Automatic analysis method and system for advanced geological prediction of tunnels using elastic wave method

Abstract

The present invention belongs to the field of geological engineering construction, and discloses an automatic analysis method and system for advanced geological prediction of tunnels using elastic wave method, which solves the problems of less training samples, poor universality of intelligent model engineering, and low automation level in the existing research on the application of artificial intelligence to advanced geological prediction analysis of tunnels. The present invention obtains elastic wave method parameter data by identifying elastic wave method parameter images, and establishes an elastic wave method parameter database; combines rock mechanics theory, geological disaster mechanism, etc. to establish a mapping relationship between key geophysical parameters of elastic wave method, geological indicators, and geological quality, to establish a classifier classification principle; finally, a machine learning method is used to construct a classifier model according to the classifier classification principle, and training and test samples are obtained based on the elastic wave method parameter database to train and verify the model. The model can be used to realize real-time automatic analysis of the entire process, thereby improving the accuracy and efficiency of the analysis.

Description

Technical Field

[0001] The invention belongs to the field of geological engineering construction, and in particular relates to an elastic wave method tunnel advanced geological prediction automatic analysis method and system. Background Art

[0002] In order to ensure the safety of tunnel construction and avoid the occurrence of various adverse geological disasters, advanced geological prediction should be carried out during the tunnel excavation process. The elastic wave method is one of the most widely used methods for long-distance advanced geological prediction of tunnels. It uses elastic waves (such as seismic waves) to test and obtain rock parameters such as surrounding rock wave velocity and Poisson's ratio, infer the surrounding rock strength, crack development and water content in front of the face, analyze adverse geological problems, and guide safe construction.

[0003] Depending on the different acquisition devices and interpretation principles, the tunnel elastic wave method of advanced geological prediction technology includes but is not limited to: elastic wave reflection method (TSP, Tunnel Seismic Prediction), true reflection tomography (TRT, True Reflection Tomography), tunnel geological prediction method (TGP, Tunnel Geology Prediction), tunnel seismic wave imaging method (TST, Tunnel Seismic Tomography), etc.

[0004] The elastic wave method in traditional technology is used for advanced geological prediction and analysis. After obtaining the rock mechanical parameter image of the elastic wave test, professionals will analyze the change amplitude, change trend, value range, etc. of multiple geophysical parameter images such as longitudinal wave velocity, transverse wave velocity, longitudinal and transverse wave velocity ratio, Young's modulus, Poisson's ratio, elastic modulus, etc., and combine the relevant conditions such as groundwater and rock integrity of the face, and describe the geological indicators such as surrounding rock integrity, water content, surrounding rock strength, etc. in the detection area through human judgment, and then further analyze the corresponding adverse geological problems.

[0005] This scheme has the following defects: (1) It relies on the operational judgment of professionals, resulting in high labor costs; (2) Since the elastic wave method parameter image data volume is large and the change trend is inconsistent, and the analyst's knowledge reserve, experience reserve and subjective ability are limited, it is easy to cause large errors in the analysis results; (3) Due to the large workload of analysis and the low efficiency of manual analysis, the timeliness of data analysis results cannot be guaranteed.

[0006] With the rapid development of artificial intelligence technologies such as machine learning in the field of tunnel engineering, the automated analysis of geophysical interpretation parameter images has become technically possible. However, existing research still faces problems such as small number of training samples, poor universality of intelligent model engineering, and low level of automation, which are specifically manifested as follows:

[0007] (1) Fewer training samples

[0008] Data is an important basis for determining the accuracy of machine learning models. In order to ensure the accuracy of intelligent computing, a large number of intelligent learning databases are required. Usually, machine learning data mainly comes from self-owned cases and reference cases. The number of self-owned cases is small and often cannot meet the needs of intelligent learning; while the parameters of reference cases are incomplete, and often only have parameter interpretation images, which are difficult to use directly.

[0009] (2) Intelligent model engineering has poor universality

[0010] Models are an important component of machine learning that reflects its learning results. The quality and stability of the model obtained through training determine the practicality of the entire machine learning method. Existing artificial intelligence prediction models directly use machine learning models to predict adverse geological problems, considering mathematical logic, with little research on the relationship between geophysical parameters, geological indicators, and geological quality. They are not based on geological logic theory, and the prediction accuracy fluctuates greatly, resulting in poor engineering universality.

[0011] (3) Low level of automation

[0012] Elastic wave method for advanced geological prediction requires the use of geophysical parameters to obtain geological indicators and further analyze the corresponding adverse geological problems. However, existing research results only conduct automated analysis on a single step, not the entire process, and the overall automation level is low. Summary of the invention

[0013] The technical problem to be solved by the present invention is to propose an automated analysis method and system for advanced tunnel geological prediction using elastic wave method, so as to solve the problems existing in the research on the application of artificial intelligence to advanced tunnel geological prediction analysis, such as small number of training samples, poor universality of intelligent model engineering, and low level of automation, and to improve the efficiency and accuracy of the analysis.

[0014] The technical solution adopted by the present invention to solve the above technical problems is:

[0015] On the one hand, the present invention provides an automatic analysis method for advanced geological prediction of tunnels using elastic wave method, comprising the following steps:

[0016] S1. Identify elastic wave method parameter images based on image processing technology to obtain elastic wave method parameter data;

[0017] S2. Establishing an elastic wave method parameter database based on elastic wave method parameter data;

[0018] S3. Select key geophysical parameters of elastic wave method for geological analysis;

[0019] S4. Establish the relationship between the key geophysical parameters of elastic wave method and geological indicators;

[0020] S5. Establish the relationship between geological indicators and geological quality;

[0021] S6. Based on the established relationship between the key geophysical parameters of the elastic wave method and the geological indicators and the relationship between the geological indicators and the geological quality, a classification principle of the classifier is established;

[0022] S7, using a machine learning method to construct a classifier model according to the classifier classification principle, obtaining training samples and test samples based on the elastic wave method parameter database, training the classifier model using the training samples, and verifying the trained classifier model using the test samples, and finally obtaining an elastic wave method tunnel advanced geological prediction analysis model that meets the requirements;

[0023] S8. According to the elastic wave method parameter image to be analyzed, the elastic wave method parameter image is processed based on image processing technology to obtain elastic wave method parameter data and extract elastic wave method key geophysical parameters; with the extracted elastic wave method key geophysical parameters as input, the elastic wave method tunnel advanced geological prediction analysis model is used to obtain the output geological quality evaluation result.

[0024] Furthermore, in step S1, the elastic wave method parameter image includes a longitudinal wave velocity image, a Poisson's ratio image and a Young's modulus image.

[0025] Furthermore, in step S1, the method of identifying the elastic wave method parameter image based on the image processing technology and obtaining the elastic wave method parameter data includes:

[0026] Through the corner point detection algorithm, the broken line corner points of the elastic wave method parameter images in historical reports and literature are automatically detected, identified and counted, the pixel coordinates of the corner points are obtained, and converted into actual coordinates to obtain the corresponding elastic wave method parameter data.

[0027] Furthermore, in step S2, the data sources in the elastic wave method parameter database are: elastic wave method parameter data directly obtained from historical reports and literature and / or elastic wave method parameter data obtained in step S1 based on image processing technology.

[0028] Furthermore, in step S3, the key geophysical parameters of the elastic wave method selected include:

[0029] P-wave velocity, P-wave velocity variation frequency and Poisson's ratio; wherein, P-wave velocity variation frequency is the ratio of the number of corner points in the selected area in the P-wave velocity image to the mileage length of the area.

[0030] Furthermore, in step S4, the geological indicators include: rock integrity, rock integrity change frequency and rock hardness; the established relationship between the key geophysical parameters of the elastic wave method and the geological indicators includes: the higher the longitudinal wave velocity, the better the rock integrity; the greater the frequency of longitudinal wave velocity change, the faster the rock integrity changes; the larger the Poisson's ratio, the more fluid content, and the softer the rock.

[0031] Further, in step S5, the geological quality includes: no bad geology, water-bearing areas, fault fracture zones or weak interlayers, water-bearing fault fracture zones or water-bearing weak interlayers, weak surrounding rocks and water-bearing caves;

[0032] The established relationships between geological indicators and geological quality include:

[0033] No adverse geology: rock integrity improves or changes little, the frequency of rock integrity changes decreases, and the fluid content decreases or remains basically unchanged;

[0034] Water-bearing area: rock integrity improves or changes little, the frequency of rock integrity changes decreases, and fluid content increases;

[0035] Fault fracture zone or weak interlayer: rock integrity deteriorates, the frequency of rock integrity changes increases or remains basically unchanged, and the fluid content remains basically unchanged;

[0036] Water-bearing fault fracture zone or water-bearing weak interlayer: rock mass integrity deteriorates, the frequency of rock mass integrity changes increases or remains basically unchanged, and fluid content increases;

[0037] Weak surrounding rock: rock mass integrity deteriorates, the frequency of rock mass integrity changes decreases, and the fluid content decreases or remains basically unchanged;

[0038] Water-bearing caves: rock integrity deteriorates, the frequency of rock integrity changes decreases, and fluid content increases.

[0039] Furthermore, in step S6, the classification principle of the established classifier includes:

[0040] Select the base value of longitudinal wave velocity , Poisson ratio base value , Base value of longitudinal wave velocity change frequency ;

[0041] calculate , , ;

[0042] in, 、 、 They are the longitudinal wave velocity increment, the longitudinal wave velocity change frequency increment and the Poisson's ratio increment respectively; 、 、 They are the current longitudinal wave velocity, the current longitudinal wave velocity change frequency and the current Poisson’s ratio respectively;

[0043] The established classification principles are:

[0044] No adverse geology: , , ;

[0045] Water-bearing area: , , ;

[0046] Fault fracture zone or weak interlayer: , , ;

[0047] Water-bearing fault fracture zone or water-bearing weak interlayer: , , ;

[0048] Weak surrounding rock: , , ;

[0049] Water-bearing caves: , , ;

[0050] in, It is the normal fluctuation ratio of the set parameters.

[0051] Furthermore, in step S7, the machine learning method includes but is not limited to: a naive Bayes algorithm, a decision tree algorithm or a random forest algorithm; the indicators used when verifying the trained classifier model using test samples include: accuracy, precision, recall, F1 value and confusion matrix.

[0052] On the other hand, the present invention also provides an automated analysis system for advanced geological prediction of tunnels using the elastic wave method, which comprises: a processor and a memory, wherein the memory stores a computer program, and the processor implements the steps of the automated analysis method for advanced geological prediction of tunnels using the elastic wave method by executing the computer program.

[0053] The beneficial effects of the present invention are:

[0054] (1) The elastic wave method parameter images in historical reports and literature are identified through image processing technology to obtain elastic wave method parameter data. Based on this, an elastic wave method parameter database is established to obtain rich training samples that can be directly applied, laying the foundation for training accurate analysis models.

[0055] (2) Combining the rock mechanics theory and geological disaster mechanism, a mapping relationship between key geophysical parameters of the elastic wave method, geological indicators and geological quality was established. Based on this, the classification principle of the classifier was established, so that the constructed classifier model has strong interpretability and generalization ability, thus making the model more universal in engineering.

[0056] (3) The elastic wave method for tunnel geological prediction analysis based on the established classifier model can realize real-time automated analysis of the entire process, thereby improving the accuracy and efficiency of the analysis and providing scientific guidance for safe and efficient tunnel construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of a training method of an elastic wave method tunnel advanced geological prediction analysis model in an embodiment of the present invention;

[0058] Figure 2 It is a mapping relationship diagram of key geophysical parameters of elastic wave method-geological indicators-geological quality in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The present invention aims to provide an automated analysis method and system for advanced tunnel geological prediction using elastic wave method, so as to solve the problems existing in the research on the application of artificial intelligence to advanced tunnel geological prediction analysis, such as small number of training samples, poor universality of intelligent model engineering, and low level of automation, and to improve the efficiency and accuracy of analysis. The core idea is: in terms of the source of training samples for the model, the elastic wave method parameter images in historical reports and literature are identified through image processing technology to obtain elastic wave method parameter data, thereby establishing an elastic wave method parameter database, thereby obtaining rich training samples that can be directly applied; in terms of the construction of the model's classifier classification principle, the mapping relationship between the key geophysical parameters of the elastic wave method-geological indicators-geological quality is established in combination with rock mechanics theory, geological disaster mechanism, etc., and the classifier classification principle is established based on this, so that the constructed classifier model has strong interpretability and generalization ability, thereby making the model have stronger engineering universality; finally, a machine learning method is used to construct a classifier model according to the classifier classification principle, and training samples and test samples are obtained based on the elastic wave method parameter database. The classifier model is trained using the training samples, and the trained classifier model is verified using the test samples, and finally a satisfactory elastic wave tunnel advanced geological prediction analysis model is obtained.

[0060] In actual application scenarios, by deploying this analysis model, it can be used to conduct real-time analysis of geological quality, specifically: according to the elastic wave method parameter image to be analyzed, the elastic wave method parameter image is processed based on image processing technology to obtain elastic wave method parameter data and extract elastic wave method key geophysical parameters; with the extracted elastic wave method key geophysical parameters as input, the output geological quality evaluation results are obtained using the above-trained elastic wave method tunnel advanced geological prediction analysis model.

[0061] The scheme of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0062] Example

[0063] In the automatic analysis method for advanced geological prediction of tunnels using elastic wave method provided in this embodiment, the elastic wave method tunnel advanced geological prediction analysis model is first trained, and then the automatic analysis is performed using the model. Figure 1 , including the following implementation steps:

[0064] S1. Identification of elastic wave parameter images

[0065] In this step, elastic wave method parameter images are identified based on image processing technology to obtain elastic wave method parameter data. In an exemplary solution, elastic wave method parameter images include longitudinal wave velocity images, Poisson's ratio images, and Young's modulus images. The corner points of the broken lines of these elastic wave method parameter images in historical reports and literature can be automatically detected, identified, and counted by using a corner point detection algorithm to obtain the pixel coordinates of the corner points, which are converted into actual coordinates to obtain the corresponding elastic wave method parameter data.

[0066] S2. Establishing elastic wave method parameter database

[0067] In this step, an elastic wave method parameter database is established based on the elastic wave method parameter data. In an exemplary solution, the elastic wave method parameter database is established directly from the elastic wave method parameter data obtained from historical reports and literature, and / or the elastic wave method parameter data obtained in step S1 based on image processing technology, and is used as a data source for subsequent model training and verification.

[0068] S3. Selecting key geophysical parameters of elastic wave method

[0069] In this step, according to the principle of advanced geophysical exploration by elastic wave method, the key geophysical exploration parameters of elastic wave method for geological analysis are selected. In an exemplary scheme, the key geophysical exploration parameters of elastic wave method selected include P-wave velocity, P-wave velocity variation frequency and Poisson's ratio, etc. Among them, P-wave velocity variation frequency is the ratio of the number of corner points of the selected area in the P-wave velocity image to the mileage length of the area. These parameters can more significantly reflect the situation of geological indicators.

[0070] S4. Establish the relationship between key geophysical parameters of elastic wave method and geological indicators

[0071] In this step, the relationship between the key geophysical parameters of the elastic wave method and the geological indicators is established based on the rock mechanics theory. In an exemplary scheme, the geological indicators include: rock integrity, rock integrity change frequency and rock hardness. These indicators play a key guiding role in determining the stability of the surrounding rock. The established relationship between the key geophysical parameters of the elastic wave method and the geological indicators includes: the higher the longitudinal wave velocity, the better the rock integrity; the greater the frequency of longitudinal wave velocity change, the faster the rock integrity changes; the larger the Poisson's ratio, the more fluid content, and the softer the rock.

[0072] S5. Establish the relationship between geological indicators and geological quality

[0073] In this step, the relationship between geological indicators and geological quality is established based on the geological disaster mechanism. In an exemplary scheme, the geological quality includes: no bad geology, water-bearing areas, fault fracture zones or weak interlayers, water-bearing fault fracture zones or water-bearing weak interlayers, weak surrounding rocks and water-bearing caves, etc. This quality evaluation system can basically cover various surrounding rock geological conditions. The established relationship between geological indicators and geological quality includes:

[0074] No adverse geology: rock integrity improves or changes little, the frequency of rock integrity changes decreases, and the fluid content decreases or remains basically unchanged;

[0075] Water-bearing area: rock integrity improves or changes little, the frequency of rock integrity changes decreases, and fluid content increases;

[0076] Fault fracture zone or weak interlayer: rock integrity deteriorates, the frequency of rock integrity changes increases or remains basically unchanged, and the fluid content remains basically unchanged;

[0077] Water-bearing fault fracture zone or water-bearing weak interlayer: rock mass integrity deteriorates, the frequency of rock mass integrity changes increases or remains basically unchanged, and fluid content increases;

[0078] Weak surrounding rock: rock mass integrity deteriorates, the frequency of rock mass integrity changes decreases, and the fluid content decreases or remains basically unchanged;

[0079] Water-bearing caves: rock integrity deteriorates, the frequency of rock integrity changes decreases, and fluid content increases.

[0080] S6. Establishing Classifier Classification Principles

[0081] In this step, the classifier classification principle is established in combination with the established relationship between the key geophysical parameters of the elastic wave method and the geological indicators, as well as the relationship between the geological indicators and the geological quality. Among them, the classification principle is the different geological quality evaluation results corresponding to the parameter ranges of the key geophysical parameters of the elastic wave method. On the basis of the aforementioned relationship between the key geophysical parameters of the elastic wave method and the geological indicators, the relationship between the geological indicators, and the relationship between the geological indicators and the geological quality, the geological indicators can be used as a bridge to build a mapping relationship between the key geophysical parameters of the elastic wave method, the geological indicators, and the geological quality, such as Figure 2 On this basis, in an exemplary solution, the process of establishing the classification principle of the classifier includes:

[0082] Select the base value of longitudinal wave velocity , Poisson ratio base value , Base value of longitudinal wave velocity change frequency ;

[0083] calculate , , ;

[0084] in, 、 、 They are the longitudinal wave velocity increment, the longitudinal wave velocity change frequency increment and the Poisson's ratio increment respectively; 、 、 They are the current longitudinal wave velocity, the current longitudinal wave velocity change frequency and the current Poisson’s ratio respectively;

[0085] The classification principles established are:

[0086] No adverse geology: , , ;

[0087] Water-bearing area: , , ;

[0088] Fault fracture zone or weak interlayer: , , ;

[0089] Water-bearing fault fracture zone or water-bearing weak interlayer: , , ;

[0090] Weak surrounding rock: , , ;

[0091] Water-bearing caves: , , ;

[0092] in, It is the normal fluctuation ratio of the set parameters.

[0093] The above-mentioned classification principle of the classifier serves as the judgment basis for the classifier when performing classification tasks.

[0094] S7. Model training and validation

[0095] In this step, a machine learning method is used to construct a classifier model according to the classifier classification principle, and training samples and test samples are obtained based on the elastic wave method parameter database. The classifier model is trained using the training samples, and the trained classifier model is verified using the test samples. If the accuracy of the model does not meet the requirements, this step is repeated until a satisfactory elastic wave tunnel advance geological prediction analysis model is finally obtained. In an exemplary scheme, the machine learning method includes but is not limited to: naive Bayes algorithm, decision tree algorithm or random forest algorithm, etc.; the indicators used to verify the trained classifier model using the test samples include: accuracy, precision, recall, F1 value and confusion matrix, etc.

[0096] The above process is the training process of the model. After obtaining the elastic wave method tunnel advanced geological prediction analysis model that meets the requirements according to the above process, the model can be deployed in the actual application scenario. Based on this, this embodiment also provides an elastic wave method tunnel advanced geological prediction analysis method, which includes: according to the elastic wave method parameter image to be analyzed, the elastic wave method parameter image is processed based on the image processing technology to obtain elastic wave method parameter data and extract elastic wave method key geophysical parameters; using the extracted elastic wave method key geophysical parameters as input, the above-trained elastic wave method tunnel advanced geological prediction analysis model is used to obtain the output geological quality evaluation results. This solution can realize real-time full-process automated analysis, thereby improving the accuracy and efficiency of the analysis and providing scientific guidance for safe and efficient tunnel construction.

[0097] Finally, it should be noted that the above embodiments are only preferred implementations and are not intended to limit the present invention. It should be pointed out that for those skilled in the art, several modifications, equivalent replacements, improvements, etc. can be made without departing from the scope of the present invention and the scope of protection of the claims, and all of these should be included in the protection scope of the present invention.

Claims

1. The automatic analysis method for advanced geological prediction of tunnels using elastic wave method is characterized by: The following steps are involved: S1. Identify elastic wave method parameter images based on image processing technology to obtain elastic wave method parameter data; S2. Establishing an elastic wave method parameter database based on elastic wave method parameter data; S3. Select key geophysical parameters of elastic wave method for geological analysis; S4. Establish the relationship between the key geophysical parameters of elastic wave method and geological indicators; S5. Establish the relationship between geological indicators and geological quality; S6. Based on the established relationship between the key geophysical parameters of the elastic wave method and the geological indicators and the relationship between the geological indicators and the geological quality, a classification principle of the classifier is established; S7, using a machine learning method to construct a classifier model according to the classifier classification principle, obtaining training samples and test samples based on the elastic wave method parameter database, training the classifier model using the training samples, and verifying the trained classifier model using the test samples, and finally obtaining an elastic wave method tunnel advanced geological prediction analysis model that meets the requirements; S8. According to the elastic wave method parameter image to be analyzed, the elastic wave method parameter image is processed based on image processing technology to obtain elastic wave method parameter data and extract key elastic wave method geophysical exploration parameters; Taking the extracted key geophysical parameters of elastic wave method as input, the output geological quality evaluation results are obtained by using the elastic wave method tunnel advanced geological prediction analysis model; In step S3, the key geophysical parameters of the elastic wave method selected include: P-wave velocity, P-wave velocity variation frequency and Poisson's ratio; wherein the P-wave velocity variation frequency is the ratio of the number of corner points in the selected area in the P-wave velocity image to the mileage length of the area; In step S4, the geological indicators include: rock integrity, rock integrity change frequency and rock hardness; the established relationship between the key geophysical parameters of the elastic wave method and the geological indicators includes: the higher the longitudinal wave velocity, the better the rock integrity; the greater the longitudinal wave velocity change frequency, the faster the rock integrity changes; the greater the Poisson's ratio, the more fluid content, and the softer the rock; In step S5, the geological quality includes: no bad geology, water-bearing areas, fault fracture zones or weak interlayers, water-bearing fault fracture zones or water-bearing weak interlayers, weak surrounding rocks and water-bearing karst caves; The established relationships between geological indicators and geological quality include: No adverse geology: rock integrity improves or changes little, the frequency of rock integrity changes decreases, and the fluid content decreases or remains basically unchanged; Water-bearing area: rock integrity improves or changes little, the frequency of rock integrity changes decreases, and fluid content increases; Fault fracture zone or weak interlayer: rock integrity deteriorates, the frequency of rock integrity changes increases or remains basically unchanged, and the fluid content remains basically unchanged; Water-bearing fault fracture zone or water-bearing weak interlayer: rock mass integrity deteriorates, the frequency of rock mass integrity changes increases or remains basically unchanged, and fluid content increases; Weak surrounding rock: rock mass integrity deteriorates, the frequency of rock mass integrity changes decreases, and the fluid content decreases or remains basically unchanged; Water-bearing caves: rock integrity deteriorates, the frequency of rock integrity changes decreases, and fluid content increases; In step S6, the classification principles of the established classifier include: Select the base value of longitudinal wave velocity v p0 , Poisson's ratio base value μ0, longitudinal wave velocity change frequency base value N0; calculationΔv p =v p -v p0 ,ΔN=N-N0,Δμ=μ-μ0; Where Δv p , ΔN, Δμ are the longitudinal wave velocity increment, the longitudinal wave velocity change frequency increment and the Poisson's ratio increment respectively; v p , N, μ are the current P-wave velocity, the current P-wave velocity change frequency and the current Poisson’s ratio respectively; The classification principles established are: No adverse geology: Δv p >-a×v p0 , -a×N>ΔN, Δμ>-a×μ; Water-bearing area: Δv p >-a×v p0 , -a×N0>ΔN, Δμ>-a×μ0; Fault fracture zone or weak interlayer: -a×v p0 >Δv p , ΔN>-a×N0, a×μ0>Δμ>-a×μ0; Water-bearing fault fracture zone or water-bearing weak interlayer: -a×v p0 >Δv p , ΔN>-a×N0, Δμ>a×μ0; Weak surrounding rock: -a×v p0 >Δv p , -a×N>ΔN, -a×μ0>Δμ; Water-bearing cave: -a×v p0 >Δv p , -a×N>ΔN, Δμ>a×μ0; Among them, a is the normal fluctuation ratio of the set parameter.

2. The automatic analysis method for advanced geological prediction of tunnels using elastic wave method according to claim 1, characterized in that: In step S1, the elastic wave method parameter images include a longitudinal wave velocity image, a Poisson's ratio image and a Young's modulus image.

3. The automatic analysis method for advanced geological prediction of tunnels using elastic wave method according to claim 1, characterized in that: In step S1, the method of identifying the elastic wave method parameter image based on the image processing technology and obtaining the elastic wave method parameter data includes: Through the corner point detection algorithm, the broken line corner points of the elastic wave method parameter images in historical reports and literature are automatically detected, identified and counted, the pixel coordinates of the corner points are obtained, and converted into actual coordinates to obtain the corresponding elastic wave method parameter data.

4. The automatic analysis method for advanced geological prediction of tunnels using elastic wave method according to claim 1, characterized in that: In step S2, the data sources in the elastic wave method parameter database are: elastic wave method parameter data directly obtained from historical reports and literature and / or elastic wave method parameter data obtained in step S1 based on image processing technology.

5. The automatic analysis method for advanced geological prediction of tunnels using elastic wave method according to any one of claims 1 to 4, characterized in that: In step S7, the machine learning method includes: naive Bayes algorithm, decision tree algorithm or random forest algorithm; the indicators used when verifying the trained classifier model using test samples include: accuracy, precision, recall, F1 value and confusion matrix.

6. Elastic wave tunnel advanced geological prediction automatic analysis system, which includes: A processor and a memory, wherein a computer program is stored in the memory, and wherein the processor implements the steps of the automated analysis method for advanced geological prediction of tunnels using the elastic wave method as described in any one of claims 1 to 5 by executing the computer program.

Images