Dual-mode shield mode conversion position determination method based on machine learning

Abstract

The invention discloses a dual-mode shield mode switching position determination method based on machine learning, and belongs to the technical field of shield construction. The method comprises the following steps that S1, stress analysis is conducted on a tunnel face at the dual-mode shield mode conversion position; s2, determining a position for switching the TBM mode to the EPB mode; and S3, determining the safe distance between the tunnel face and the interface of the soft rock and the hard rock when the EPB mode is converted into the TBM mode. Compared with the prior art, the method has the beneficial effects that the method is more economical, and the construction safety can be effectively improved; by the adoption of the method, the cutter can be prevented from being damaged, cutter abrasion is reduced, construction efficiency is improved, and shield mode conversion failure can be effectively prevented.

Description

Technical Field

[0001] The present invention belongs to the technical field of shield construction, and in particular relates to a method for determining a dual-mode shield mode conversion position based on machine learning. Background Art

[0002] In the process of shield selection, slurry shield or earth pressure balance shield is generally selected according to hydrogeological conditions and surrounding environmental factors. However, once a single-mode shield is determined, it cannot be changed. If it must be changed, it will cost a very large economic cost and construction time. To improve this situation, dual-mode shield technology came into being.

[0003] TBM is mainly suitable for hard rock. It can give full play to its advantages when constructing relatively complete surrounding rock with certain self-stability. In subway construction, TBM mode is usually used in full-section hard rock sections or strata with good self-stability of surrounding rock. The EPB mode directly controls the pressure in the soil compartment by adjusting the amount of soil discharge or excavation, and balances it with the water and soil pressure of the excavation surface strata. At the same time, the soil in the soil compartment is directly used to support the excavation surface strata, so as to keep the excavation surface soil layer stable. Therefore, the EPB mode is selected for excavation in soft rock, soil and other strata. When the tunnel passes through poor geological conditions such as karst caves and fault zones, the EPB mode is usually used to ensure construction safety.

[0004] There are two types of mode conversions for dual-mode shield machines: one is the conversion from open mode to earth pressure mode, that is, the conversion from TBM mode to EPB mode; the other is the conversion from earth pressure mode to open mode, that is, the conversion from EPB mode to TBM mode. If the hard rock section in front of the shield cutterhead is too long during mode conversion (bidirectional conversion), the cutter will be worn out before or after the mode conversion, or even damaged, and the construction efficiency will decrease, thereby increasing the construction cost; if the hard rock section in front of the shield cutterhead is too short during mode conversion, due to the sudden change of the stratum at the conversion position, the face is prone to instability during mode conversion, resulting in mode conversion failure. Therefore, the selection of the mode conversion position in dual-mode shield construction is extremely important. Summary of the invention

[0005] The purpose of an embodiment of the present invention is to provide a method for determining a dual-mode shield mode conversion position based on machine learning, which selects the mode conversion position in the hard rock section and determines through numerical analysis the safe distance between the heading and the interface between the soft and hard rocks when the dual-mode shield machine mode is converted in both directions, thereby solving at least one technical problem involved in the background technology.

[0006] In order to solve the above-mentioned technical problems, the present invention is achieved as follows:

[0007] The embodiment of the present invention provides a method for determining a dual-mode shield mode conversion position based on machine learning, comprising the following steps:

[0008] Step S1, force analysis of the tunnel face at the dual-mode shield mode switching position;

[0009] Step S2, determining the position where the TBM mode switches to the EPB mode, specifically includes:

[0010] Step S21, selecting relevant excavation parameters and formation parameters;

[0011] Step S22, establishing a numerical model of different working conditions based on plaxs i3d to simulate the excavation process and obtain surface settlement simulation results;

[0012] Step S23, constructing a BP neural network model, debugging model parameters, using the obtained surface settlement simulation results to train the model, and predicting the project to obtain prediction results;

[0013] Step S24, selecting a conversion position based on the prediction result and experience;

[0014] Step S3, determining the safe distance between the tunnel face and the interface between the soft rock and the hard rock when the EPB mode is switched to the TBM mode.

[0015] Optionally, in step S1, the force analysis of the tunnel face at the dual-mode shield mode switching position specifically includes:

[0016] During tunnel construction, if strongly weathered granite is encountered as the geological condition in front of the heading face, and the thickness of the hard rock layer above the tunnel is not sufficient to provide sufficient support, a saturated state is formed under the action of groundwater, and the bearing capacity of the strongly weathered granite will drop rapidly; under the dual effects of water and soil pressure, the stability of the heading face is reduced, and there is a risk of instability, which may cause safety accidents and pose a serious threat to construction safety.

[0017] Optionally, in step S1, the force analysis of the tunnel face at the dual-mode shield mode switching position specifically includes:

[0018] Provided that the net overburden pressure does not exceed six times the undrained shear strength of the soil, the problem of soil flow through the openings in the tunnel lining or in the tunnel front is effectively controlled, avoiding difficulties during construction.

[0019] Optionally, in step S1, the force analysis of the tunnel face at the dual-mode shield mode switching position specifically includes:

[0020] Based on the burial depth of the tunnel and the mechanical properties of the stratum in which it is located, it is evaluated whether the thickness of the hard rock layer at the top of the tunnel meets the conditions required for mode conversion.

[0021] Optionally, in step S1, the force analysis of the tunnel face at the dual-mode shield mode switching position specifically includes:

[0022] When the ratio of total cover pressure to undrained shear strength exceeds six times, soil inflow problems will be encountered unless special precautions are taken.

[0023] Optionally, in step S21, the excavation parameters include cutter head torque, thrust speed, thrust pressure and cutter head rotation speed.

[0024] Optionally, in step S22, the different working conditions include geological information, tunnel burial depth and shield diameter.

[0025] Optionally, in step S22, obtaining a surface settlement simulation result specifically includes:

[0026] Determine model size and mesh division;

[0027] Set material model and parameters;

[0028] Simulation process.

[0029] Optionally, in step S3, determining a safe distance between the tunnel face and the interface between soft rock and hard rock when the EPB mode is switched to the TBM mode specifically includes:

[0030] When the EPB mode is switched to the TBM mode, the length of the shield machine entering the hard rock section should not be less than the sum of the total length of the main machine plus the distance from the water stop ring to the shield tail plus the length of the water stop ring in the hard rock section;

[0031] Before switching to shield mode, it is necessary to strengthen the grouting in the tunnel, and to supplement the grouting at the tail of the shield and apply a water stop ring.

[0032] Optionally, in the hard rock section, water stop rings are installed 5 to 7 rings behind the shield tail, and at least one water stop ring is located in the hard rock section.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. The present invention relates to the analysis and optimization of key working parameters (such as cutter head torque, propulsion speed, propulsion pressure and cutter head rotation speed) during the tunneling process of a shield machine, and provides a machine learning model (BP neural network) that can predict the surface settlement that may be caused during the construction of a shield machine, which helps to prevent and reduce the impact of construction on the surrounding environment, so as to achieve a more stable and efficient construction process.

[0035] 2. The present invention uses numerical simulation results based on different working conditions to train the model, which can more accurately simulate the surface settlement during the actual construction process and provide a scientific basis for construction decision-making. Through the prediction model, it is possible to take into account economic benefits while ensuring construction safety, and reasonably select the switching point of the slurry / earth pressure dual-mode shield to adapt to different geological conditions.

[0036] 3. The present invention takes into account multiple factors such as geological information, tunnel depth, shield diameter, etc., making the prediction model more comprehensive and accurate. The present invention provides decision support for the conversion of earth pressure / mud water dual-mode shield modes, helping the construction team to make more reasonable construction mode selection under different geological conditions. At the same time, by optimizing parameters such as cutter head speed and propulsion pressure, the tunneling efficiency of the shield machine can be improved and the construction period can be shortened. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0038] Figure 1 A flow chart of a method for determining a dual-mode shield mode switching position based on machine learning provided by the present invention;

[0039] Figure 2 A schematic diagram of the lateral pressure of the cutter disc provided by the present invention;

[0040] Figure 3 A numerical simulation model diagram provided by the present invention;

[0041] Figure 4 A schematic diagram of the lateral distribution of surface subsidence provided by the present invention;

[0042] Figure 5 A schematic diagram of the longitudinal settlement of the ground surface provided by the present invention;

[0043] Figure 6 This is a flow chart of the maximum surface settlement prediction modeling provided by the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are 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 creative work are within the scope of protection of the present invention.

[0045] The terms "first", "second", etc. in the specification and claims of the present invention are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0046] See also Figure 1 As shown, an embodiment of the present invention provides a method for determining a dual-mode shield mode conversion position based on machine learning, comprising the following steps:

[0047] Step S1, force analysis of the tunnel face at the dual-mode shield mode switching position;

[0048] Step S2, determining the position where the TBM mode switches to the EPB mode, specifically includes:

[0049] Step S21, selecting relevant excavation parameters and formation parameters;

[0050] Step S22, establishing a numerical model of different working conditions based on plaxs i3d to simulate the excavation process and obtain surface settlement simulation results;

[0051] Step S23, constructing a BP neural network model, debugging model parameters, using the obtained surface subsidence simulation results to train the model, and predicting the surface subsidence to obtain prediction results;

[0052] Step S24, selecting a conversion position based on the prediction result and experience;

[0053] Step S3, determining the safe distance between the tunnel face and the interface between the soft rock and the hard rock when the EPB mode is switched to the TBM mode.

[0054] In step S1, the force analysis of the tunnel face at the dual-mode shield mode conversion position specifically includes:

[0055] During tunnel construction, if strongly weathered granite is encountered as the geological condition in front of the heading face, and the thickness of the hard rock layer above the tunnel is not sufficient to provide sufficient support, a saturated state is formed under the action of groundwater, and the bearing capacity of the strongly weathered granite will drop rapidly; under the dual effects of water and soil pressure, the stability of the heading face is reduced, and there is a risk of instability, which may cause safety accidents and pose a serious threat to construction safety.

[0056] Provided that the net overburden pressure does not exceed six times the undrained shear strength of the soil, the problem of soil flow through the openings in the tunnel lining or in the tunnel front is effectively controlled, avoiding difficulties during construction.

[0057] Based on the burial depth of the tunnel and the mechanical properties of the stratum in which it is located, it is evaluated whether the thickness of the hard rock layer at the top of the tunnel meets the conditions required for mode conversion.

[0058] When the ratio of total cover pressure to undrained shear strength exceeds six times, soil inflow problems will be encountered unless special precautions are taken.

[0059] In step S2, it should be noted that the surface settlement is caused by the loss of water and soil or stratum disturbance caused by shield machine construction, which destroys the stable stress state of the original stratum. Its influencing factors are mainly related to the nature of the stratum and the overlying stratum through which the tunnel passes, the shield machine excavation parameter settings, and the thickness of the overlying stratum. According to the excavation parameters and geological information during the shield excavation process, it is feasible to use machine learning models to predict surface settlement.

[0060] In step S21, there are many working parameters of the shield machine during the excavation process, but the ones most closely related to the formation changes are the cutter head torque, propulsion speed, propulsion pressure and cutter head rotation speed.

[0061] (1) Cutter torque: When the cutter disc is cutting the soil, both its front and side surfaces will rub against the soil and generate friction torque. Therefore, the magnitude of the cutter disc torque is equal to the sum of the front and side friction torques. The friction torque between the lateral circumference of the cutter disc and the soil consists of two parts.

[0062] like Figure 2 As shown, part of it is generated by vertical earth pressure, and the other part is generated by lateral earth pressure. Obviously, the size of the cutter head torque is related to factors such as the cutter head diameter, the friction coefficient between the cutter head and the soil, and the shield burial depth.

[0063] (2) Advance speed: In order to control the stability of the excavation face and the deformation of the ground during shield operation, the amount of soil fed into and discharged from the soil bin must match. When the speed of the shield cutterhead is constant, the amount of soil fed is proportional to the forward speed, and the amount of soil discharged is proportional to the speed of the screw conveyor. When the speed of the screw conveyor is too fast compared to the forward speed, that is, the amount of soil discharged is too large, the soil pressure in the soil bin will decrease accordingly, resulting in the soil pressure in the shield sealing bin being less than the sum of the soil pressure and water pressure in front of the excavation, which can easily lead to instability of the excavation face and subsidence of the ground surface; when the speed of the screw conveyor is too slow compared to the forward speed, that is, the amount of soil fed is too much, the pressure in the soil bin will increase to a level greater than the sum of the water and soil pressure in front of the shield, resulting in ground uplift. The construction process needs to adjust the advancement speed and the speed of the screw conveyor to balance the two and obtain the desired soil bin pressure.

[0064] The biggest factor affecting the advancement speed is the amount of soil input, that is, the excavation capacity of the shield cutter head.

[0065] (3) Propulsion pressure: Propulsion force is the forward excavation power provided by the hydraulic system to the shield cutterhead during the forward advancement of the shield machine. It is used to overcome various resistances during the excavation of the shield machine, including but not limited to the friction resistance or adhesion between the shield machine and the soil or rock mass, the propulsion resistance of the excavation surface on the cutting disc, and the traction force of the trolley. The resistance of the excavation surface on the front of the cutterhead is the water and soil pressure on the cutterhead surface.

[0066] The thrust pressure is mainly related to soil layer parameters such as lateral earth pressure coefficient, soil weight, and active earth pressure coefficient.

[0067] (4) Cutterhead speed: The cutterhead speed of a shield machine is one of the key factors affecting tunneling efficiency and construction quality. In a hard rock tunnel boring machine (TBM), the cutterhead speed is a function of the size of the disc cutter, the upper speed limit, and the diameter of the TBM. The larger the cutter size, the higher the upper speed limit, which is suitable for larger TBMs.

[0068] In step S22, a multi-layer BP neural network is built using the Python-based keras deep learning library. The parameters for model debugging include the number of neurons in the hidden layer, activation function, learning rate, number of iterations, data batches, and the optimization and loss function of the model are selected.

[0069] The model is trained using the numerical simulation results of different working conditions based on plaxis3d, such as Figure 3 As shown, the working conditions include geological information, tunnel depth and shield diameter, etc. According to the geological survey data, the model size is determined, the outer diameter, thickness and width of the segment are set, the Mohr-Coulomb model is used for the soil, and the linear elastic model is used for the segment. The normal displacement is constrained around the model and at the bottom, and all are impermeable boundaries. The surface settlement simulation results are obtained when the shield is excavated to different distances from the interface between soft and hard rocks.

[0070] The simulation steps are as follows:

[0071] (1) Determine the model size and mesh division: In actual projects, the thickness and properties of rock and soil layers vary greatly. If the model is established and calculated and analyzed completely based on the actual strata and construction conditions, not only will the size of the established model be large and the number of unit mesh divisions be large, but it will also affect the efficiency of numerical calculations. Therefore, the numerical calculation model needs to be simplified. Considering factors such as the buried depth of the tunnel vault, the width of the pipe segment, and the influence range of the tunnel excavation on the surrounding rock and soil of 3-5 times the tunnel diameter, the appropriate model size is finally established, and relevant working conditions are established through different models.

[0072] (2) Setting material models and parameters: The Mohr-Coulomb model (MC) is used for the rock and soil model. The soil layer calculation parameters are determined by combining the recommended values ​​of the physical and mechanical parameters of the rock and soil in the geological survey report with relevant experience. The subway shield tunnel and other structures are simulated using 2D plate elements. The deformation mode of the plate is set to linear elasticity, and the values ​​of some concrete parameters are based on the Code for Design of Concrete Structures (GB 50010-2010). The model boundary conditions are lateral constraints on the four sides of the overall model to simulate the situation in which only settlement deformation and no lateral displacement occur at the soil boundary in actual conditions. Fixed constraints are applied to the bottom of the overall model to simulate the situation in which neither lateral displacement nor vertical settlement deformation occurs at the bedrock in actual conditions.

[0073] (3) Simulation process: In actual engineering, the shield machine excavates with the segment width as one advance. When the excavation face soil is removed, the segment is immediately installed and grouting is carried out, and then the next advance is excavated. In the numerical simulation process, this construction cycle is realized by passivating and activating different units and boundary conditions.

[0074] In step S23, the surface settlement is predicted, including: prediction research based on the BP neural network model, tunneling parameters, stratum parameters and numerical simulation results during tunnel traffic shield construction. Specifically including:

[0075] (1) Lateral distribution of surface subsidence

[0076] The main characteristic of the lateral distribution of surface subsidence is that the closer to the central axis, the greater the subsidence value, such as Figure 4 shown.

[0077] (2) Vertical distribution of surface subsidence

[0078] The longitudinal distribution of surface settlement is mainly related to the shield crossing time. The surface deformation at a point above the tunnel can be divided into five stages, namely, the first deformation, deformation before the shield passes, settlement when the shield passes, settlement after the shield tail passes, and late consolidation deformation of the soil. The longitudinal distribution of surface settlement caused by tunnel construction can be expressed by a cumulative probability curve, such as Figure 5 shown.

[0079] Combined with the engineering data, the selected input parameters are tunnel depth, shield diameter, soil layer information, cutterhead torque, cutterhead speed, propulsion speed and propulsion force, and the output parameter is the maximum surface settlement.

[0080] Among them, the maximum surface settlement prediction modeling process is as follows Figure 6 shown.

[0081] In step S24, the safety and economy of the construction process are mainly considered when determining the reasonable switching point selection principle of the slurry / earth pressure dual-mode shield, and the economic benefits are considered under the condition of ensuring safety. The EPB mode is more economical in soft soil strata, while the TBM mode is more economical in hard rock. Therefore, from an economic perspective, the mode switching point is most suitable at the stratum interface, but in order to ensure construction safety, factors such as vault settlement should also be considered to comprehensively determine the switching position.

[0082] The stability of the heading face during the excavation process is the key control factor for the mode conversion of the earth pressure / mud-water dual-mode shield. The conversion point should be selected in a stable stratum at a certain distance from the stratum interface. The curve of the crown settlement changing with the excavation distance is obtained based on the model surface settlement prediction results and the longitudinal distribution curve. The point where the crown settlement change rate begins to increase sharply and the settlement in the hard rock is relatively minimum is taken as the position of the mode conversion.

[0083] In step S3, determining the safe distance between the tunnel face and the interface between soft rock and hard rock when the EPB mode is switched to the TBM mode includes:

[0084] When the EPB mode is changed to the TBM mode, it is necessary not only to keep the surrounding rock of the tunnel face stable, but also to keep the surrounding rock stable in the mode conversion operation space, and to carry out the screw conveyor removal, belt conveyor installation, passive stirring rod cutting, slag chute and slag plate welding and other soil bunker transformation work. Therefore, when the mode is changed, the length of the shield machine entering the hard rock section should not be less than the sum of the total length of the main machine plus the distance from the water stop ring to the shield tail plus the length of the water stop ring in the hard rock section;

[0085] During the mode conversion, the front and middle shields of the shield machine have no segment support. In order to prevent the groundwater behind the segment wall from entering the soil bin during the mode conversion and affecting the mode conversion construction, it is necessary to strengthen the grouting in the tunnel before the shield machine mode conversion, and to supplement the grouting at the tail of the shield and apply a water stop ring.

[0086] The location of the water stop ring is extremely important. If the water stop ring is too close to the shield tail, the grouting slurry has strong fluidity. If it flows to the shield tail, the shield tail is easily blocked after the mode conversion is completed; if the water stop ring is too far from the shield tail, the water stop effect is not good. Therefore, it is generally chosen to apply the water stop ring 5 to 7 rings behind the shield tail.

[0087] When the EPB mode is switched to the TBM mode, after the shield machine enters the hard rock section, if the water stop ring is constructed in the soft rock section, the groundwater will flow from the segment wall to the shield tail along the interface between the soft and hard rocks. If the shield tail has poor sealing and waterproofing effects, the groundwater will flow along the outer wall of the shield to the cutterhead, affecting the mode conversion work. Therefore, the water stop ring should be made in the hard rock section, and the construction should start 5 rings after the shield tail, and at least one water stop ring should be ensured to be located in the hard rock section.

[0088] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0089] In addition, it should be noted that the scope of the methods and systems in the embodiments of the present invention is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.

[0090] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A method for determining the mode conversion position of a dual-mode shield machine based on machine learning, characterized in that: The steps include: Step S1, force analysis of the tunnel face at the dual-mode shield mode switching position; Step S2, determining the position where the TBM mode switches to the EPB mode, specifically includes: Step S21, selecting relevant excavation parameters and formation parameters; Step S22, establishing a numerical model based on different working conditions of plaxsi3d to simulate the excavation process and obtain the surface settlement simulation results; Step S23, constructing a BP neural network model, debugging model parameters, using the obtained surface settlement simulation results to train the model, and predicting the project to obtain prediction results; Step S24, selecting a conversion position based on the prediction result and experience; Step S3, determining the safe distance between the tunnel face and the interface between the soft rock and the hard rock when the EPB mode is switched to the TBM mode.

2. The method according to claim 1, characterized in that: In step S1, the force analysis of the tunnel face at the dual-mode shield mode conversion position specifically includes: During tunnel construction, if strongly weathered granite is encountered as the geological condition in front of the heading face, and the thickness of the hard rock layer above the tunnel is not sufficient to provide sufficient support, a saturated state is formed under the action of groundwater, and the bearing capacity of the strongly weathered granite will drop rapidly; under the dual effects of water and soil pressure, the stability of the heading face is reduced, and there is a risk of instability, which may cause safety accidents and pose a serious threat to construction safety.

3. The method according to claim 2, characterized in that In step S1, the force analysis of the tunnel face at the dual-mode shield mode conversion position specifically includes: Provided that the net overburden pressure does not exceed six times the undrained shear strength of the soil, the problem of soil flow through the openings in the tunnel lining or in the tunnel front is effectively controlled, avoiding difficulties during construction.

4. The method according to claim 3, characterized in that: In step S1, the force analysis of the tunnel face at the dual-mode shield mode conversion position specifically includes: Based on the burial depth of the tunnel and the mechanical properties of the stratum in which it is located, it is evaluated whether the thickness of the hard rock layer at the top of the tunnel meets the conditions required for mode conversion.

5. The method according to claim 4, characterized in that In step S1, the force analysis of the tunnel face at the dual-mode shield mode conversion position specifically includes: When the ratio of total cover pressure to undrained shear strength exceeds six times, soil inflow problems will be encountered unless special precautions are taken.

6. The method according to claim 5, characterized in that In step S21, the excavation parameters include cutter head torque, thrust speed, thrust pressure and cutter head rotation speed.

7. The method according to claim 6, characterized in that In step S22, different working conditions include geological information, tunnel burial depth and shield diameter.

8. The method according to claim 7, characterized in that In step S22, the surface settlement simulation results are obtained, which specifically include: Determine model size and mesh division; Set material model and parameters; Simulation process.

9. The method according to claim 1, characterized in that: In step S3, determining the safe distance between the tunnel face and the interface between soft rock and hard rock when the EPB mode is switched to the TBM mode includes: When the EPB mode is switched to the TBM mode, the length of the shield machine entering the hard rock section should not be less than the sum of the total length of the main machine plus the distance from the water stop ring to the shield tail plus the length of the water stop ring in the hard rock section; Before switching to shield mode, it is necessary to strengthen the grouting in the tunnel, and to supplement the grouting at the tail of the shield and apply a water stop ring.

10. The method according to claim 9, characterized in that In the hard rock section, water stop rings are installed 5 to 7 rings behind the shield tail, and at least one water stop ring is located in the hard rock section.

Images