In-situ detection method for rock mass mechanical parameters in initial stress state based on drilling technology

Abstract

The invention provides a method for acquiring rock mass mechanical parameters in an initial crustal stress state based on a digital drilling technology, and relates to the technical field of geotechnical engineering reconnaissance, the method comprises the following steps: 1, acquiring four drilling parameters of propulsive force F, torque M, drilling speed V and rotating speed N through monitoring equipment; 2, based on the drilling parameters and the indoor rock strength parameters, the initial ground stress of the surrounding rock of the tunnel face is obtained; thirdly, the rock mass compressive strength of each construction cycle tunnel face is analyzed based on the drilling parameters and the initial ground stress, and the rock mass elasticity modulus is determined based on energy conservation; and 4, constructing a tunnel face rock mass mechanical parameter and initial crustal stress identification model based on deep learning, training the model by using a sample library, and further accurately obtaining each cycle crustal stress and rock mass mechanical parameter. The method can effectively help constructors to judge the overall stability of the tunnel, and in-situ quantitative detection of the strength parameters and the crustal stress state of surrounding rock of the tunnel face of tunnel engineering can be achieved.

Description

Technical Field

[0001] The invention relates to the technical field of geotechnical engineering investigation, and in particular to an in-situ detection method for rock mass mechanical parameters under an initial stress state based on drilling technology. Background Art

[0002] Tunnel engineering often involves rock drilling technology construction. Rapidly obtaining the engineering rock mechanics parameters corresponding to the initial ground stress conditions during the rock drilling process is an important prerequisite for guiding the safe construction of drilling projects and rock excavation projects of different scales.

[0003] Rock compressive strength and elastic modulus are one of the basic parameters of rock mechanics, and rock mass compressive strength and elastic modulus can better reflect the mechanical properties of the surrounding rock itself. This parameter has an important impact on tunnel support design, tunnel engineering stability, excavation speed and monitoring measurement. However, the traditional rock strength test method is limited by the size and number of cores, and it is difficult to cover the entire project area. In addition, the availability of drilling and coring is low, time-consuming, and cumbersome. At the same time, traditional rock mechanics indoor tests consume a lot of manpower and material resources, and the measured rock mechanical properties are essentially different from the engineering rock mass, making it difficult to consider the comprehensive impact of the structural surface. Therefore, if the basic mechanical properties of the surrounding rock mass of the tunnel face can be obtained through drilling rig parameters, and the integrated exploration, design and construction management system can be used, it is of great significance to ensure construction safety and speed up construction progress. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention provides an in-situ detection method of rock mechanical parameters under the initial stress state based on drilling technology, which can realize intelligent analysis of rock mechanical properties under the initial stress of the face, and solve the problems mentioned in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an in-situ detection method of rock mass mechanical parameters under initial stress state based on drilling technology, comprising the following steps:

[0006] S1. Obtaining drilling parameters during the drilling process of the drilling rig, including: obtaining real-time torque through a dynamic torque sensor, obtaining real-time speed through a speed sensor, obtaining real-time propulsion force through a pressure sensor, and obtaining real-time drilling speed through a displacement sensor;

[0007] S2. When the drill rig is drilling in the rock mass, the rock mass is subjected to the thrust force exerted by the drill rig, and the drill bit is subjected to the resistance of the rock mass when advancing. According to the force balance of the drill bit, the relationship between the rock mass compressive strength, ground stress and the drilling parameters of the drill bit is obtained;

[0008] S3. Analyze the rock mass compressive strength of the tunnel face in each construction cycle based on drilling parameters and initial ground stress, and determine the rock mass elastic modulus based on energy conservation analysis;

[0009] S4. Based on deep learning, a model for identifying the mechanical parameters of the rock mass at the tunnel face and the initial in-situ stress is constructed, and the model is trained using a sample library to further accurately obtain the in-situ stress and mechanical parameters of the rock mass in each cycle. Preferably, step S2 is based on the force balance of the drill bit to establish the relationship between the rock mass compressive strength, in-situ stress and drilling parameters, which specifically includes the following:

[0010] According to the force balance of the drill bit, the propulsion force output by the drill rig during the drill bit advancement process is F, and the torque is M, and there is the following formula:

[0011]

[0012] F1 is the cutting resistance of the rock mass to the drill bit, and the calculation formula is as follows:

[0013]

[0014] F2 is the positive pressure of the cutting edge on the rock mass, and the calculation formula is as follows:

[0015]

[0016] Then we have:

[0017]

[0018] The formula includes parameters related to rock mass properties: rock mass cohesion c, internal friction angle φ, friction angle between the blade and the rock mass and the angle χ between the rock shear plane and the cutting plane; cutting blade width b, cutting depth h, tool rake angle γ; maximum stress σ m , σ m = kσ s , σ s is the compressive strength of the rock mass, k depends on the geometry of the tool and the friction of the interface; the friction length l on the cutting edge f ; Friction coefficient μ between the hole wall rock and the drill bit surface; lateral pressures p1 and p2 on the cutting edge; drill bit radius R, contact area S between the drill bit side and the rock mass;

[0019] In the above formula, p1 and p2 correspond to the horizontal and vertical ground stresses at the tunnel face, respectively, and p1 = k0p2, k0 is the lateral pressure coefficient. The relationship between the compressive strength of the surrounding rock mass at the tunnel face and the ground stress can be established by formula (4), which can be expressed as σ s =f(F, M, ξ, η, σ0), where ξ represents parameters related to rock properties, η represents parameters related to drill bit shape, and σ0 represents parameters related to ground stress.

[0020] Preferably, the step S3 analyzes the rock mass compressive strength of the tunnel face of each construction cycle based on the drilling parameters and the initial ground stress, and determines the rock mass elastic modulus based on the energy conservation analysis, specifically including:

[0021] After obtaining the rock mass compressive strength and the ground stress of the tunnel face, the rock mass elastic modulus can be further obtained through the output data of the fully computerized three-arm drilling rig; the method includes: different from the aforementioned drilling parameters, the drilling rig can obtain the impact-propulsion-rotation specific energy output by the rig drill bit during the impact-propulsion-rotation rock breaking process; the impact-propulsion specific energy ISE is composed of the impact specific energy, propulsion specific energy and rotation specific energy, and the calculation formula of the impact-propulsion specific energy ISE is as follows:

[0022]

[0023] The formula includes 13 variables including machine parameters, drilling parameters, and drilling parameters, among which the rear end diameter D of the thrust piston is t , drilling tool speed V r , motor displacement q r , reduction ratio i r , impact piston rear end diameter D cA , impact piston front end diameter D cB , Impact stroke S c , impact piston mass m c It is the machine parameter of the drilling rig, which is set when the drilling rig leaves the factory or before drilling; feed speed V d , Propulsion pressure P t , impact pressure P c and rotary pressure P r It is the drilling parameter collected by the sensor in real time during the drilling process of the drill bit; the borehole diameter D is a drilling parameter used to calculate the drilling area;

[0024] The rock mass unit per unit volume is deformed under mechanical action, and the energy output by the machine is used to break the rock. This energy is completely converted into the internal energy of the rock mass, including the dissipated energy of the rock mass and the elastic deformation energy of the rock mass. By introducing the elastic deformation energy conversion coefficient K, the elastic modulus of the rock mass can be obtained by the following formula:

[0025]

[0026] In the formula, E s is the elastic modulus of the rock mass, in Pa; σ s is the compressive strength of rock mass, unit: Pa.

[0027] Preferably, the step S4 specifically includes: analyzing the rock mechanical parameters and initial geostress parameters of the face according to the force balance analysis and energy analysis of the drill bit during the rock drilling process, relying on the rock mechanical parameters obtained from the rock mechanics indoor test of some tunnel sections, constructing a rock mechanical parameters and initial geostress identification model of the face based on deep learning, using the sample library to train the model, and further accurately obtaining the geostress and rock mechanical parameters of the face of each cycle tunnel; the rock mechanical parameters include rock compressive strength and rock elastic modulus; the rock mechanical parameters include rock compressive strength and rock elastic modulus.

[0028] The network layer number of the identification model of the rock mass mechanical parameters and initial geostress of the tunnel face is 7 layers, namely 2 input layers, 3 hidden layers and 2 output layers; a series of drilling parameters, including torque, rotation speed, thrust, drilling speed and rock mechanical parameters of some sections are taken as two groups of input sets, and geostress and rock mechanical parameters are taken as two groups of output sets. At the same time, geostress and rock mechanical parameters are both output parameters and input parameters in the calculation process; some indoor test data obtained by rock drilling coring are also used as the sample library of this model to train the identification model, and fitting regression analysis and robot learning regression method are used to predict the relevant data of rock compressive strength and elastic modulus under the original stress state of the tunnel face.

[0029] Preferably, under the condition of known drilling parameters, the ground stress parameters are not independent of the compressive strength of the surrounding rock mass, that is, the compressive strength of the rock mass changes with the ground stress. Through deep learning, the rock mass mechanical parameters under the real-time ground stress conditions of each tunnel face can be obtained; at the same time, the strength of the rock mass at failure and the relationship between the uniaxial compressive strength of the rock are: σ3 is the minimum principal stress on the tunnel face, σ ci is the uniaxial compressive strength of rock, m and s are constants related to lithology and structural surface.

[0030] The beneficial effects of the present invention are as follows: the present invention analyzes the face rock mechanical parameters and initial geostress parameters according to the force balance analysis and energy analysis of the drill bit during rock drilling, constructs a face rock mechanical parameters and initial geostress identification model based on deep learning, accurately obtains the geostress and rock mechanical parameters of each cycle face, reduces the input of manpower and material resources in rock geological surveys, rock mechanics indoor tests and initial geostress tests at the construction site, avoids design and construction misjudgments caused by differences in the physical and mechanical properties of rocks and rock masses, facilitates integrated tunnel exploration, design and construction management, and promotes the intelligent construction process of tunnels. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The figure is a schematic flow chart of a method for in-situ detection of rock mechanical parameters under an initial stress state based on drilling technology according to an exemplary embodiment.

[0032] Figure 2 The figure is a schematic diagram of a model for identifying mechanical parameters of surrounding rock mass under two-way initial ground stress conditions at a tunnel face based on deep learning according to an exemplary embodiment. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In order to more clearly illustrate the technical solutions of the implementation methods of the present application, the drawings required for use in the implementation methods will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can also be obtained based on these drawings without creative work.

[0034] With the increasing construction scale of extra-long tunnels, high-altitude tunnels and special geological tunnels, the accurate acquisition of the initial geostress of the surrounding rock of the tunnel face and the physical and mechanical properties of the rock mass directly affects the stability and safety of the tunnel project. The traditional method of on-site sampling and indoor testing is difficult to simulate the mechanical properties of the rock mass under the real on-site initial geostress conditions, has certain errors, and is difficult to meet the current intelligent design and construction needs of tunnels. To this end, the inventors have proposed an in-situ detection method for rock mass mechanical parameters under the initial stress state based on drilling technology through long-term research and practice. The rock mass mechanical parameters and initial geostress parameters of the face are analyzed according to the force balance analysis and energy analysis of the drill bit during the rock drilling process. The rock mass mechanical parameters and initial geostress identification model of the face is constructed based on deep learning, and the geostress and rock mass mechanical parameters of each cycle face are accurately obtained. The invention aims to provide a fast and convenient method for analyzing rock mass mechanical parameters under the initial geostress conditions of the surrounding rock of the on-site face, to ensure the safety of tunnel construction, improve the level of intelligent tunnel construction, speed up the construction progress, and reduce the cost of tunnel construction.

[0035] like Figure 1 As shown, this embodiment provides a technical solution: an in-situ detection method for rock mechanical parameters under initial stress state based on drilling technology, which includes the following steps:

[0036] Step S1: Obtain drilling parameters during the drilling process.

[0037] The real-time torque M is obtained through the dynamic torque sensor, the real-time speed N is obtained through the speed sensor, the real-time propulsion force F is obtained through the pressure sensor, and the real-time drilling speed V is obtained through the displacement sensor.

[0038] Step S2: According to the force balance of the drill bit, the relationship between the rock mass compressive strength, the ground stress and the drilling parameters of the drill bit is obtained.

[0039] 1) When the drill rig is drilling in the rock mass, the rock mass is subjected to the thrust exerted by the drill rig, and the drill bit is subjected to the resistance of the rock mass when advancing. According to the force balance of the drill bit, the thrust output by the drill rig during the advancement of the drill bit is F, and the torque is M, and the following formula is used:

[0040]

[0041] F1 is the cutting resistance of the rock mass to the drill bit, and the calculation formula is as follows:

[0042]

[0043] F2 is the positive pressure of the cutting edge on the rock mass, and the calculation formula is as follows:

[0044]

[0045] Then there is

[0046]

[0047] The formula includes parameters related to rock mass properties: rock mass cohesion c, internal friction angle φ, friction angle between the blade and the rock mass and the angle χ between the rock shear surface and the cutting surface; the cutting blade width b, the cutting depth h, the tool rake angle γ; the maximum stress σm, σm = kσs, σs is the rock compressive strength, k depends on the geometry of the tool and the friction of the interface; the friction length lf on the cutting edge; the friction coefficient μ between the hole wall rock and the drill bit surface; the lateral pressures p1 and p2 on the cutting edge; the drill bit radius R, the contact area S between the drill bit side and the rock mass;

[0048] In the above formula, p1 and p2 correspond to the horizontal and vertical ground stresses at the tunnel face, respectively, and p1=k0p2, k0 is the lateral pressure coefficient. The relationship between the compressive strength of the surrounding rock mass at the tunnel face and the ground stress can be established by formula (4), which can be expressed as σs=f(F, M, ξ, η, σ0), where ξ represents a parameter related to the rock mass properties, η represents a parameter related to the drill bit shape, and σ0 represents a parameter related to the ground stress.

[0049] 2) Examples of drilling parameters obtained by statistics of certain drilling processes are shown in Table 1.

[0050] Table 9

[0051]

[0052] By analyzing the above parameters, according to formula (4), we can get the relationship σs=f(F, M, ξ, η, σ0), and thus calculate the initial stress value of the tunnel face.

[0053] S3: Analyze the rock mass compressive strength of the tunnel face in each construction cycle based on drilling parameters and initial ground stress, and determine the rock mass elastic modulus based on energy conservation;

[0054] 1) Since the drilling data is discontinuous, it is assumed that the ground stress value between the two boreholes remains unchanged. According to the change of drilling parameters, the compressive strength of the rock mass at each tunnel face is calculated using formula (4).

[0055] 2) Based on the impact and rotation specific energy of the drilling rig, calculate the elastic modulus of the surrounding rock mass.

[0056] The rock drilling rig can obtain the impact-propulsion-rotation specific energy output by the rig drill bit during the rock breaking process of impact-propulsion-rotation. The impact-propulsion specific energy ISE is composed of impact specific energy, propulsion specific energy and rotation specific energy. The calculation formula of the impact-propulsion specific energy ISE is as follows:

[0057]

[0058] The formula includes 13 variables including machine parameters, drilling parameters, and drilling parameters, among which the rear end diameter D of the thrust piston is t , drilling tool speed V r , motor displacement q r , reduction ratio i r , impact piston rear end diameter D cA , impact piston front end diameter D cB , Impact stroke S c , impact piston mass m c It is the machine parameter of the drilling rig, which is set when the drilling rig leaves the factory or before drilling; feed speed V d , Propulsion pressure P t , impact pressure P c and rotary pressure P r It is the drilling parameter collected by the sensor in real time during the drilling process of the drill bit; the borehole diameter D is a drilling parameter used to calculate the drilling area;

[0059] The rock mass unit per unit volume is deformed under mechanical action, and the energy output by the machine is used to break the rock. This energy is completely converted into the internal energy of the rock mass, including the dissipated energy of the rock mass and the elastic deformation energy of the rock mass. By introducing the elastic deformation energy conversion coefficient K, the elastic modulus of the rock mass can be obtained by the following formula:

[0060] The rock mass unit per unit volume is deformed under mechanical action, and the energy output by the machine is used to break the rock. This energy is completely converted into the internal energy of the rock mass, including the dissipated energy of the rock mass and the elastic deformation energy of the rock mass. By introducing the elastic deformation energy conversion coefficient K, the elastic modulus of the rock mass can be obtained by the following formula:

[0061]

[0062] In the formula, E s is the elastic modulus of the rock mass, in Pa; σ s is the rock mass compressive strength, in units of Pa. S4: The step S4 specifically comprises: analyzing the rock mass mechanical parameters and initial geostress parameters of the face according to the force balance analysis and energy analysis of the drill bit during rock drilling, relying on the rock mechanical parameters obtained from the rock mechanics indoor test of some tunnel sections, constructing a rock mass mechanical parameters and initial geostress identification model of the face based on deep learning, using the sample library to train the model, and further accurately obtaining the geostress and rock mass mechanical parameters of the face of each cycle tunnel; the rock mass mechanical parameters include rock mass compressive strength and rock mass elastic modulus; the rock mass mechanical parameters include rock compressive strength and rock elastic modulus. The network layer number of the identification model of the rock mechanical parameters and initial geostress of the tunnel face is 7 layers, namely 2 input layers, 3 hidden layers and 2 output layers; a series of drilling parameters, including torque, rotation speed, thrust, drilling speed and rock mechanical parameters of some sections are taken as two sets of input sets, and geostress and rock mechanical parameters are taken as two sets of output sets. At the same time, geostress and rock mechanical parameters are both output parameters and input parameters in the calculation process; some indoor test data obtained by rock drilling coring are also used as the sample library of this model to train the identification model, and the fitting regression analysis and robot learning regression method are used to predict the relevant data of the rock compressive strength and elastic modulus under the original stress state of the tunnel face.

[0063] In step S4, the face rock mass compressive strength identification model can adopt a BP neural network model, use tanh as the activation function of the hidden layer, use sigmoid as the activation function of the output layer, set the maximum number of iterations to 1000, and the training error target to 10 -6 , and the learning rate is 0.01.

[0064] It should be noted that under the condition of known drilling parameters, the ground stress parameters and the compressive strength of the surrounding rock mass are not independent, that is, the compressive strength of the rock mass changes with the ground stress. Through deep learning, the rock mass mechanical parameters under the real-time ground stress conditions of each tunnel face can be obtained; at the same time, the relationship between the strength of the rock mass at failure and the uniaxial compressive strength of the rock is: σ3 is the minimum principal stress on the tunnel face, σ ci is the uniaxial compressive strength of rock, m and s are constants related to lithology and structural surface.

[0065] The present invention analyzes the relationship between the drilling parameters generated during the drilling process and the initial geostress of the surrounding rock and the rock mechanical parameters, which can effectively help construction personnel judge the overall stability of the tunnel, reduce the investment in the engineering survey stage, and realize in-situ quantitative detection of the strength parameters and geostress state of the surrounding rock at the tunnel engineering face.

[0066] 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 protection scope of the present invention.

Claims

1. A method for in-situ detection of rock mass mechanical parameters under initial stress state based on drilling technology, characterized in that: The following steps are involved: S1. Obtaining drilling parameters during the drilling process of the drilling rig, including: obtaining real-time torque through a dynamic torque sensor, obtaining real-time speed through a speed sensor, obtaining real-time propulsion force through a pressure sensor, and obtaining real-time drilling speed through a displacement sensor; S2. When the drill rig is drilling in the rock mass, the rock mass is subjected to the thrust force exerted by the drill rig, and the drill bit is subjected to the resistance of the rock mass when advancing. According to the force balance of the drill bit, the relationship between the rock mass compressive strength, ground stress and the drilling parameters of the drill bit is obtained; S3. Analyze the rock mass compressive strength of the tunnel face in each construction cycle based on drilling parameters and initial ground stress, and determine the rock mass elastic modulus based on energy conservation analysis; S4. Based on deep learning, a model for identifying the mechanical parameters of the rock mass and initial geostress of the tunnel face is constructed, and the model is trained using a sample library to further accurately obtain the geostress and rock mechanical parameters of each cycle.

2. The in-situ detection method of rock mass mechanical parameters under initial stress state based on drilling technology according to claim 1 is characterized in that: The step S2 specifically includes: According to the force balance of the drill bit, the propulsion force output by the drill rig during the drill bit advancement process is F, and the torque is M, and there is the following formula: F1 is the cutting resistance of the rock mass to the drill bit, and the calculation formula is as follows: F2 is the positive pressure of the cutting edge on the rock mass, and the calculation formula is as follows: Then we have: The formula includes parameters related to the properties of the rock mass: rock mass cohesion c, internal friction angle φ, friction angle between the blade and the rock mass and the angle χ between the rock shear plane and the cutting plane; cutting blade width b, cutting depth h, tool rake angle γ; maximum stress σ m , σ m = kσ s , σ s is the compressive strength of the rock mass, k depends on the geometry of the tool and the friction of the interface; the friction length l on the cutting edge f ; Friction coefficient μ between the hole wall rock and the drill bit surface; lateral pressures p1 and p2 on the cutting edge; drill bit radius R, contact area S between the drill bit side and the rock mass; In the above formula, p1 and p2 correspond to the horizontal and vertical ground stresses at the tunnel face, respectively, and p1 = k0p2, k0 is the lateral pressure coefficient. The relationship between the compressive strength of the surrounding rock mass at the tunnel face and the ground stress can be established by formula (4), which can be expressed as σ s =f(F, M, ξ, η, σ0), where ξ represents parameters related to rock properties, η represents parameters related to drill bit shape, and σ0 represents parameters related to ground stress.

3. The in-situ detection method of rock mass mechanical parameters under initial stress state based on drilling technology according to claim 1 is characterized in that: The step S3 specifically includes: After obtaining the rock mass compressive strength and the ground stress of the tunnel face, the rock mass elastic modulus can be further obtained through the output data of the fully computerized three-arm drilling rig; the method includes: different from the aforementioned drilling parameters, the drilling rig can obtain the impact-propulsion-rotation specific energy output by the rig drill bit during the impact-propulsion-rotation rock breaking process; the impact-propulsion specific energy ISE is composed of the impact specific energy, propulsion specific energy and rotation specific energy, and the calculation formula of the impact-propulsion specific energy ISE is as follows: The formula includes 13 variables including machine parameters, drilling parameters, and drilling parameters, among which the rear end diameter D of the thrust piston is t , drilling tool speed V r , motor displacement q r , reduction ratio i r , impact piston rear end diameter D cA , impact piston front end diameter D cB , Impact stroke S c , impact piston mass m c It is the machine parameter of the drilling rig, which is set when the drilling rig leaves the factory or before drilling; feed speed V d , Propulsion pressure P t , impact pressure P c and rotary pressure P r It is the drilling parameter collected by the sensor in real time during the drilling process of the drill bit; the borehole diameter D is a drilling parameter used to calculate the drilling area; The rock mass unit per unit volume is deformed under mechanical action, and the energy output by the machine is used to break the rock. This energy is completely converted into the internal energy of the rock mass, including the dissipated energy of the rock mass and the elastic deformation energy of the rock mass. By introducing the elastic deformation energy conversion coefficient K, the elastic modulus of the rock mass can be obtained by the following formula: In the formula, E s is the elastic modulus of the rock mass, in Pa; σ s It is the compressive strength of rock mass, unit is Pa.

4. The in-situ detection method of rock mass mechanical parameters under initial stress state based on drilling technology according to claim 1 is characterized in that: The step S4 specifically includes: analyzing the rock mechanical parameters and initial geostress parameters of the face according to the force balance analysis and energy analysis of the drill bit during the rock drilling process, relying on the rock mechanical parameters obtained from the rock mechanics indoor test of some tunnel sections, building a rock mechanical parameters and initial geostress identification model of the face based on deep learning, using the sample library to train the model, and further accurately obtaining the geostress and rock mechanical parameters of the face of each cycle tunnel; the rock mechanical parameters include rock compressive strength and rock elastic modulus; the rock mechanical parameters include rock compressive strength and rock elastic modulus.

5. The in-situ detection method of rock mass mechanical parameters under initial stress state based on drilling technology according to claim 4 is characterized in that: In the step S4, the network layer number of the identification model of the rock mass mechanical parameters and initial geostress of the tunnel face is 7 layers, namely, 2 input layers, 3 hidden layers and 2 output layers; a series of drilling parameters, including torque, rotation speed, thrust, drilling speed and rock mechanical parameters of some sections are used as two groups of input sets, and geostress and rock mechanical parameters are used as two groups of output sets. At the same time, geostress and rock mechanical parameters are both output parameters and input parameters in the calculation process; some indoor test data obtained by rock drilling and coring are also used as the sample library of this model to train the identification model, and fitting regression analysis and robot learning regression method are used to predict the relevant data of rock compressive strength and elastic modulus under the original stress state of the tunnel face.

6. The in-situ detection method of rock mass mechanical parameters under initial stress state based on drilling technology according to any one of claims 1 to 5, characterized in that: In step S4, the face rock mass compressive strength identification model can adopt a BP neural network model, use tanh as the activation function of the hidden layer, use sigmoid as the activation function of the output layer, set the maximum number of iterations to 1000, and the training error target to 10 -6 , and the learning rate is 0.

01.

7. The in-situ detection method of rock mass mechanical parameters under initial stress state based on drilling technology according to claim 2 is characterized in that: Under the condition of known drilling parameters, the ground stress parameters and the compressive strength of the surrounding rock mass are not independent, that is, the compressive strength of the rock mass changes with the ground stress. Through deep learning, the rock mass mechanical parameters under the real-time ground stress conditions of each tunnel face can be obtained; at the same time, the relationship between the strength of the rock mass at failure and the uniaxial compressive strength of the rock is: σ3 is the minimum principal stress on the tunnel face, σ ci is the uniaxial compressive strength of rock, m and s are constants related to lithology and structural surface.

Images