A tunnel seismic wave detection data preprocessing method, device and storage device

Abstract

The application provides a tunnel seismic wave detection data preprocessing method, equipment and storage equipment, first, the rock physical parameters are extracted from the seismic wave detection result map based on the OCR method, the original seismic wave detection data is extracted from the original rock attribute record table based on the keyword matching method, the former is used as a template, and the abnormal values in the original seismic wave detection data are removed by using the template matching method; then, when the advanced geological prediction is carried out, experts interpret the seismic wave detection data by the change of the rock physical parameters of a certain to-be-predicted mileage point relative to the construction surface to upgrade or downgrade the surrounding rock grade, and the trend characteristics of the rock physical parameters are extracted to carry out the tunnel advanced geological prediction. The application has the beneficial effects that: the abnormal values existing in the seismic wave detection data are effectively removed, and the data trend characteristics are extracted, which lays an important foundation for the intelligent advanced geological prediction of the tunnel.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of complex tunnel advanced geological prediction, and particularly relates to a tunnel seismic wave detection data preprocessing method, equipment and storage equipment. BACKGROUND

[0002] In the process of traffic engineering construction, the influence of adverse geology such as high ground stress and high ground temperature is more prominent, and engineering personnel often encounter adverse geological disasters such as fault fracture zone, karst, water and mud inrush, and soft rock stratum during construction, so it is necessary to predict the adverse geological conditions in front of the tunnel. As a long-distance detection method, seismic wave is currently the most commonly used advanced geological prediction method in deep buried tunnels. Due to the complexity of underground rock strata and the precision and sensitivity of measuring instruments, some errors may occur during seismic wave detection, and some abnormal mutation values may exist in the obtained rock physical parameters. These abnormal parameters will have a certain influence on the judgment of the geological conditions in front of the tunnel, so it is necessary to preprocess the rock physical parameters obtained by seismic wave detection to remove abnormal values before advanced geological prediction. SUMMARY

[0003] In order to solve the problem that the existing tunnel seismic wave detection data has abnormal values, resulting in low accuracy of advanced geological prediction, the present application provides a tunnel seismic wave detection data preprocessing method, equipment and storage equipment. The tunnel seismic wave detection data preprocessing method mainly comprises:

[0004] S1: extracting the result rock physical parameters corresponding to the specified mileage from the seismic wave detection result map, including the numerical information of the longitudinal wave velocity Vp, the transverse wave velocity Vs, the longitudinal transverse wave velocity ratio Vp / Vs, the Poisson's ratio, and the dynamic Young's modulus; extracting the original seismic wave detection data from the original rock attribute record table, including the numerical information of the tunnel mileage and its corresponding Vp, Vs, Vp / Vs, Poisson's ratio, and dynamic Young's modulus, and then correcting the original seismic wave detection data by taking the seismic wave detection result map data as a template, so as to remove the abnormal values in the original seismic wave detection data;

[0005] S2: extracting the trend characteristics relative to the construction surface based on the seismic wave detection data after removing the abnormal values based on engineering experience;

[0006] S3: standardizing the seismic wave trend characteristics to obtain a seismic wave trend characteristic database;

[0007] S4: interpreting the characteristic data in the database to realize advanced geological prediction of the tunnel.

[0008] Further, in step S1, the Vp, Vs, Vp / Vs, Poisson's ratio, dynamic Young's modulus corresponding to the specified mileage are extracted from the seismic wave detection result map by an optical character recognition (OCR) method, and the specific process is as follows:

[0009] First, an offline OCR model is trained, and according to actual needs, the text information and its corresponding position information in the seismic wave detection result map are extracted. After some screening and preprocessing of the extracted text information, a standard text information containing mileage and rock physical parameters is obtained.

[0010] Then, the position information of the coordinate scale is matched with the position information of the parameters. The image needs to be binarized. For a color image, only when the pixels of the R, G and B channels are all 0, black is obtained. All the curves and scales that need to be extracted in the seismic wave detection result map are black. The following formula is used to obtain them.

[0011]

[0012] Where g(i,j) is the pixel of a certain point (i,j) in the result map, R(i,j), G(i,j) and B(i,j) are the pixel values of the R, G and B channels of a certain point (i,j) in the result map.

[0013] Finally, using the matched mileage and coordinate scale, any given mileage can be used to calculate the distance from the initial position, so as to determine its position information in the picture. Then, by drawing a vertical line along the mileage, the intersection position information of Vp, Vs, Vp / Vs, Poisson's ratio and dynamic Young's modulus can be obtained. Using the position information of the rock physical parameters matched with the coordinate scale, the numerical information of the rock physical parameters corresponding to the intersection points can be obtained. According to the above OCR method, the detection mileage and its corresponding Vp, Vs, Vp / Vs, Poisson's ratio and dynamic Young's modulus can be obtained from the seismic wave detection result map.

[0014] Further, in step S1, the Vp, Vs, Vp / Vs, Poisson's ratio, dynamic Young's modulus corresponding to the specified mileage are extracted from the seismic wave detection result map by an optical character recognition (OCR) method, and the specific process is as follows:

[0015] Firstly, all the keywords in the information to be extracted need to be indexed, including indexes such as "tunnel axis intersection mileage", "Vp", "Vs", "Vp / Vs", "Poisson's ratio", "dynamic Young's modulus", and the like. The keywords contain all the parameter values to be extracted. Then, only the table content needs to be traversed to obtain the position information of all the keywords. Finally, the rock physical parameter values corresponding to the keywords are extracted to obtain the original seismic wave detection data.

[0016] Further, in step S1, an AC automaton is used to establish the index.

[0017] Further, in step S1, the seismic wave detection result map data obtained by the OCR method is used as a template to correct the original seismic wave detection data, so as to remove the abnormal values in the original seismic wave detection data. The process is as follows:

[0018] Firstly, the matching of the seismic wave result data and the original seismic wave detection data is completed through the mileage. Then, the seismic wave detection result map data obtained by the OCR method is used as a template to subtract the data matched by the mileage. The abnormal values are removed by setting the threshold values of Vp, Vs, Vp / Vs, Poisson's ratio, and dynamic Young's modulus. Specifically, the upper and lower thresholds are set. When the difference value is less than the lower threshold or the difference value is greater than the upper threshold, it is an abnormal detection data, and the abnormal detection data is deleted.

[0019] A storage device, which stores instructions and data for implementing a tunnel seismic wave detection data preprocessing method.

[0020] A tunnel seismic wave detection data preprocessing method device, comprising a processor and the storage device; the processor loads and executes the instructions and data in the storage device to implement a tunnel seismic wave detection data preprocessing method.

[0021] The beneficial effects brought by the technical scheme provided by the present application are: the present application firstly considers seismic wave detection result data, extracts text information from the seismic wave detection result graph by using an OCR method, can obtain result rock physical parameters, that is, obtains a detection mileage and corresponding Vp, Vs, Vp / Vs, Poisson's ratio and dynamic Young's modulus, can form a seismic wave detection result database, and can lay a good foundation for subsequent template matching work. The corresponding original seismic wave detection data under the corresponding keywords is obtained from the seismic wave detection original record table by the keyword matching method, and the original seismic wave detection database can be formed. The seismic wave detection result database and the original seismic wave detection database are used, the seismic wave detection result database is used as a template, the template matching method is used to automatically remove the abnormal values in the original seismic wave detection database, and the curve graph of the data after removing the abnormal values is drawn for verification. The trend change feature of the corrected seismic wave detection data is extracted, the feature is consistent with the mechanism and specification of the advanced geological prediction, and therefore the seismic wave trend feature database is formed, which is beneficial to the application of the present application in actual production, and lays an important foundation for the intelligent advanced geological prediction of the tunnel. BRIEF DESCRIPTION OF DRAWINGS

[0022] The present application will be further described below in combination with the drawings and embodiments, and the drawings are as follows:

[0023] Figure 1 Fig. 1 is a framework diagram of a tunnel seismic wave detection data preprocessing method in an embodiment of the present application.

[0024] Figure 2 Fig. 2 is an index establishment schematic diagram of an AC automatic machine in an embodiment of the present application.

[0025] Figure 3 Fig. 3 is a seismic wave parameter comparison diagram before and after removing the abnormal values in an embodiment of the present application, (1) is an original seismic wave detection data graph, (2) is a rock physical parameter curve graph after removing the abnormal values, and (3) is a seismic wave detection result graph.

[0026] Figure 4 Fig. 4 is a schematic diagram of the working of a hardware device in an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to have a clearer understanding of the technical features, objects and effects of the present application, the specific embodiments of the present application will be described in detail with reference to the drawings.

[0028] The embodiment of the present application provides a tunnel seismic wave detection data preprocessing method, device and storage device.

[0029] Please refer to Figure 1 , Figure 1is a framework diagram of a tunnel seismic wave detection data preprocessing method in an embodiment of the present application, and the preprocessing process is divided into two parts:

[0030] The first part includes extracting text information from the seismic wave detection result map by using the OCR method, obtaining the detection mileage and the corresponding Vp, Vs, Vp / Vs, Poisson ratio, and dynamic Young's modulus numerical information, forming a seismic wave detection result database; extracting the original seismic wave detection data from the original rock attribute record table by using the keyword matching method, forming an original seismic wave detection database, and an AC automatic machine is used to establish an index as shown in Figure 2 ; then the seismic wave detection result database is used as a template to correct the original seismic wave detection data, and the template matching method is used to automatically remove the abnormal values in the original seismic wave detection data, specifically: setting the upper and lower thresholds, when the difference is less than the lower threshold, or the difference is greater than the upper threshold, it is an abnormal detection data, the abnormal detection data is deleted, and a curve graph of the seismic wave detection data after removing the abnormal values is drawn Figure 3 (3)), and compared with the seismic wave detection result map Figure 3 (2)) and the curve graph of the original seismic wave detection data Figure 3 (1)), the results show that the method has a good effect on removing abnormal values, and is conducive to the application of the present application in actual production.

[0031] The second part is to extract the trend change characteristics of the rock physical parameters of the seismic wave detection data after removing the abnormal values based on engineering experience, that is, the trend characteristics relative to the construction surface, and after standardizing the seismic wave trend characteristics, a seismic wave trend characteristics database after removing the abnormal values is constructed.

[0032] The second part is to extract the trend characteristics of Vp, Vs, Vp / Vs, Poisson ratio, and dynamic Young's modulus relative to the construction surface according to the seismic wave detection data after removing the abnormal values, and after standardizing the seismic wave trend characteristics, a seismic wave trend characteristics database is constructed. That is, according to the expert rules, when the advanced geological prediction is carried out in the actual construction site, the geological conditions and rock physical parameters of the construction surface are first judged, and then when the prediction is carried out, the geological conditions of the to-be-predicted mileage point are predicted according to the change of the rock physical parameters of the to-be-predicted point relative to the construction surface.

[0033] The min-max standardization is adopted in the embodiment, because five parameters can be extracted from the seismic wave data, and the five parameters are respectively processed when the standardization is performed, taking Vp as an example, first, the maximum value and the minimum value of all Vp are calculated, then the formula Vp=(Vp(i)-min(Vp)) / (max(Vp)-Vp(i)) is used for standardization, and the other four parameters are standardized by using the same method.

[0034] At present, the advanced geological prediction is mainly performed by feature interpretation on the detection data, and the features of the detection parameters are extracted, and then the geological condition can be predicted, because the features can be associated with the rock mass integrity and the underground water condition in mechanism, and a feature library can be established to facilitate modeling and prediction.

[0035] The surrounding rock grade prediction experiment is designed for the trend feature, and the simulation experiment results are shown in Table 1.

[0036] Table 1 Simulation experiment results

[0037]

[0038] As shown in Table 1, the prediction accuracy is improved by 6.5% by using the method provided by the embodiment.

[0039] Please refer to Figure 4 , Figure 4 is a hardware device working schematic diagram of the embodiment of the application, and the hardware device specifically comprises: a tunnel seismic wave detection data preprocessing method device 401, a processor 402 and a storage device 403.

[0040] The tunnel seismic wave detection data preprocessing method device 401: the tunnel seismic wave detection data preprocessing method device 401 realizes the tunnel seismic wave detection data preprocessing method.

[0041] The processor 402: the processor 402 loads and executes the instructions and data in the storage device 403 to realize the tunnel seismic wave detection data preprocessing method.

[0042] The storage device 403: the storage device 403 stores instructions and data; and the storage device 403 is used to realize the tunnel seismic wave detection data preprocessing method.

[0043] The present application has the beneficial effects that: the present application firstly considers the seismic wave detection result data, uses the OCR method to extract the text information from the seismic wave detection result map, can obtain the result rock physical parameter, that is, obtains the detection mileage and its corresponding Vp, Vs, Vp / Vs, Poisson ratio, dynamic Young's modulus, can form the seismic wave detection result database, can lay a good foundation for the following template matching work. The corresponding original seismic wave detection data under the corresponding key word is obtained from the seismic wave detection original record table through the keyword matching method, and the original seismic wave detection database can be formed. The seismic wave detection result database and the original seismic wave detection database are used, the seismic wave detection result database is used as a template, the template matching method is used to automatically remove the abnormal value in the original seismic wave detection database, and the curve graph of the data after removing the abnormal value is drawn to verify. The trend change feature of the corrected seismic wave detection data is extracted, the feature is in line with the mechanism and specification of the advanced geological prediction, so the seismic wave trend feature database is formed, which is beneficial to the application of the present application in the actual production, and lays an important foundation for the intelligent advanced geological prediction of the tunnel.

[0044] The above only describes the preferred embodiments of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for pre-processing tunnel seismic wave exploration data, characterized in that: Comprise: S1: Extract the specified mileage corresponding to the rock physical parameters of the seismic wave detection results from the seismic wave detection results map, including the numerical information of the longitudinal wave velocity Vp, the transverse wave velocity Vs, the longitudinal transverse wave velocity ratio Vp / Vs, the Poisson ratio, and the dynamic Young's modulus; Extract the original seismic wave detection data from the original rock attribute record table, including the tunnel mileage and its corresponding Vp, Vs, Vp / Vs, Poisson ratio, and dynamic Young's modulus numerical information, and then correct the original seismic wave detection data with the seismic wave detection results map data as a template, so as to remove the abnormal values in the original seismic wave detection data; S2: Based on engineering experience, the trend characteristics of the seismic wave detection data after removing the abnormal values are extracted relative to the construction surface; S3: The seismic wave trend characteristics are standardized to obtain a seismic wave trend characteristic database; S4: By interpreting the characteristic data in the database, the advanced geological prediction of the tunnel is realized.

2. The tunnel seismic wave detection data preprocessing method according to claim 1, characterized in that: In step S1, the numerical information of Vp, Vs, Vp / Vs, Poisson ratio, and dynamic Young's modulus corresponding to the specified mileage is extracted from the seismic wave detection results map by the OCR method, and the specific process is as follows: First, an offline OCR model is trained, the text information and its corresponding position information in the seismic wave detection results map are extracted according to actual needs, and the extracted text information is screened and preprocessed, and a standard text information containing mileage and rock physical parameters is obtained after screening and preprocessing; Then, the position information of the coordinate scale is matched with the position information of the parameters, and the image is binarized, and for a color image, only when the pixels of R, G, and B channels are all 0, black is obtained, and all the curves and scales that need to be extracted in the seismic wave detection results map are black, and the following formula is used to obtain them: Where g(i,j) is the pixel of a certain point (i,j) in the results map, B(i,j), G(i,j), and B(i,j) are the pixel values of R, G, and B channels of a certain point (i,j) in the results map, respectively; Finally, the distance of a given mileage from the initial position is calculated using the matched mileage and coordinate scale, so as to determine its position information in the picture, and then a vertical line is drawn along the mileage to obtain the intersection position information of Vp, Vs, Vp / Vs, Poisson ratio, and dynamic Young's modulus, and then the position information of the rock physical parameters corresponding to the intersection points is obtained using the position information of the rock physical parameters that have been matched with the coordinate scale; The above OCR method can be used to obtain the detection mileage and its corresponding Vp, Vs, Vp / Vs, Poisson ratio, and dynamic Young's modulus from the seismic wave detection results map.

3. The tunnel seismic wave detection data preprocessing method according to claim 2, characterized in that: In step S1, the original seismic wave detection data is extracted from the irregular seismic wave detection original record table provided by the railway advanced geological prediction project department by the keyword matching method, and the specific process is as follows: Firstly, all the keywords in the information to be extracted need to be indexed, including indexes such as "tunnel axis intersection mileage", "Vp", "Vs", "Vp / Vs", "Poisson's ratio", "dynamic Young's modulus", and the like, and all the parameter values to be extracted are contained below the keywords, and then only the table content needs to be traversed to obtain the position information of all the keywords; finally, the rock physical parameter values corresponding to the keywords are extracted to obtain the original seismic wave detection data.

4. The tunnel seismic wave detection data preprocessing method of claim 3, characterized in that: In step S1, an AC automaton is used to establish the index.

5. The tunnel seismic wave detection data preprocessing method of claim 3, characterized in that: In step S1, the seismic wave detection result map data obtained by the OCR method is used as a template to correct the original seismic wave detection data, so as to remove the abnormal values in the original seismic wave detection data, and the process is as follows: Firstly, the matching of the seismic wave result data and the original seismic wave detection data is completed by the mileage, and then the seismic wave detection result map data obtained by the OCR method is used as a template to subtract the data matched by the mileage, and the abnormal values are removed by setting the threshold values of Vp, Vs, Vp / Vs, Poisson's ratio and dynamic Young's modulus, and the specific process is as follows: when the difference value is less than the lower threshold value or the difference value is greater than the upper threshold value, it is an abnormal detection data, and the abnormal detection data is deleted.

6. A storage device, characterized by: The storage device stores instructions and data for implementing the tunnel seismic wave detection data preprocessing method of any one of claims 1-5.

7. A device for pre-processing tunnel seismic wave exploration data, characterized in that it comprises: It comprises: a processor and a storage device; The processor loads and executes the instructions and data in the storage device to implement the tunnel seismic wave detection data preprocessing method of any one of claims 1-5.

Images