structure is a high-risk area, i.e., an active fault. In the calculation of the corresponding G value, the crustal movement rate is >3 mm / yr and the slip rate is >2 mm / yr. The value is relatively large, and the weighting increases the value of G; the stress distribution is close to the compressive strength of the rock mass. Larger This will amplify the impact; the seismic data fluctuates wildly, and the difference between max(D) and min(D) is large, making... The value is large, the weight This amplifies the contribution. The resulting G value indicates that the geological structure is in an active state and prone to disasters such as earthquakes and landslides.

[0154] When the value of G is in the range of (0.3, 0.6), it indicates that the geological structure is a medium-risk area, i.e., a semi-active fault. In the corresponding calculation process, the crustal movement rate is 1-3 mm / yr and the slip rate is 0.5-2 mm / yr. The value is moderate; the ground stress is 15-25 MPa, which has not reached the critical value. The contribution is moderate; the seismic data shows gentle fluctuations. The value is moderate. Therefore, the calculated G value under these conditions indicates that the geological structure has some activity and requires regular monitoring.

[0155] When the value of G is in the range [0.0, 0.3], it indicates that the geological structure is a low-risk area, i.e., an ancient fault. In the corresponding calculation process, the crustal movement rate is <1 mm / yr and the slip rate is <0.5 mm / yr. The value is relatively small; the geological evolution time T > 50 Ma. The value is stable; the rock mass is dense (density > 2.7 g / cm³). The value is small. Therefore, the G value in this case is calculated, indicating that the geological structure is stable and there are no obvious signs of activity.

[0156] By using the above methods, different types of geological data can be comprehensively calculated and analyzed to calculate the corresponding G value, thereby gaining an understanding of the geological structure.

[0157] (2) Folded structures

[0158] When the value of G is in the range (0.7, 1.0), it indicates that the geological structure is a high-risk area, meaning that there are fractured folds. In the corresponding calculation, the geological exploration data shows significant local anomalies, with a large value for sum(E) - median(E). The value is relatively high; the elastic modulus is <20 GPa, indicating low rock mass strength. The value is relatively high; the crustal movement rate is >2 mm / yr, and the folds are subjected to compression / tension deformation. A relatively large value indicates that the folds in this area have broken and are prone to collapse.

[0159] When the value of G is in the range (0.4, 0.7), it indicates that the geological structure is in a medium-risk area, meaning the folds are relatively stable. In the corresponding calculation, the value of sum(E) - median(E) is moderate. The value is moderate; viscosity 10 5 -10 6 Pa·s, the rock mass shows slight plastic deformation. The stress value is moderate; the stress distribution is uniform (tensile / compressive stress difference < 5 MPa). The value is moderate. This indicates that the folds in this geological structure contain a small number of fractures, and there is no significant risk at present, but regular observation is required.

[0160] When the value of G is in the range [0.0, 0.4], it indicates that the geological structure is a low-risk area, meaning that there are fractured folds. In the corresponding calculation process, the geological exploration data showed no local anomalies, and the value of sum(E)-median(E) was relatively small. The value is relatively small; the elastic modulus is >30 GPa, indicating high rock mass strength. The value is relatively small; the evolution time T>30Ma, and the structure has been formed and stabilized. This indicates that the folds of this geological structure are intact, with basically no fractures or fissures.

[0161] Based on the above method, geological structures such as igneous rocks and cavities can also be analyzed. For example, regarding igneous rocks, when the detected density is high, the evolution time is long, and the difference in exploration data is small, the value of G is smaller, indicating that the rock mass is hard, dense, and highly stable, corresponding to dense igneous rocks such as granite; conversely, it indicates loose clastic rocks. Regarding cavities, when there are significant local anomalies in the exploration data, with low density and uneven stress distribution, a larger value of G indicates the presence of underground cavities and a risk of collapse; when the difference in magnetic field strength is large, the density and stress are high, and... and When the value is moderate, it indicates that there is a high probability of the presence of a mineral vein, and there is basically no significant risk of collapse.

[0162] Therefore, the G value can be calculated using the above method in combination with the actual application scenario, thereby obtaining the corresponding geological structure identification results.

[0163] This invention also discloses an automatic geological structure identification system, based on the above-mentioned automatic geological structure identification method, comprising:

[0164] Data acquisition module: used to acquire geological structure data, wherein the geological structure data includes: seismic data set, geological exploration data set, evolution time of geological processes, physical parameter set of geological materials, parameter set of crustal movement, and parameter set of stress distribution;

[0165] Model setting module: used to set up a geological structure recognition model, input the geological structure data into the geological structure recognition model, calculate the feature values ​​of the geological structure through the geological structure recognition model, the geological structure recognition model consists of multiple different types of analysis functions, the analysis functions are used to analyze the geological structure data;

[0166] Geological structure identification module: used to fit the feature values ​​of historical geological structures with the corresponding real feature values ​​of geological structures, and adjust the geological structure identification model according to the fitting results to reduce the calculation error of the geological structure identification model, and identify the current geological structure through the adjusted geological structure identification model.

[0167] Preferably, the weights in the geological structure identification model and analysis function can be fitted using the least squares method. The true feature values ​​are verified data obtained through drilling, field surveys, etc., and are objectively existing and accurate data.

[0168] Preferably, historical data is used as the training set, and the system automatically learns by minimizing the error between the predicted result G and the true label. - Weights, for example, least squares, gradient descent, etc.

[0169] For example, suppose there are several geological samples (e.g., from different boreholes, different survey areas, or different time points). For each sample, there are six input feature values ​​and one output value, where:

[0170] The six input feature values ​​are as follows:

[0171] Seismic data processing results f1(D); geological exploration data processing results f2(E); time factor processing results f3(T); geophysical parameter processing results f4(P); crustal movement processing results f5(M); stress distribution processing results f6(S).

[0172] One output value: representing the true geological structure score or identification result of the sample, denoted as .

[0173] Multiple sets of such sample data are required (at least 6 sets, more than 10 sets are recommended).

[0174] Next, an input matrix needs to be constructed based on the above data: arrange the six feature values ​​of each sample into a row, and multiple samples form an input matrix.

[0175] Constructing the target vector: This involves assigning the true structural score value of each sample to the target vector. They are collected into a column vector.

[0176] Minimize the sum of squared errors: Assume there is already a set of candidate weights - Then the predicted value for each sample is:

[0177] The sum of the squared differences between the predicted and actual values ​​of all samples is denoted as the error function:

[0178]

[0179] By mathematically solving for the minimum point of this error function, a set of optimal w1~w6 weight values ​​can be obtained. Then, the weight values ​​in the geological structure identification model are replaced with the calculated w1~w6 weight values.

[0180] for - For optimization, the following methods can be used for calculation:

[0181]

[0182] Therefore, methods such as weight value calculation can be used to calculate the optimal value. - That is, the power exponent parameter.

[0183] In summary, this scheme optimizes the weight values ​​and power exponent parameters to make the numerical values ​​in the model more accurate, thereby optimizing the geological structure identification model and making the calculation results closer to the true values.

[0184] The method provided by this invention can be implemented in a terminal environment, which may include one or more components such as a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0185] A processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the terminal, and performs various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and by calling data stored in the storage medium.

[0186] Storage media can include random access memory (RAM) or read-only memory (ROM). Storage media can be used to store instructions, programs, code, code sets, or instructions.

[0187] The display screen is used to show the user interface of each application.

[0188] In the formula of this invention, all subscripts are only used to distinguish parameters and have no actual meaning.

[0189] In addition, those skilled in the art will understand that the structure of the terminal described above does not constitute a limitation on the terminal. The terminal may include more or fewer components, or combine certain components, or have different component arrangements. For example, the terminal may also include radio frequency circuits, input units, sensors, audio circuits, power supplies, and other components, which will not be described in detail here.

[0190] The technical features not described in detail in this solution are based on the conventional operation and general understanding of those skilled in the art and are derived from existing technologies, and will not be elaborated further here.

[0191] The above description is merely an exemplary embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the technical concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for automatic identification of geological structures, characterized in that, Includes the following steps: Historical geological structure data is acquired, and the collected data is preprocessed. For different types of geological structure data, multiple geological structure analysis functions are established. The types of these geological structure analysis functions correspond to the types of geological structure data and are used to calculate the characteristics of different types of geological structures. A geological structure identification model is established to combine various geological structure analysis functions to comprehensively calculate geological structure characteristics; The preprocessed historical geological structure data is input into the geological structure identification model to calculate the feature values ​​of the geological structure; The calculated feature values ​​of the geological structure are fitted with the actual real feature values, and the geological structure identification model is adjusted based on the fitting results. Collect current geological structure data and input the current geological structure data into the adjusted geological structure recognition model. The adjusted geological structure recognition model is used to identify the current geological structure and obtain the feature value of the current geological structure. The type of the current geological structure is determined based on the magnitude of its characteristic values.

2. The automatic geological structure identification method according to claim 1, characterized in that, The geological structure data includes seismic data set D, geological exploration data set E, geological process evolution time T, physical parameter set P of geological materials, parameter set M of crustal movement, and parameter set S of stress distribution.

3. The automatic geological structure identification method according to claim 2, characterized in that, The geological structure identification model is as follows: ； Wherein, G represents the characteristic value of the geological structure; Represents earthquake data analysis functions; Functions representing geological exploration data analysis; This represents a function for analyzing time factors. Functions representing the analysis of physical parameters of geological materials; Represents the function for analyzing crustal movement; Represents the stress distribution analysis function; - This represents the weight value of the corresponding geological structure data; - This represents the power exponent parameter of the corresponding geological structure data.

4. The automatic geological structure identification method according to claim 3, characterized in that, The specific calculation formula for the earthquake data analysis function is as follows: ； in, The scaling factor represents the seismic data; The maximum value of the data in the earthquake dataset; The minimum value of the data in the earthquake dataset; This represents the average value of the data in the earthquake dataset. This represents the standard deviation of the data in the earthquake dataset.

5. The automatic geological structure identification method according to claim 3, characterized in that, The specific calculation formula for the geological exploration data analysis function is as follows: ; in, The scaling factor for geological exploration data. It is the sum of data in the geological exploration dataset. The median of the data in the geological exploration dataset. The maximum value of the data in the geological exploration dataset. It represents the minimum value of the data in the geological exploration dataset.

6. The automatic geological structure identification method according to claim 3, characterized in that, The specific calculation formula for the time factor analysis function is as follows: ； in, The scaling factor is a time factor.

7. The automatic geological structure identification method according to claim 3, characterized in that, The specific calculation formula for the physical parameter analysis function of the geological materials is as follows: ； in, As the weight of geological density, For geological density, As the weight of the geological elastic modulus, For geological elastic modulus, As the weight of geological viscosity, It is geologically cohesive.

8. The automatic geological structure identification method according to claim 3, characterized in that, The specific calculation formula for the crustal movement analysis function is as follows: ； The weighting of crustal movement rate, The rate of crustal movement, The weights for seismic slip velocity, This represents the seismic slip rate.

9. The automatic geological structure identification method according to claim 3, characterized in that, The specific calculation formula for the stress distribution analysis function is as follows: ； The weights of the geostress field are given. This represents the measured value of the geostress field. As the weight of tensile / compressive stress, These are the measured values ​​of tensile / compressive stress.

10. An automatic geological structure identification system, based on the automatic geological structure identification method according to any one of claims 1-9, characterized in that, include: Data acquisition module: used to acquire geological structure data, wherein the geological structure data includes: seismic data set, geological exploration data set, evolution time of geological processes, physical parameter set of geological materials, parameter set of crustal movement, and parameter set of stress distribution; Model setting module: used to set up a geological structure recognition model, input the geological structure data into the geological structure recognition model, calculate the feature values ​​of the geological structure through the geological structure recognition model, the geological structure recognition model consists of multiple different types of analysis functions, the analysis functions are used to analyze the geological structure data; Geological structure identification module: used to fit the feature values ​​of historical geological structures with the corresponding real feature values ​​of geological structures, and adjust the geological structure identification model according to the fitting results to reduce the calculation error of the geological structure identification model, and identify the current geological structure through the adjusted geological structure identification model.

Images