Tunnel surrounding rock mechanical parameter intelligent analysis method and system based on while-drilling parameters

Abstract

The invention belongs to the technical field of tunnel intelligent construction, and discloses a tunnel surrounding rock mechanical parameter intelligent analysis method and system based on parameters while drilling, and the method comprises the following steps: constructing a tunnel surrounding rock mechanical parameter intelligent analysis sample database and feature system based on the parameters while drilling; adopting a transfer learning algorithm to construct a tunnel surrounding rock sublevel intelligent grading model based on the while-drilling parameters; and based on the grading probability vector output by the tunnel surrounding rock sublevel intelligent grading model and the surrounding rock mechanical parameter characteristic value of each sublevel, analyzing to obtain the elastic modulus, Poisson's ratio, cohesive force and internal friction angle of the surrounding rock. According to the grading probability vector output by the tunnel surrounding rock sublevel intelligent grading model based on the while-drilling parameters, the mechanical parameter characteristic value of each sublevel surrounding rock is combined, and intelligent analysis of the mechanical parameters of the tunnel surrounding rock is achieved. The tunnel surrounding rock mechanical parameter acquisition efficiency is improved, the tunnel construction cost is reduced, the construction efficiency and safety are improved, and guidance is provided for intelligent tunnel construction.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent tunnel construction, and in particular to a method and system for intelligently analyzing mechanical parameters of tunnel surrounding rocks based on drilling parameters. Background Art

[0002] With the accelerating aging of the population and the growing labor shortage, the transition of tunnel engineering toward less- and even unmanned construction models has become a key development direction for the industry. Furthermore, as tunnel engineering continues to expand into deeper and more geologically complex areas, the acquisition of surrounding rock mechanical parameters provides a scientific and reliable reference for the formulation of subsequent construction plans, playing a vital role in ensuring safe and efficient tunnel construction.

[0003] Traditional methods for obtaining tunnel surrounding rock mechanical parameters, such as on-site sampling followed by indoor or field testing, are costly and time-consuming, making them inadequate for intelligent construction. With the advancement of tunneling technology, a large number of tunnel face parameters can be acquired in real time while drilling. Using artificial intelligence technologies such as machine learning, intelligent analysis of surrounding rock mechanical parameters can be performed rapidly, providing a scientific and reliable reference for the formulation of subsequent construction plans. This effectively ensures the safety of intelligent tunnel construction and reduces project costs. Summary of the Invention

[0004] In order to solve the problems existing in the prior art, the present invention provides a method and system for intelligent analysis of tunnel surrounding rock mechanical parameters based on drilling parameters, which can realize intelligent analysis of tunnel surrounding rock mechanical parameters and solve the problems mentioned in the above background technology.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solution: an intelligent analysis method for tunnel surrounding rock mechanical parameters based on drilling parameters, comprising the following steps:

[0006] S1. Construct a sample database for intelligent analysis of tunnel surrounding rock mechanical parameters based on drilling parameters;

[0007] S2. Construct an intelligent analytical feature system for tunnel surrounding rock mechanical parameters based on drilling parameters;

[0008] S3. Using transfer learning algorithm, construct an intelligent classification model of tunnel surrounding rock sub-grades based on drilling parameters;

[0009] S4. Based on the classification probability vector output by the tunnel surrounding rock sub-class intelligent classification model and the characteristic values ​​of the surrounding rock mechanical parameters of each sub-class, the surrounding rock elastic modulus, Poisson's ratio, cohesion, and internal friction angle are analytically obtained.

[0010] Preferably, in step S1, the sample database includes four original drilling parameters of rotation pressure, feed speed, impact pressure and thrust pressure automatically collected by the intelligent drilling rig; the surrounding rock levels of the face include level II, level III1, level III2, level IV1, level IV2, level V1 and level V2.

[0011] Preferably, in step S2, an intelligent analytical feature system for tunnel surrounding rock mechanical parameters based on drilling parameters is constructed, which specifically includes the following steps:

[0012] S21. Based on the four original drilling parameters, five rock drillability indices (equivalent feed rate, equivalent propulsion force, difficulty-to-drill index, E index, and equivalent impact force) are calculated; four energy indices (propulsion specific energy, impact specific energy, rotation specific energy, and mechanical specific energy) are calculated.

[0013] S22. Taking the tunnel face excavation cycle as the unit, six statistical characteristics of four original drilling parameters, five rock drillability indicators, and four energy indicators are collected, including standard deviation, first quartile, second quartile, third quartile, coefficient of variation, and mean; and a 78-characteristic system of drilling parameters is constructed.

[0014] Preferably, in step S21, the calculation formulas for the five rock drillability indices and the four energy indices are:

[0015]

[0016]

[0017] Where: δ ff is the equivalent propulsion force; D is the drill diameter; P r is the rotation pressure; q r is the motor displacement; i r is the reduction ratio; P f is the propulsion pressure; D f is the rear end diameter of the thrust piston; δ fh is the equivalent striking force; D ca 、D cb Respectively, the diameter of the rear and front ends of the striking cylinder piston; m h is the piston mass; t h is the time of impact on the fiber tail; P h is the striking pressure; v is the equivalent feed rate; V r is the drilling tool speed; V d is the feed rate; δ hard is the difficulty to drill indicator; E is the E indicator; Ψ h is the specific energy of strike; S h is the impact stroke; r is the rotational specific energy;t is the propulsion specific energy; m is the mechanical specific energy.

[0018] Preferably, in step S22, the data points of the five rock drillability indicators and the four energy indicators are arranged in ascending order, and then the data are evenly divided into four intervals of equal length, thereby determining the values ​​at the three dividing points; among them, the first quartile Q1 is at the 25th position after the data is sorted, the second quartile Q2, which is the median, is at the 50th position; the third quartile Q3 corresponds to the 75th position; the mean reflects the trend of the data set, the standard deviation measures the degree of dispersion, and the coefficient of variation is the ratio of the standard deviation to the mean, which is used to compare the discreteness of data of different dimensions.

[0019] Preferably, in step S3, a transfer learning algorithm is used to construct a tunnel surrounding rock sub-class intelligent classification model based on drilling parameters, which specifically includes the following steps:

[0020] S31. Training a source model for intelligent sub-classification of tunnel surrounding rock based on source-domain while-drilling parameters: The sample database is normalized and proportionally divided into two datasets: the source domain and the target domain. The 78 features of the while-drilling parameters in the source domain dataset are horizontally concatenated into a 1×78 input feature vector as input, and a convolutional neural network (CNN) is used for feature extraction and classification. The network structure includes convolutional layers, pooling layers, and fully connected layers. The fully connected layers use the Softmax activation function, and employ the cross-entropy loss function and Adam adaptive learning rate optimization to calculate the loss and optimize the model's weights and bias parameters to complete source model training.

[0021] S32. Train a transfer learning model for intelligent sub-class grading of tunnel surrounding rock based on target domain LWD parameters: Load the source model trained on the source domain dataset, then freeze the convolutional layer parameters and the fully connected layer parameters except for the last two layers, so that the weights and bias parameters of these layers remain unchanged during the transfer learning process and are not updated. Concatenate the 78 LWD parameter features of the target domain dataset horizontally into a 1×78 input feature vector as the target data input, then fine-tune based on the target data. Use the Adam optimizer to retrain the fully connected layer of the model to adjust the fully connected layer parameters.

[0022] Through the above steps, the source model is adapted to the target domain data, and a tunnel surrounding rock sub-level intelligent classification transfer learning model is obtained, completing the construction of a tunnel surrounding rock sub-level intelligent classification model based on drilling parameters.

[0023] Preferably, in step S4, based on the classification probability vector output by the tunnel surrounding rock sub-class intelligent classification model and the characteristic values ​​of the surrounding rock mechanical parameters of each sub-class, the elastic modulus, Poisson's ratio, cohesion, and internal friction angle of the surrounding rock are analytically obtained, specifically including the following:

[0024] S41. Calculate the characteristic values ​​of the mechanical parameters of the surrounding rock, such as elastic modulus, Poisson's ratio, cohesion, and internal friction angle, for each sub-level;

[0025] S42. Analyze the surrounding rock mechanical parameters based on the classification probability vector output by the tunnel surrounding rock sub-class intelligent classification model and the characteristic values ​​of the mechanical parameters of each sub-class surrounding rock.

[0026] Preferably, in step S41: according to the "Code for Design of Railway Tunnel TB 10003-2016", the average of the upper and lower limits of the surrounding rock mechanical parameters of each sub-level is taken as its characteristic value; the characteristic values ​​of the elastic modulus corresponding to levels II to V2 in the tunnel face surrounding rock levels are 25, 15.35, 8.35, 4.9, 2.55, 1.65, and 1.15 GPa respectively; the characteristic values ​​of the Poisson's ratio corresponding to levels II to V2 are 0.225, 0.255, 0.28, 0.305, 0.33, 0.37, and 0.42; the characteristic values ​​of the internal friction angle corresponding to levels II to V2 are 55, 47, 41.5, 37, 31, 24.5, and 21°; and the characteristic values ​​of the cohesion corresponding to levels II to V2 are 1.8, 1.3, 0.9, 0.6, 0.35, 0.16, and 0.085 MPa.

[0027] Preferably, in step S42, the elastic modulus, Poisson's ratio, cohesion, and internal friction angle of the surrounding rock are calculated based on the classification probability vector of the surrounding rock sub-class and the characteristic values ​​of each surrounding rock mechanical parameter. The calculation formula is:

[0028]

[0029] Where: y E 、y ν 、y C 、 are the analytical results of the tunnel surrounding rock elastic modulus, Poisson's ratio, cohesion, and internal friction angle; E i 、ν i 、C i 、 are the characteristic values ​​of subtype i; P i is the probability of sub-class type i output by the tunnel surrounding rock sub-class intelligent classification model.

[0030] On the other hand, to achieve the above-mentioned purpose, the present invention further provides the following technical solution: an intelligent analysis system for tunnel surrounding rock mechanical parameters based on drilling parameters, comprising the following modules:

[0031] Sample database construction module: Build a sample database, which includes the drilling parameters of the tunnel face and the corresponding surrounding rock sub-grades of the tunnel face;

[0032] Characteristic system construction module: calculates rock drillability index, energy index, statistical characteristics, and constructs a 78-characteristic system of drilling parameters;

[0033] Tunnel surrounding rock sub-class intelligent classification model construction module: uses transfer learning algorithm to build a tunnel surrounding rock sub-class intelligent classification model based on drilling parameters;

[0034] Tunnel surrounding rock mechanical parameter analysis module: Based on the surrounding rock sub-class classification probability vector and the characteristic values ​​of the surrounding rock mechanical parameters of each sub-class, the surrounding rock mechanical parameters are obtained through analysis.

[0035] The beneficial effects of the present invention are: based on the drilling parameters of the tunnel face collected during the tunnel construction process, the present invention realizes rapid and real-time intelligent analysis of the surrounding rock mechanical parameters through artificial intelligence technologies such as transfer learning, providing a scientific and reliable reference basis for the formulation of subsequent construction plans, effectively ensuring the safety of intelligent tunnel construction, reducing engineering costs, and facilitating intelligent tunnel construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Schematic diagram of the flow of an intelligent analysis method for tunnel surrounding rock mechanical parameters based on drilling parameters in an embodiment of the present invention;

[0037] Figure 2 A schematic diagram of a process for constructing a while-drilling parameter feature system in an embodiment of the present invention;

[0038] Figure 3 A schematic diagram of a process for constructing a tunnel surrounding rock sub-classification model based on drilling parameters and analyzing surrounding rock mechanical parameters in an embodiment of the present invention;

[0039] Figure 4 Schematic diagram of a module of an intelligent analysis system for tunnel surrounding rock mechanical parameters based on while-drilling parameters in an embodiment of the present invention;

[0040] In the figure, 110 is a sample database construction module; 120 is a feature system construction module; 130 is a tunnel surrounding rock sub-class intelligent classification model construction module; and 140 is a tunnel surrounding rock mechanical parameter analysis module. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] Example 1

[0043] During tunnel construction, the mechanical parameters of the tunnel's surrounding rock reflect its physical and mechanical properties and serve as a crucial basis for subsequent construction planning. Traditional methods for obtaining these parameters, such as on-site sampling followed by laboratory testing or field testing, are costly and time-consuming, making them inadequate for intelligent construction.

[0044] To this end, through extensive research and practice, the inventors have proposed a method for rapidly and intelligently analyzing tunnel rock mechanical parameters using a transfer learning algorithm, based on face-while-drilling parameters collected during tunnel excavation. This method aims to provide a fast, automated method for analyzing tunnel rock mechanical parameters, improve the efficiency of obtaining tunnel rock mechanical parameters, reduce tunnel construction costs, and enhance the level of intelligent tunnel construction.

[0045] See also Figure 1 The present invention provides a technical solution: a method and system for intelligently analyzing mechanical parameters of tunnel surrounding rocks based on drilling parameters, which includes the following steps:

[0046] Step S1: Construct a sample database of intelligent analysis of tunnel surrounding rock mechanical parameters based on drilling parameters.

[0047] The sample database includes four drilling parameters automatically collected by the intelligent drilling rig: rotation pressure, feed speed, impact pressure, and thrust pressure. The surrounding rock grade of the tunnel face is determined using standard methods and includes grades II, III1, III2, IV1, IV2, V1, and V2. Table 1 shows an example of data recorded by the intelligent drilling rig for a certain blasthole on a certain tunnel face.

[0048] Table 1 Example of data of a certain blasthole on a tunnel face recorded by the intelligent drilling rig

[0049]

[0050]

[0051] The surrounding rock grades of the tunnel faces in the sample library are classified according to the Railway Tunnel Design Code TB10003-2016. The surrounding rock grades can be divided into I to VI. In actual projects, the common surrounding rock grades are II to V. Grades II to V are further sub-classified according to the code, and the surrounding rock sub-grades are III1 to V2.

[0052] Step S2: Construct an intelligent analytical feature system for tunnel surrounding rock mechanical parameters based on drilling parameters.

[0053] S21, such as Figure 2As shown in the figure, based on the four original drilling parameters, five rock drillability indices, including equivalent feed rate, equivalent propulsion force, difficulty-to-drill index, E index, and equivalent impact force, are calculated; four energy indices, including propulsion specific energy, impact specific energy, rotation specific energy, and mechanical specific energy, are calculated using the following formula:

[0054]

[0055] Where: δ ff is the equivalent thrust, in N; D is the drill diameter, in mm; P r is the rotation pressure, unit Pa; q r is the motor displacement, unit is ml / r; i r is the reduction ratio; P f is the propulsion pressure, in bar; D f is the rear end diameter of the propulsion piston, in mm; δ fh is the equivalent striking force, unit N; D ca 、D cb Respectively, the diameter of the rear and front ends of the striking cylinder piston, in mm; m h is the piston mass; t h is the time of impacting the fiber tail, in seconds; P h is the impact pressure, unit bar; δ v is the equivalent feed rate; V r is the drilling tool speed, unit is r / min; V d is the feed rate, in m / min; δ hard is the difficulty to drill indicator; E is the E indicator; Ψ h is the impact specific energy, unit is Pa; S h is the impact stroke, in mm; r is the rotational specific energy, unit is Pa; t is the propulsion specific energy, in Pa; m It is the mechanical specific energy, unit is Pa.

[0056] S22, such as Figure 2 As shown in the figure, taking the tunnel face excavation cycle as the unit, six statistical features of four original drilling parameters, five rock drillability indices, and four energy indices are statistically analyzed, including standard deviation, first quartile, second quartile, third quartile, coefficient of variation, and mean, to construct a 78-feature system of drilling parameters.

[0057] The data points of the five rock drillability indicators and four energy indicators were arranged in ascending order, and then the data were evenly divided into four equal-length intervals to determine the values ​​at the three division points. Among them, the first quartile Q1 is located at the 25th percentile position after the data is sorted, the second quartile Q2 is the median, located at the 50th percentile position; and the third quartile Q3 corresponds to the 75th percentile position. The mean reflects the trend of the data set, the standard deviation measures the degree of dispersion, and the coefficient of variation is the ratio of the standard deviation to the mean, which is used to compare the dispersion of data of different dimensions.

[0058] Step S3: Using the transfer learning algorithm, a tunnel surrounding rock sub-class intelligent classification model based on drilling parameters is constructed.

[0059] S31. Training a tunnel surrounding rock sub-level intelligent classification source model based on source domain drilling parameters;

[0060] like Figure 3 As shown in the figure, the sample database is normalized and proportionally divided into two datasets: the source domain and the target domain. The 78 features of the downhole parameters in the source domain are horizontally concatenated into a 1×78 input feature vector. A convolutional neural network (CNN) is used for feature extraction and classification. The network structure consists of convolutional layers, pooling layers, and fully connected layers. The fully connected layers use the Softmax activation function. Cross-entropy loss and Adam adaptive learning rate optimization are used to calculate the loss and optimize model parameters such as weights and biases to complete model training.

[0061] S32, training a tunnel surrounding rock sub-level intelligent classification transfer learning model based on target domain drilling parameters;

[0062] like Figure 3 As shown in the figure, the source model trained on the source domain data is loaded, and the parameters of the convolutional layers and the fully connected layers (except for the last two layers) are frozen, so that the weights and bias parameters of these layers remain unchanged and are not updated during the transfer learning process. The 78 features of the target dataset's while-drilling parameters are horizontally concatenated into a 1×78 input feature vector, which is then used as the target data input. Fine-tuning is then performed based on the target data, and the fully connected layers of the model are retrained using the Adam optimizer to adjust the parameters of the fully connected layers. Through these steps, the source model is adapted to the target domain data, resulting in a transfer learning model for intelligent sub-classification of tunnel surrounding rock. This completes the construction of an intelligent sub-classification model for tunnel surrounding rock based on while-drilling parameters.

[0063] Step S4: Based on the classification probability vector output by the tunnel surrounding rock sub-class intelligent classification model and the characteristic values ​​of the surrounding rock mechanical parameters of each sub-class, the surrounding rock elastic modulus, Poisson's ratio, cohesion, and internal friction angle are analytically obtained.

[0064] S41, such as Figure 3As shown in the figure, the characteristic values ​​of mechanical parameters such as elastic modulus, Poisson's ratio, cohesion, internal friction angle, etc. of surrounding rock of each sub-level are calculated; according to the Railway Tunnel Design Code TB 10003-2016", the average of the upper and lower limits of the surrounding rock mechanical parameters of each sub-level is taken as its characteristic value; the characteristic values ​​of elastic modulus corresponding to levels II to V2 are 25, 15.35, 8.35, 4.9, 2.55, 1.65, and 1.15 GPa, respectively; the characteristic values ​​of Poisson's ratio corresponding to levels II to V2 are 0.225, 0.255, 0.28, 0.305, 0.33, 0.37, and 0.42, the characteristic values ​​of internal friction angle corresponding to levels II to V2 are 55, 47, 41.5, 37, 31, 24.5, and 21°, and the characteristic values ​​of cohesion corresponding to levels II to V2 are 1.8, 1.3, 0.9, 0.6, 0.35, 0.16, and 0.085 MPa.

[0065] S42, such as Figure 3 As shown in Figure 2, the elastic modulus, Poisson's ratio, cohesion, and internal friction angle of the surrounding rock are calculated based on the classification probability vector of the surrounding rock subclass and the characteristic values ​​of each surrounding rock mechanical parameter. The calculation formula is:

[0066]

[0067] Where: y E 、y ν 、y C 、 They are the analytical results of the tunnel surrounding rock elastic modulus, Poisson's ratio, cohesion, and internal friction angle, respectively. E 、y C 、 The units are GPa, MPa, °; E i 、ν i 、C i 、 are the eigenvalues ​​of subtype i, E i 、C i 、 The units are GPa, MPa, °; P i is the probability of sub-class type i output by the tunnel surrounding rock sub-class intelligent classification model.

[0068] Table 2 shows an example of the intelligent analysis results of the mechanical parameters of the surrounding rock of a typical tunnel.

[0069] Table 2 Examples of intelligent analysis results of mechanical parameters of surrounding rock of typical tunnels

[0070]

[0071] Based on the same inventive concept as the above method embodiment, the embodiment of the present application also provides an intelligent analysis system for tunnel surrounding rock mechanical parameters based on drilling parameters, which can realize the functions provided by the above method embodiment, such as Figure 4 As shown, the system includes the following modules:

[0072] Sample database construction module 110: constructs a sample database, which includes the drilling parameters of the tunnel face and the corresponding tunnel face surrounding rock sub-grades;

[0073] Feature system construction module 120: calculates drillability, energy index, statistical characteristics, and constructs a feature system of while-drilling parameters 78;

[0074] Tunnel surrounding rock sub-class intelligent classification model construction module 130: uses a transfer learning algorithm to construct a tunnel surrounding rock sub-class intelligent classification model based on drilling parameters;

[0075] Tunnel surrounding rock mechanical parameter analysis module 140: Based on the surrounding rock sub-class classification probability vector and the surrounding rock mechanical parameter characteristic value of each sub-class, the surrounding rock mechanical parameters are obtained through analysis.

[0076] The present invention is based on the face drilling parameters collected by the intelligent drilling rig during the tunnel construction process, and realizes rapid and real-time intelligent analysis of the surrounding rock mechanical parameters through artificial intelligence technologies such as transfer learning. This method can complete the dynamic analysis of the surrounding rock parameters during the drilling process, which greatly improves the efficiency and reduces the cost compared with traditional testing methods, provides a scientific and reliable reference basis for the formulation of subsequent construction plans, effectively ensures the safety of intelligent tunnel construction, and facilitates intelligent tunnel construction.

[0077] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0078] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "an", "the" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise.

[0079] It should be understood that the term "and / or" as used herein is merely a description of the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0080] The word "if," as used herein, may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0081] The references to "first" and "second" in the embodiments merely distinguish similar objects and do not represent a specific ordering of the objects. It is understood that the specific order or precedence of "first" and "second" can be interchanged where appropriate. It should be understood that the objects distinguished by "first" and "second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein.

[0082] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent analysis method for tunnel surrounding rock mechanical parameters based on drilling parameters, characterized by: The steps include: S1. Construct a sample database for intelligent analysis of tunnel surrounding rock mechanical parameters based on drilling parameters; S2. Construct an intelligent analytical feature system for tunnel surrounding rock mechanical parameters based on drilling parameters; S3. Using transfer learning algorithm, construct an intelligent classification model of tunnel surrounding rock sub-grades based on drilling parameters; S4. Based on the classification probability vector output by the tunnel surrounding rock sub-class intelligent classification model and the characteristic values ​​of the surrounding rock mechanical parameters of each sub-class, the surrounding rock elastic modulus, Poisson's ratio, cohesion, and internal friction angle are analytically obtained.

2. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 1 is characterized by: In step S1, the sample database includes four original drilling parameters automatically collected by the intelligent drilling rig, namely, rotation pressure, feed speed, impact pressure, and thrust pressure; the surrounding rock levels of the face include II, III1, III2, IV1, IV2, V1, and V2.

3. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 1 is characterized by: The step S2 constructs an intelligent analytical feature system for tunnel surrounding rock mechanical parameters based on drilling parameters, specifically including the following: S21. Based on the four original drilling parameters, five rock drillability indices (equivalent feed rate, equivalent propulsion force, difficulty-to-drill index, E index, and equivalent impact force) are calculated; four energy indices (propulsion specific energy, impact specific energy, rotation specific energy, and mechanical specific energy) are calculated. S22. Taking the tunnel face excavation cycle as the unit, six statistical characteristics of four original drilling parameters, five rock drillability indicators, and four energy indicators are collected, including standard deviation, first quartile, second quartile, third quartile, coefficient of variation, and mean; and a 78-characteristic system of drilling parameters is constructed.

4. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 3 is characterized by: In step S21, the calculation formulas for the five rock drillability indices and the four energy indices are: δ v =V p / (V r D) d hard =d ff d fh / d v Where: δ ff is the equivalent propulsion force; D is the drill bit diameter; p r is the rotation pressure; q r is the motor displacement; i r is the reduction ratio; P f is the propulsion pressure; D f is the rear end diameter of the thrust piston; δ fh is the equivalent striking force; D ca 、D cb Respectively, the diameter of the rear and front ends of the striking cylinder piston; m h is the piston mass; t h is the time of impact on the fiber tail; P h is the striking pressure; v is the equivalent feed rate; V r is the drilling tool speed; V d is the feed rate; δ hard is the difficulty to drill indicator; E is the E indicator; Ψ h is the specific energy of strike; S h is the impact stroke; r is the rotational specific energy; t is the propulsion specific energy; m is the mechanical specific energy.

5. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 3 is characterized by: In step S22, the data points of the five rock drillability indicators and the four energy indicators are arranged in ascending order, and then the data are evenly divided into four equal-length intervals, thereby determining the values ​​at the three division points; among them, the first quartile Q1 is at the 25th percentile position after the data is sorted, the second quartile Q2, which is the median, is at the 50th percentile position; the third quartile Q3 corresponds to the 75th percentile position; the mean reflects the trend of the data set, the standard deviation measures the degree of dispersion, and the coefficient of variation is the ratio of the standard deviation to the mean, which is used to compare the discreteness of data of different dimensions.

6. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 1 is characterized by: In step S3, a transfer learning algorithm is used to construct an intelligent classification model of tunnel surrounding rock sub-classes based on drilling parameters, which specifically includes the following steps: S31. Training a source model for intelligent sub-classification of tunnel surrounding rock based on source-domain while-drilling parameters: The sample database is normalized and proportionally divided into two datasets: the source domain and the target domain. The 78 features of the while-drilling parameters in the source domain dataset are horizontally concatenated into a 1×78 input feature vector as input, and a convolutional neural network (CNN) is used for feature extraction and classification. The network structure includes convolutional layers, pooling layers, and fully connected layers. The fully connected layers use the Softmax activation function, and employ the cross-entropy loss function and Adam adaptive learning rate optimization to calculate the loss and optimize the model's weights and bias parameters to complete source model training. S32. Train a transfer learning model for intelligent sub-class grading of tunnel surrounding rock based on target domain LWD parameters: Load the source model trained on the source domain dataset, then freeze the convolutional layer parameters and the fully connected layer parameters except for the last two layers, so that the weights and bias parameters of these layers remain unchanged during the transfer learning process and are not updated. Concatenate the 78 LWD parameter features of the target domain dataset horizontally into a 1×78 input feature vector as the target data input, then fine-tune based on the target data. Use the Adam optimizer to retrain the fully connected layer of the model to adjust the fully connected layer parameters. Through the above steps, the source model is adapted to the target domain data, and a tunnel surrounding rock sub-level intelligent classification transfer learning model is obtained, completing the construction of a tunnel surrounding rock sub-level intelligent classification model based on drilling parameters.

7. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 1 is characterized by: In step S4, based on the classification probability vector output by the tunnel surrounding rock sub-class intelligent classification model and the characteristic values ​​of the surrounding rock mechanical parameters of each sub-class, the surrounding rock elastic modulus, Poisson's ratio, cohesion, and internal friction angle are obtained by analysis, specifically including the following: S41. Calculate the characteristic values ​​of the mechanical parameters of the surrounding rock, such as elastic modulus, Poisson's ratio, cohesion, and internal friction angle, for each sub-level; S42. Analyze the surrounding rock mechanical parameters based on the classification probability vector output by the tunnel surrounding rock sub-class intelligent classification model and the characteristic values ​​of the mechanical parameters of each sub-class surrounding rock.

8. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 7 is characterized by: In step S41: the upper and lower limits of the surrounding rock mechanical parameters of each sub-level are averaged as its characteristic value; The characteristic values ​​of elastic modulus corresponding to grades II to V2 of the surrounding rock mass of the tunnel face are 25, 15.35, 8.35, 4.9, 2.55, 1.65, and 1.15 GPa, respectively; the characteristic values ​​of Poisson's ratio corresponding to grades II to V2 are 0.225, 0.255, 0.28, 0.305, 0.33, 0.37, and 0.42, respectively; the characteristic values ​​of internal friction angle corresponding to grades II to V2 are 55, 47, 41.5, 37, 31, 24.5, and 21°, respectively; and the characteristic values ​​of cohesion corresponding to grades II to V2 are 1.8, 1.3, 0.9, 0.6, 0.35, 0.16, and 0.085 MPa, respectively.

9. The intelligent analysis method of tunnel surrounding rock mechanical parameters based on drilling parameters according to claim 7 is characterized by: In step S42, the elastic modulus, Poisson's ratio, cohesion, and internal friction angle of the surrounding rock are calculated based on the classification probability vector of the surrounding rock subclass and the characteristic values ​​of each surrounding rock mechanical parameter. The calculation formula is: Where: y E 、y ν 、y C 、 are the analytical results of the tunnel surrounding rock elastic modulus, Poisson's ratio, cohesion, and internal friction angle; E i 、ν i 、C i 、 are the eigenvalues ​​of subtype i respectively; P i is the probability of sub-class type i output by the tunnel surrounding rock sub-class intelligent classification model.

10. A system for intelligent analysis of tunnel surrounding rock mechanical parameters based on drilling parameters according to any one of claims 1 to 9, characterized in that: Includes the following modules: Sample database construction module (110): constructing a sample database, the sample database including the drilling parameters of the tunnel face and the corresponding tunnel face surrounding rock sub-grades; Characteristic system construction module (120): calculates rock drillability index, energy index, statistical characteristics, and constructs a characteristic system of drilling parameters 78; Tunnel surrounding rock sub-class intelligent classification model construction module (130): using transfer learning algorithm to construct a tunnel surrounding rock sub-class intelligent classification model based on drilling parameters; Tunnel surrounding rock mechanical parameter analysis module (140): based on the surrounding rock sub-class classification probability vector and the surrounding rock mechanical parameter characteristic value of each sub-class, the surrounding rock mechanical parameters are obtained through analysis.

Images