A method for karst geophysical prospecting and data coupling processing of a power transmission line

Abstract

The application discloses a power transmission line karst geophysical prospecting and data coupling processing method, comprising the following steps: initially determining a karst development area through a high-density surface wave method, further determining the karst development area through a high-density electrical method, and obtaining bottom layer detection data; coupling processing the detection data by using a machine learning coupling algorithm to determine the area, scale and size of the karst development; and detecting key areas by using a geological radar to verify the coupling result of the bottom layer detection data. The method is characterized in that: a plurality of geophysical prospecting methods are formulated to detect the power transmission line area, the same area is detected from the perspective of multiple means, the multidimensionality of data is improved, the machine learning coupling algorithm is used to couple process the bottom layer data obtained through detection, the coupling effect of data is improved, the utilization rate of effective data is increased, the geological radar is used to detect the key areas, the data coupling result is further verified, and the accuracy of geophysical prospecting is improved.

Description

Technical Field

[0001] This invention relates to the field of karst exploration data processing, and in particular to a method for geophysical exploration and data coupling processing of karst for power transmission lines. Background Technology

[0002] With the increasing need for rural revitalization in my country, power facilities are being deployed more and more into remote areas. However, my country's karst area exceeds 3 million km², accounting for nearly one-third of the country's land area. Power engineering projects inevitably traverse karst regions. Due to the diverse and complex nature of karst development, the construction of transmission lines in karst areas faces significant risks to national property and personnel safety, including high risks of major casualties, due to unclear karst cave detection and improper handling measures. Furthermore, secondary disasters caused by karst after operation will also affect operational safety. Therefore, investigating the degree of karst development, karst characteristics, and the impact of karst on the stability of power facility foundations is a key focus of karst foundation engineering investigation.

[0003] Existing karst exploration typically employs only a single geophysical method, resulting in relatively limited data. Furthermore, different geophysical methods are used for exploration followed by comprehensive analysis. This combined approach can be considered a "pseudo-coupled" comprehensive exploration method. When the final results of several exploration methods differ, it is difficult to make a choice, which can easily lead to misjudgments or incorrect assessments.

[0004] For example, a "High-Density Surface Wave Exploration Data Acquisition and Processing Method" disclosed in Chinese patent literature, publication number CN107807382A, includes two parts: data acquisition and data processing. The data processing is further divided into preprocessing, waveform processing, dispersion analysis, inversion analysis, and visualization processing. The specific steps of this high-density surface wave exploration data acquisition and processing method are as follows: S1: Data acquisition adopts an exploration method using encrypted and overlapping measurement points. In the first step, 24 detectors are arranged at certain intervals on the measurement line to form a sensor array. However, the above scheme uses a single high-density surface wave exploration data acquisition method. Transmission lines are long and narrow, with substations and tower foundations, making geophysical exploration somewhat complex and subject to interference factors. A single geophysical exploration method is difficult to meet the high-quality and high-efficiency geological exploration requirements of transmission lines. Summary of the Invention

[0005] This invention aims to address the problem of existing technologies using only one geophysical method for karst exploration of transmission lines, or using multiple geophysical methods but with errors caused by false coupling in data processing. It provides a method for geophysical exploration and data coupling processing of karst in transmission lines, which formulates targeted geophysical exploration and data processing methods based on the characteristics of transmission lines, improves data coupling effect, increases the utilization rate of effective data, and improves geophysical accuracy. A further objective of this invention is to employ a co-excitation source data processing algorithm to process data from different sources collected at the same detection point. These are superimposed to obtain a record of "co-excited seismic sources"; Another objective of this invention is to apply the same treatment to every detection point throughout the entire exploration area, thereby enabling the detection of thousands of [data points]. By effectively utilizing the waveform data, accurate waveform results can be obtained, thereby increasing the effectiveness of the inversion. A further objective of this invention is to address the issue of long and narrow transmission lines with inherent interference by moving the entire seismic source and detector array forward so that it overlaps with the positions of some detectors in the previous row. This process is repeated until data acquisition for the entire survey line is completed, and the rolling method is used for surveying.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for geophysical exploration and data coupling processing of karst in power transmission lines, comprising: Different detection methods are used to conduct overall detection of the transmission line area and obtain the underlying detection data; The detection data is coupled using a machine learning coupling algorithm to determine the region, scale, and size of karst development. Ground-penetrating radar (GPR) was used to probe key areas to verify the coupling results of the underlying exploration data. Multiple targeted geophysical methods were developed to probe the transmission line area, exploring the same area from multiple perspectives to improve data multidimensionality. Machine learning coupling algorithms were employed to couple the underlying exploration data, enhancing the coupling effect and increasing the utilization rate of effective data. GPR was then used to probe key areas to further verify the data coupling results and improve the accuracy of geophysical exploration.

[0007] Preferably, the step of using different detection methods to perform overall detection of the transmission line area and obtain the underlying detection data includes: The high-density surface wave method was used to conduct a comprehensive detection of the transmission line area, and the data was superimposed using a co-excitation source data processing algorithm. High-density electrical resistivity tomography (EDS) was used to obtain the two-dimensional distribution of resistivity on the geological slope. The high-density surface wave method was used to preliminarily determine the karst development area, and the high-density EDS was used to further determine the karst development area. The high-density surface wave method employed a DCX-1A type centralized mainframe; the high-density EDS used a Winner-Schlumberger electrode array and employed a rolling method for surveying.

[0008] Preferably, the data superposition using a co-excitation source data processing algorithm includes: Perform dispersion analysis on the surface wave signal and extract the surface wave dispersion curve; Cross-sectional structural analysis was performed, and the underlying shear wave velocity was inferred from the dispersion curve.

[0009] Furthermore, including: The surface wave dispersion curve was extracted using the high-resolution FK method (frequency-wavenumber spectrum (FK) analysis), and then analyzed using the maximum amplitude velocity sweep method. This process can be described by the following formula: In the formula, It removes the influence of amplitude; A hybrid algorithm combining genetic algorithm and least squares method is adopted. First, the genetic algorithm performs inversion. When the error converges to a certain level, the output model at this point is used as the initial model for least squares method. Finally, the least squares method is used for inversion to accelerate the convergence of error and minimize the error between the inverted curve data and the actual measurement data.

[0010] Preferably, the step of coupling the probe data using a machine learning coupling algorithm to determine the karst development area, scale, and size includes: The resistivity acquired and processed by high-density electrical resistivity method and the shear wave velocity acquired and processed by high-density surface wave method are used as the sample set for the clustering model. Initialize the k-value and weight phasor of the clustering model, and determine the k-value using the elbow method. By determining the k-value using the elbow method, the true number of clusters in the data can be obtained.

[0011] Preferably, the surface wave signal is subjected to dispersion analysis to extract the surface wave dispersion curve, including the use of frequency-wavenumber spectrum analysis. The cross-sectional structure analysis includes the conversion of Rayleigh surface wave phase velocity into shear wave velocity.

[0012] Furthermore, the phase velocity of Rayleigh surface waves in layered media satisfies: ; Among them, V si Let V be the shear wave velocity of the i-th layer. pi Let D be the longitudinal wave velocity of the i-th layer. i H is the density of the i-th layer. i Let be the thickness of the i-th layer, where i is from 1 to N.

[0013] Preferably, the k-value and weight phasor of the initial clustering model are determined using the elbow method, including: The resistivity and wave velocity of the underlying data in the high-density point method and the high-density surface wave method are initialized as follows: ; right Recalculate the new centroid for all sample points. ; in, ; The data above is iterated until the number of iterations exceeds the preset maximum number of iterations N or the k centroids obtained from the iterations no longer change, at which point the final category classification is output.

[0014] Preferably, the objective of the clustering model is to minimize the weighted squared error E to determine the final result. In the formula Indicate category The mean, Let be the weight vector, and .

[0015] Preferably, the key areas include the tower foundations and high-voltage substation areas. Given the characteristics of long and narrow transmission lines with distribution of substations and tower foundations, multiple detection methods are employed, and underlying data coupling processing is used to improve data utilization and coupling accuracy, thereby enhancing the precision of geophysical exploration.

[0016] In summary, the present invention has the following beneficial effects: (1) It formulates a variety of targeted geophysical exploration methods to explore the transmission line area, explores the same area from multiple perspectives, and uses machine learning coupling algorithms to couple the underlying data of the exploration, thereby improving the data coupling effect. (2) It preliminarily determines the karst development area through high-density surface wave method and further determines the karst development area through high-density electrical resistivity tomography, thereby improving the multidimensionality of the data. (3) It explores key areas through ground-penetrating radar, further verifying the data coupling results and improving the accuracy of geophysical exploration. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the high-density surface wave method detection results of a method for geophysical exploration and data coupling processing of karst in transmission lines according to an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of the high-density point method detection results of a method for geophysical exploration and data coupling processing of karst in transmission lines according to an embodiment of the present invention.

[0019] Figure 3 This is a flowchart of the underlying data fusion processing of a method for geophysical exploration and data coupling processing of karst in transmission lines according to an embodiment of the present invention.

[0020] Figure 4 This is a schematic diagram of the elbow method for determining the k value in a method for geophysical exploration and data coupling processing of karst in transmission lines according to an embodiment of the present invention.

[0021] Figure 5This is a schematic diagram of the bottom-level data fusion detection results of a method for geophysical exploration and data coupling processing of karst in transmission lines according to an embodiment of the present invention.

[0022] Figure 6 This is a schematic diagram of the geological radar detection results of a method for geophysical exploration and data coupling processing of karst in power transmission lines according to an embodiment of the present invention.

[0023] In the picture: 1. Elbow. Detailed Implementation

[0024] The present invention will now be further described with reference to the accompanying drawings and specific embodiments.

[0025] Example: A method for integrated geophysical exploration and data coupling processing of karst in power transmission lines includes the following steps: S1: A high-density surface wave method is used to conduct a comprehensive survey of the transmission line area to preliminarily determine the lava development zone; the specific principles and operating methods are as follows: S11: First, set up a detector array at the starting point of the survey line, with 12 or 24 detectors; at a certain offset distance from the detectors, excite elastic waves and record the elastic waves received by the detectors to complete the data acquisition of the first arrangement; S12: Move the seismic source and detector array forward as a whole, so that it coincides with the position of some detectors in the previous row, and excite it to collect data. Repeat this step until the data collection for the entire survey line is completed.

[0026] Surface wave signals acquired by different detectors under the same seismic source have a certain phase difference. After the excitation source and detector are moved 1m along the survey line at the same time, the distance between the same detection point and the seismic source changes in the two data acquisitions. Therefore, the surface wave signals measured in the two times will also have a phase difference.

[0027] S13: A co-excitation source data processing algorithm is used to superimpose data from different sources collected at the same detection point, thus obtaining a "co-excitation source" record. Since the energy of artificially applied excitation sources cannot be completely consistent, surface wave signals from the same detection point cannot be simply superimposed when comprehensively processing data from the entire survey line. By performing the same processing on every detection point throughout the entire exploration area, thousands of waveform data points can be effectively utilized to obtain accurate waveform results, thereby increasing the effectiveness of the inversion.

[0028] The specific algorithms for processing co-excitation seismic sources include: S131: Perform dispersion analysis on the surface wave signal to extract the surface wave phase velocity information: High-resolution frequency-wavenumber spectrum analysis was used to extract the surface wave dispersion curve, and then the maximum amplitude velocity scanning method was employed for analysis. This process can be described by the following formula: In the formula, It removes the influence of amplitude; S132: Perform cross-sectional structural analysis: The underlying shear wave velocity structure is deduced from the dispersion curve. The phase velocity of the Rayleigh surface wave in the layered medium can be expressed as: Among them, V si Let V be the shear wave velocity of the i-th layer. pi Let D be the longitudinal wave velocity of the i-th layer. i H is the density of the i-th layer. i Let be the thickness of the i-th layer, where i ranges from 1 to N. A hybrid algorithm combining genetic algorithm and least squares method is employed. First, the genetic algorithm performs inversion. Once the error converges to a certain level, the output model at this point is used as the initial model for the least squares method. Finally, the least squares method is used for inversion to accelerate error convergence and minimize the error between the inverted curve data and the actual measurement data. The detection results of the high-density surface wave method are as follows: Figure 1 As shown.

[0029] S2: High-density electrical resistivity tomography (EDT) was used for measurement, and the measurement data was analyzed to obtain the two-dimensional distribution of resistivity in the geological profile. Specifically, data processing using hybrid high-density resistivity tomography included: apparent resistivity calculation, topographic correction, format conversion, filtering, forward and inverse modeling, finally yielding a resistivity distribution cross-sectional map. The high-density EDT results are shown below. Figure 2 As shown.

[0030] S3: Couple the underlying data from the high-density surface wave method and the high-density electrical resistivity method, including the following sub-steps: S31: The resistivity acquired and processed by high-density electrical resistivity method and the shear wave velocity acquired and processed by high-density surface wave method are used as the sample set for the Wkmeans clustering model. S32: Initialize the k-value and weight phasor of Wkmeans, using the elbow method to determine the k-value, such as... Figure 4 As shown; specifically including: such as Figure 3 As shown, the resistivity and wave velocity of the underlying data in the high-density electrical resistivity method and the high-density surface wave method are initialized as follows: for Calculate samples and each center of mass Distance: ,Will The smallest one is Corresponding category Update the category at this time. for right Recalculate the new centroid for all sample points. ; The above data is iterated until the number of iterations exceeds the maximum number of iterations N set in advance or the k centroids obtained from the iterations no longer change. Then the final category classification can be output. The value of k is measured here using the elbow method. The process of determining k using the elbow method is as follows: Figure 2 As shown.

[0031] The goal of using Wkmenas is to minimize the weighted squared error E to determine the final result: In the formula Indicates category The mean, Let be the weight vector, and .

[0032] The classification levels for resistivity and shear wave are determined based on local geological conditions, such as karst caves, rock masses, and soil masses. The image after fusion processing of the underlying data is as follows: Figure 5 As shown.

[0033] S4: Employs Swedish MALA ground-penetrating radar and a 270MHz shielded antenna; uses range-triggered mode for continuous detection; The basic parameters of ground-penetrating radar are as follows: Electromagnetic pulse wave travel time (ns) In the formula: — Depth of the target being surveyed; x — Distance between the transmitting and receiving antennas (x can be ignored since z > x); —The speed at which electromagnetic waves propagate in a medium.

[0034] speed of electromagnetic waves in a medium Where: C—the speed of electromagnetic wave propagation in vacuum (0.29979 m / ns); —The relative permittivity of the medium, —The relative permeability of the medium, generally speaking ; Data processing for ground-penetrating radar (GPR) detection specifically includes: static correction and removal (i.e., time zero-point correction), DC offset to DC offset, gain adjustment, horizontal interference removal, bandpass filtering, moving average, and range normalization. A GPR detection map is shown below. Figure 6 As shown.

[0035] S5: Compare and verify the radar maps of key areas with the underlying data processing results from high-density surface wave method and density electrical resistivity tomography (DMT). Use high-density surface wave method to initially identify karst development areas, and use high-density DMT to further confirm these areas, improving the multidimensionality of the data. Develop multiple targeted geophysical methods to explore the transmission line area, exploring the same area from multiple perspectives. Use machine learning coupling algorithms to couple the underlying data, improving the data coupling effect. Use ground-penetrating radar to explore key areas, further verifying the data coupling results and improving the accuracy of geophysical exploration.

[0036] The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

[0037] Although this paper uses terms such as co-excitation source data processing algorithm, coupling processing, high-density surface method, and elbow method extensively, the possibility of using other terms is not excluded. These terms are used merely for the convenience of describing and explaining the essence of this invention; interpreting them as any additional limitation would contradict the spirit of this invention.

Claims

1. A method for geophysical exploration and data coupling processing of karst in transmission lines, characterized in that, include: Different detection methods are used to conduct overall detection of the transmission line area and obtain the underlying detection data; The detection data was coupled using the Wkmeans machine learning coupling algorithm to determine the region, scale, and size of karst development. Ground-penetrating radar was used to probe key areas to verify the coupling of machine learning between surface wave data and electrical resistivity data from the underlying exploration data. Combined results; The acquisition of the underlying detection data includes: A high-density surface wave method was used to conduct a comprehensive survey of the transmission line area, and a co-excitation source data processing algorithm was employed to analyze the data. Overlay; The resistivity distribution of the geological slope was obtained by high-density electrical resistivity measurement. The karst development area was initially determined by the high-density surface wave method, and the karst development area was further determined by the high-density electrical resistivity tomography method. The high-density surface wave method is used to conduct overall detection of the transmission line area. A detector array is set up at the beginning of the survey line. Elastic waves are excited at a certain offset distance from the detectors and the elastic waves received by the detectors are recorded. The source and detector array are moved forward as a whole so that they coincide with the position of some detectors in the previous row, and the waves are excited and data is collected. This process is repeated until the data collection for the entire survey line is completed.

2. The method for geophysical exploration and data coupling processing of karst in transmission lines according to claim 1, characterized in that, The high-density surface wave method uses a DCX-1A centralized mainframe; The high-density electrophysiological method employs a Winner-Schlumberger electrode device.

3. The method for geophysical exploration and data coupling processing of karst in transmission lines according to claim 1, characterized in that, The data overlay process using a co-excitation source data processing algorithm includes: Dispersion analysis is performed on the surface wave signal to extract the surface wave dispersion curve. Cross-sectional structure analysis is then performed to infer the velocity of the underlying shear wave from the dispersion curve. The cross-sectional structure analysis employs the maximum amplitude velocity scanning method. The reverse calculation of the under-shear wave velocity uses a hybrid algorithm of genetic algorithm and least squares method; By superimposing different source data collected at the same detection point, a co-excitation source record is obtained; the same processing is performed on each detection point in the entire exploration area to effectively utilize waveform data and obtain accurate waveform results.

4. The method for geophysical exploration and data coupling processing of karst in transmission lines according to claim 3, characterized in that, The process of coupling the detected data using a machine learning coupling algorithm to determine the karst development area, scale, and size includes: The resistivity acquired using high-density electrical resistivity method and the shear wave velocity acquired using high-density surface wave method were used as the sample set for the Wkmeans clustering model. Initialize the k-value and weight phasor of the Wkmeans clustering model, and determine the k-value using the elbow method; The classification levels of resistivity and shear wave are determined based on local geological conditions.

5. The method for geophysical exploration and data coupling processing of karst in transmission lines according to claim 4, characterized in that, The surface wave signal is subjected to dispersion analysis to extract the surface wave dispersion curve, including the use of frequency-wavenumber spectrum analysis. The cross-sectional structure analysis includes the conversion of Rayleigh surface wave phase velocity into shear wave velocity.

6. The method for geophysical exploration and data coupling processing of karst in transmission lines according to claim 5, characterized in that, The initialization of the clustering model's k-value and weight phasor, using the elbow method to determine the k-value, includes: The resistivity and wave velocity of the underlying data in the high-density point method and the high-density surface wave method are initialized as follows: ； right Recalculate the new centroid for all sample points. ; in, ; The data above is iterated until the number of iterations exceeds the preset maximum number of iterations N or the k centroids obtained from the iterations no longer change, at which point the final category classification is output.

7. The method for geophysical exploration and data coupling processing of karst in transmission lines according to claim 6, characterized in that, The objective of the Wkmeans clustering model is to minimize the weighted squared error E to determine the final result. in, Indicates category The mean, Let be the weight vector, and .

8. The method for geophysical exploration and data coupling processing of karst in transmission lines according to claim 7, characterized in that, The key areas include the tower foundations and the high-voltage transmission and transformation station areas.

Images