Tunnel grouting method and device

Abstract

The embodiment of the invention provides a tunnel grouting method and device, and a grouting balance prediction system is constructed by combining a machine learning model and a numerical manifold method. By analyzing critical value characteristics of a pressure-volume curve and combining grouting flow and rock mass displacement data, model training is performed, and accurate prediction of the grouting state is realized. The system adopts a Bingham fluid model and a fluid-solid coupling model to calculate optimal grouting parameters, correction is carried out through a zero-thickness cohesion unit model, and automatic regulation and control of a grouting pump are achieved. According to the method, the defects of the traditional technology in parameter optimization and process control are effectively overcome, a scientific basis is provided for grouting construction of tunnel engineering, and the grouting quality and efficiency are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a tunnel grouting method and device. Background Art

[0002] Existing tunnel grouting technology has obvious shortcomings. Traditional grouting methods mostly rely on manual experience control and lack intelligent prediction and precise control of the grouting process, resulting in unstable grouting effects and material waste. At the same time, existing methods fail to fully consider the complexity of the rock mass fracture network and it is difficult to accurately simulate the slurry migration path.

[0003] In addition, existing technologies also have bottlenecks in determining the grouting equilibrium state and optimizing parameters. Most methods lack in-depth analysis of the characteristics of the pressure-volume curve and cannot accurately grasp the critical grouting state. The adjustment of grouting parameters lacks scientific basis, making it difficult to achieve the optimal match between grouting pressure and flow rate.

[0004] The existing system has technical shortcomings in fluid-solid coupling modeling and automatic control. The lack of comprehensive consideration of Bingham fluid characteristics and zero-thickness cohesive effect makes it difficult to accurately predict slurry diffusion behavior and bonding reinforcement effect. Solving these problems is of great significance to improving the grouting quality and construction efficiency of tunnel engineering. Summary of the invention

[0005] In response to the problems in the prior art, the present application provides a tunnel grouting method and device, which can effectively solve the shortcomings of traditional technology in parameter optimization and process control, provide a scientific basis for grouting construction of tunnel engineering, and significantly improve the grouting quality and efficiency.

[0006] In order to solve at least one of the above problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a tunnel grouting method, comprising:

[0008] The grouting pressure, grouting flow rate, rock displacement and grouting volume data in the tunnel project are collected, a pressure-volume curve is constructed for the grouting pressure and the grouting volume data, a critical value feature is extracted from the pressure-volume curve, a training sample set is constructed using the critical value feature, the grouting flow rate and the rock displacement data, and the training sample set is input into a preset machine learning model for training to obtain a grouting balance prediction model;

[0009] Establishing a communication connection between the grouting balance prediction model and the grouting pump controller, collecting the grouting pressure and grouting flow rate data of the grouting pump in real time, inputting the grouting pressure and the grouting flow rate data into the grouting balance prediction model, calculating the slurry migration path in the rock mass fracture network based on the numerical manifold method, substituting the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculating the optimal grouting pressure and the optimal grouting flow rate under the grouting equilibrium state;

[0010] A bonding reinforcement algorithm is constructed based on a zero-thickness cohesive unit model, and the optimal grouting pressure and the optimal grouting flow are corrected using the bonding reinforcement algorithm. The corrected grouting pressure and grouting flow data are sent to the grouting pump controller, and the grouting pump controller is driven to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

[0011] Furthermore, the collecting of grouting pressure, grouting flow, rock displacement and grouting volume data in the tunnel engineering, constructing a pressure-volume curve for the grouting pressure and the grouting volume data, and extracting critical value features from the pressure-volume curve include:

[0012] Setting a pressure sensor, a flow sensor, a displacement sensor and a volume sensor, performing multi-point time domain sampling on the data collected by the pressure sensor, the flow sensor, the displacement sensor and the volume sensor, performing time sequence alignment and data preprocessing on the sampled data according to timestamp information, and using a wavelet transform algorithm to perform noise reduction on the sampled data to remove abnormal values ​​and redundant values;

[0013] A coordinate system is established by taking the grouting pressure data after noise reduction as the ordinate and the grouting volume data as the abscissa, and a pressure-volume curve is generated in the coordinate system using a cubic spline interpolation algorithm. The inflection point and extreme point of the pressure-volume curve are identified based on the rate of change of the slope of the curve, and the coordinate values ​​of the inflection point and the extreme point are used as critical value features.

[0014] Furthermore, the critical value feature, the grouting flow rate and the rock mass displacement data are used to construct a training sample set, and the training sample set is input into a preset machine learning model for training to obtain a grouting balance prediction model, including:

[0015] The critical value feature, the grouting flow rate and the rock mass displacement data are associated in time series to construct a feature vector matrix, the feature vector matrix is ​​normalized and standardized, the processed feature vector matrix is ​​divided into a training set and a validation set, label data of the grouting equilibrium state are annotated in the training set, and a training sample set with supervision information is generated;

[0016] A deep neural network including an input layer, a hidden layer and an output layer is constructed, a recurrent neural network unit and a long short-term memory unit are set in the hidden layer, the training sample set is input into the deep neural network, the network parameters are optimized and trained using a back propagation algorithm and a gradient descent method, the training results are cross-validated based on the validation set, and a grouting balance prediction model is obtained.

[0017] Furthermore, the grouting balance prediction model is connected in communication with a grouting pump controller, grouting pressure and grouting flow rate data of the grouting pump are collected in real time, and the grouting pressure and grouting flow rate data are input into the grouting balance prediction model, including:

[0018] Configure the communication interface parameters of the grouting pump controller, including the communication protocol type, data transmission rate and data format, deploy the grouting balance prediction model in the edge computing device, connect the edge computing device with the grouting pump controller using industrial Ethernet, establish a bidirectional data transmission channel based on the TCP / IP protocol, and realize real-time communication between the grouting balance prediction model and the grouting pump controller;

[0019] A data acquisition task is set in the grouting pump controller, and the grouting pressure and grouting flow data are collected from the grouting pump according to a preset sampling period. The grouting pressure and the grouting flow data are digitally filtered and signal conditioned, and the processed data are sent to the grouting balance prediction model in real time through the data transmission channel.

[0020] Furthermore, the slurry migration path in the rock mass fracture network is calculated based on the numerical manifold method, the slurry migration path is substituted into the Bingham fluid model and the fluid-solid coupling model, and the optimal grouting pressure and the optimal grouting flow rate under the grouting equilibrium state are calculated, including:

[0021] A rock mass fracture network model is established based on a numerical manifold method, a grouting hole generation algorithm is combined with the rock mass fracture network model, a slurry flow path is constructed in the rock mass fracture network model using a migration network search algorithm, the slurry flow path is discretized into a finite number of calculation units, boundary constraints and initial conditions are imposed on each calculation unit, and a slurry migration path in the fracture network is generated;

[0022] The migration path is input into the Bingham fluid model as the solution domain, and a control equation group of slurry flow is established. The fluid-solid coupling iterative algorithm is used to numerically solve the control equation group, and the pressure distribution and flow velocity distribution of each calculation unit under the grouting equilibrium state are obtained. The optimal grouting pressure and the optimal grouting flow rate are calculated based on the pressure distribution and the flow velocity distribution.

[0023] Furthermore, the bonding reinforcement algorithm is constructed based on the zero-thickness cohesive unit model, and the optimal grouting pressure and the optimal grouting flow rate are corrected using the bonding reinforcement algorithm, including:

[0024] A material constitutive equation is established in a zero-thickness cohesive unit model, rock mass strength parameters and grouting material performance parameters are substituted into the material constitutive equation, a bonding reinforcement algorithm describing the slurry solidification process is constructed, and an explicit integration scheme is used to numerically solve the bonding reinforcement algorithm to obtain the stress-strain relationship and displacement change law during the slurry solidification process;

[0025] A grouting parameter correction matrix is ​​established according to the stress-strain relationship and the displacement variation law, the optimal grouting pressure and the optimal grouting flow rate are substituted into the grouting parameter correction matrix for calculation, and the calculation results are corrected by using an iterative convergence method to obtain a corrected grouting pressure and a corrected grouting flow rate after considering the slurry solidification effect.

[0026] Furthermore, the corrected grouting pressure and grouting flow rate data are sent to the grouting pump controller, and the grouting pump controller is driven to automatically adjust the grouting pressure and grouting flow rate until a grouting equilibrium state is reached, including:

[0027] The corrected grouting pressure and the corrected grouting flow rate data are encapsulated according to a preset communication protocol format using an industrial data bus, and the encapsulated data are sent to the grouting pump controller in real time through the data transmission channel, and a parameter parsing module is established in the grouting pump controller to parse and verify the validity of the received data;

[0028] The parsed grouting pressure and grouting flow data are written into the parameter register of the grouting pump controller, and the inverter output frequency and servo motor speed of the grouting pump are adjusted in real time based on the proportional integral differential control algorithm, so that the actual grouting pressure and grouting flow of the grouting pump gradually approach the correction value until the grouting equilibrium state is reached.

[0029] In a second aspect, the present application provides a tunnel grouting device, comprising:

[0030] A balance model building module is used to collect grouting pressure, grouting flow, rock displacement and grouting volume data in tunnel engineering, build a pressure-volume curve for the grouting pressure and the grouting volume data, extract critical value features from the pressure-volume curve, use the critical value features, the grouting flow and the rock displacement data to build a training sample set, input the training sample set into a preset machine learning model for training, and obtain a grouting balance prediction model;

[0031] A grouting parameter determination module is used to establish a communication connection between the grouting balance prediction model and the grouting pump controller, collect the grouting pressure and grouting flow rate data of the grouting pump in real time, input the grouting pressure and the grouting flow rate data into the grouting balance prediction model, calculate the slurry migration path in the rock mass fracture network based on the numerical manifold method, substitute the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculate the optimal grouting pressure and the optimal grouting flow rate under the grouting equilibrium state;

[0032] The automatic grouting module is used to construct a bonding reinforcement algorithm based on a zero-thickness cohesive unit model, use the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow, send the corrected grouting pressure and grouting flow data to the grouting pump controller, and drive the grouting pump controller to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

[0033] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the tunnel grouting method when executing the program.

[0034] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the tunnel grouting method when executed by a processor.

[0035] In a fifth aspect, the present application provides a computer program product, comprising a computer program / instructions, which implement the steps of the tunnel grouting method when executed by a processor.

[0036] It can be seen from the above technical solution that the present application provides a tunnel grouting method and device, and constructs a grouting balance prediction system by combining a machine learning model and a numerical manifold method. By analyzing the critical value characteristics of the pressure-volume curve and combining the grouting flow rate and rock displacement data for model training, accurate prediction of the grouting state is achieved. The system uses the Bingham fluid model and fluid-solid coupling model to calculate the optimal grouting parameters, and corrects them through the zero-thickness cohesive unit model to achieve automatic control of the grouting pump. This method effectively solves the shortcomings of traditional technologies in parameter optimization and process control, provides a scientific basis for tunnel engineering grouting construction, and significantly improves the grouting quality and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 One of the flow diagrams of the tunnel grouting method in the embodiment of the present application;

[0039] Figure 2 This is the second flow chart of the tunnel grouting method in the embodiment of the present application;

[0040] Figure 3 The third flow chart of the tunnel grouting method in the embodiment of the present application;

[0041] Figure 4 The fourth flow chart of the tunnel grouting method in the embodiment of the present application;

[0042] Figure 5 The fifth flow chart of the tunnel grouting method in the embodiment of the present application;

[0043] Figure 6 This is the sixth flow chart of the tunnel grouting method in the embodiment of the present application;

[0044] Figure 7 FIG7 is a flow chart of the tunnel grouting method in the embodiment of the present application;

[0045] Figure 8 is a structural diagram of a tunnel grouting device in an embodiment of the present application;

[0046] Fig. 9 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0047] Reference numerals:

[0048] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0051] Taking into account the problems existing in the prior art, the present application provides a tunnel grouting method and device, and constructs a grouting balance prediction system by combining a machine learning model and a numerical manifold method. By analyzing the critical value characteristics of the pressure-volume curve and combining the grouting flow rate and rock displacement data for model training, accurate prediction of the grouting state is achieved. The system uses the Bingham fluid model and the fluid-solid coupling model to calculate the optimal grouting parameters, and corrects them through the zero-thickness cohesive unit model to achieve automatic control of the grouting pump. This method effectively solves the shortcomings of traditional technologies in parameter optimization and process control, provides a scientific basis for grouting construction in tunnel engineering, and significantly improves the grouting quality and efficiency.

[0052] In order to effectively solve the deficiencies of traditional technologies in parameter optimization and process control, provide a scientific basis for tunnel engineering grouting construction, and significantly improve the grouting quality and efficiency, the present application provides an embodiment of a tunnel grouting method, see Figure 1 The tunnel grouting method specifically includes the following contents:

[0053] Step S101: collecting grouting pressure, grouting flow, rock displacement and grouting volume data in the tunnel project, constructing a pressure-volume curve for the grouting pressure and the grouting volume data, extracting a critical value feature from the pressure-volume curve, constructing a training sample set using the critical value feature, the grouting flow and the rock displacement data, inputting the training sample set into a preset machine learning model for training, and obtaining a grouting balance prediction model;

[0054] Optionally, this embodiment collects tunnel engineering grouting parameters through a high-precision sensor network. In the grouting pipeline system, a pressure sensor node is arranged every 2 meters, and a piezoelectric pressure sensor is adopted, with a range of 0-20MPa and an accuracy of 0.1%FS. The grouting flow is monitored by a bidirectional electromagnetic flowmeter, one is installed at the inlet and outlet of the grouting pipe, and the sampling frequency is 10Hz. Rock displacement monitoring adopts high-precision fiber grating sensors, and monitoring sections are set at the tunnel vault, side walls and bottom. 8 measuring points are arranged in each section to form a circumferential distribution. The grouting volume is recorded in real time by a special metering device, and the minimum division value is 0.1L.

[0055] This embodiment implements the data acquisition preprocessing module. First, the raw data is sampled in the time domain with a sampling period of 0.1s to ensure that transient changes in the grouting process are captured. The wavelet threshold denoising method is used to process the measurement noise, and the db4 wavelet is selected, and the decomposition level is set to 4 layers. The Hampel filter is used to identify and remove outliers, and the sliding window length is set to 50 data points. The preprocessed data is time-series aligned to establish a unified time base to ensure the synchronization between different parameters.

[0056] This embodiment innovatively constructs a pressure-volume curve generation algorithm. The pre-processed grouting pressure data is used as the ordinate and the grouting volume data is used as the abscissa to establish a two-dimensional coordinate system. The discrete data points are connected using the cubic spline interpolation method, and the interpolation node spacing is set to 0.5L. In each interpolation interval, the relationship between pressure and volume is described by a cubic polynomial to ensure that the first-order derivative and the second-order derivative of the curve at the node are continuous. The tension coefficient is introduced to control the shape of the curve to avoid unreasonable oscillations.

[0057] This embodiment designs a critical value feature identification method. The pressure-volume curve is numerically differentiated, and the first-order derivative and the second-order derivative are calculated using the central difference format. The inflection point position is identified by the change in the derivative sign. When the first-order derivative is zero and the second-order derivative is not zero, it is determined to be an extreme point. A dynamic threshold is set to determine the validity of the inflection point, and the threshold value is adaptively adjusted according to the local characteristics of the curve. The extracted critical value features include: initial filling pressure point P1 (corresponding to volume V1), crack expansion pressure point P2 (corresponding to volume V2), and pressure stabilization point P3 (corresponding to volume V3).

[0058] This embodiment implements a training sample set construction strategy. The extracted critical value features are composed into a feature vector [P1, V1, P2, V2, P3, V3], which is associated with the grouting flow Q and rock displacement D at the corresponding moment. The sliding time window method is used to construct the time series features, and the window length is 10 sampling cycles. The feature data is normalized, and the min-max method is used to map the values ​​to the [0,1] interval. When constructing the label data, the grouting process is divided into three states: unbalanced, near-balanced, and fully balanced according to engineering experience.

[0059] This embodiment innovatively designs the machine learning model training process. A deep neural network is selected as the basic model. The number of nodes in the input layer is the same as the feature dimension. The hidden layer adopts a two-layer structure, each layer contains 64 neurons, and the output layer corresponds to the equilibrium state classification. LSTM units are introduced in the hidden layer to capture the time correlation of the parameter sequence. The training process adopts batch gradient descent method, the batch size is set to 32, the initial value of the learning rate is 0.001, and the Adam optimizer is used. The dropout mechanism (dropout rate = 0.3) is used to prevent overfitting.

[0060] This embodiment achieves accurate prediction of the grouting equilibrium state through innovative data processing and model training schemes. The scheme makes full use of multi-source sensor data and establishes a complete feature extraction and model training process. The generalization ability of the model is improved through reasonable network structure design and training strategy optimization. In actual engineering applications, the prediction model can accurately identify the equilibrium state during the grouting process, provide a reliable basis for grouting parameter optimization, and significantly improve the intelligent level of grouting construction.

[0061] Step S102: establishing a communication connection between the grouting balance prediction model and the grouting pump controller, collecting the grouting pressure and grouting flow data of the grouting pump in real time, inputting the grouting pressure and the grouting flow data into the grouting balance prediction model, calculating the slurry migration path in the rock mass fracture network based on the numerical manifold method, substituting the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculating the optimal grouting pressure and the optimal grouting flow under the grouting equilibrium state;

[0062] Optionally, this embodiment first implements the communication connection scheme between the grouting balance prediction model and the grouting pump controller. An industrial-grade Ethernet switch is used to build a field communication network, and the network topology adopts a star structure to support Gigabit Ethernet transmission. The communication protocol adopts Modbus TCP / IP, and the data frame format includes a function code, a start address, a data length, and a CRC check. At the grouting pump controller end, the communication serial port baud rate is configured to be 115200bps, 8 data bits, 1 stop bit, and no check bit. Data interaction is achieved through the OPC UA server, and real-time data transmission is performed in publish / subscribe mode.

[0063] This embodiment designs a real-time data acquisition module. A high-precision pressure transmitter is installed on the grouting pump, with a range of 0-30MPa, a response time of less than 10ms, and a temperature compensation range of -20℃ to 60℃. The flow acquisition uses a Coriolis mass flowmeter with a measurement accuracy of ±0.1% and a repeatability better than 0.05%. The data sampling period is set to 50ms, and a double buffer mechanism is used to achieve lossless data acquisition. The original data is processed by digital filtering, using a Butterworth low-pass filter with a cutoff frequency of 10Hz.

[0064] This embodiment innovatively implements the numerical manifold method to construct a rock mass fracture network model. First, based on the on-site geological survey data, a three-dimensional fracture statistical model is established, including parameters such as fracture direction, inclination, length and opening. The Monte Carlo method is used to generate a random fracture network, and the fracture density is determined according to the RQD index of the engineering area. The fracture network is discretized by the Delaunay triangulation algorithm to generate an unstructured grid. Displacement degrees of freedom and pressure degrees of freedom are set at the nodes to construct a complete calculation domain.

[0065] This embodiment designs a method for calculating the migration path of slurry. The minimum energy principle is used to construct a path search algorithm, and the diffusion trajectory of the slurry in the fracture network is calculated with the grouting hole as the starting point. Considering the conductivity of the fracture, the fracture is equivalent to a one-dimensional pipe unit, and the flow resistance is inversely proportional to the cube of the fracture opening. The bifurcation probability model is introduced to describe the flow direction selection of the slurry at the node, and the probability value is proportional to the conductivity of the fracture. The main migration path is determined by iterative calculation.

[0066] This embodiment realizes the numerical solution of the Bingham fluid model. As a non-Newtonian fluid, the slurry has a yield stress characteristic and flows only when the shear stress exceeds the yield stress. In tubular flow, the relationship between shear stress and velocity gradient is described by the constitutive equation, which contains two key parameters: plastic viscosity and yield stress. The control equation is discretized using the finite volume method, and the velocity field and pressure field are stored using staggered grids. The coupled problem of momentum equation and continuity equation is solved using the SIMPLE algorithm.

[0067] This embodiment innovatively designs a fluid-solid coupling analysis method. Considering the influence of slurry pressure on crack deformation, the coupling relationship between rock deformation and slurry flow is established. The rock deformation adopts an elastoplastic constitutive model, and the material parameters include elastic modulus, Poisson's ratio and cohesion. The coupled equations are solved by the separation iteration method. In each time step, the flow field is first solved, and then the crack deformation is updated until convergence. The relaxation factor is introduced to control the iteration process to ensure the stability of the numerical calculation.

[0068] This embodiment implements the optimal grouting parameter calculation strategy. Based on the results of fluid-solid coupling analysis, the mapping relationship between grouting pressure, flow rate and filling effect is established. The filling effect is characterized by the slurry diffusion radius and the crack filling rate. The multi-objective optimization method is used to solve the optimal parameters, and the objective function includes the filling effect term and the energy consumption term. The optimal solution is searched by the particle swarm algorithm, the population size is set to 50, and the maximum number of iterations is 100.

[0069] This embodiment realizes intelligent optimization of grouting parameters through innovative communication schemes and numerical calculation methods. This scheme combines modern communication technology with complex flow analysis to establish a complete numerical simulation system for the grouting process. Through reasonable model simplification and efficient solution strategies, real-time optimization of grouting parameters is achieved. In engineering practice, this scheme significantly improves the controllability of the grouting process and the predictability of the grouting effect, providing reliable technical support for intelligent grouting construction.

[0070] Step S103: construct a bonding reinforcement algorithm based on the zero-thickness cohesive unit model, use the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow, send the corrected grouting pressure and grouting flow data to the grouting pump controller, and drive the grouting pump controller to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

[0071] Optionally, this embodiment innovatively implements a method for constructing a zero-thickness cohesive unit model. In the model, rock mass cracks are discretized into multiple zero-thickness units, the length of each unit is set to 50 mm, and the unit nodes use 8-node isoparametric units. For the material properties of the unit, basic parameters such as elastic modulus E, Poisson's ratio μ, and cohesion c are determined through on-site rock sampling tests. Contact nonlinearity is introduced at the unit interface, and the Coulomb friction criterion is used to describe the mechanical behavior of the interface, and the friction coefficient is calibrated through experiments.

[0072] This embodiment designs a sophisticated bonding reinforcement algorithm. First, the constitutive relationship of the slurry-rock interface is established, considering the two key parameters of the forward stiffness kn and the tangential stiffness ks. Through on-site grouting tests, the strength development curves of the slurry at different ages are obtained, and the strength time-varying function is established. On this basis, a set of nonlinear equations describing the evolution of interface damage is constructed, including the damage initiation criterion and the damage development equation. The evolution of the damage parameter D adopts the form of an exponential function, which takes into account the coupling effect of stress level and time factors.

[0073] This embodiment implements an innovative numerical solution strategy. The control equations are solved by explicit integration method, and the time step is set to 0.001s. In each time step, the displacement increment of the unit node is first calculated, and then the stress state and damage variables are updated. The stress return algorithm is introduced to handle plastic flow to ensure that the calculation results meet the yield criterion. The solution of the nonlinear equations is achieved by Newton-Raphson iteration, and the convergence accuracy is set to 10^-6.

[0074] This embodiment designs a grouting parameter correction matrix. Based on the numerical calculation results, a mapping relationship between the stress field and the displacement field is established. The dimension of the correction matrix M is 6×6, which contains normal stress components, shear stress components and displacement components. The correction coefficient is obtained by fitting the stress-strain curve by the least squares method. For the optimal grouting pressure P and flow rate Q, the correction matrix is ​​used to transform and obtain the correction values ​​Pc and Qc that take into account the slurry solidification effect.

[0075] This embodiment implements a parameter iterative optimization mechanism. The dynamic relaxation method is used for iterative calculation, and the initial value of the relaxation factor ω is set to 0.8. In each iteration, the relative error of pressure and flow is calculated. When the error is less than a preset threshold (usually 0.1%), the iteration is considered to have converged. Adaptive step size control is introduced to dynamically adjust the iteration step size according to the convergence speed to improve the calculation efficiency.

[0076] This embodiment innovatively designs a data transmission interface. Industrial Ethernet is used to achieve communication with the grouting pump controller, and the communication protocol adopts Modbus TCP / IP. The data frame format includes: frame header (2 bytes), function code (1 byte), data length (2 bytes), pressure value (4 bytes), flow value (4 bytes) and check code (2 bytes). The transmission rate is set to 100Mbps, and CRC16 check is used to ensure data integrity.

[0077] This embodiment implements the grouting pump control algorithm. The controller adopts an incremental PID control strategy, in which the proportional coefficient Kp = 0.8, the integral coefficient Ki = 0.05, and the differential coefficient Kd = 0.1. The motor speed is adjusted by the frequency converter to achieve closed-loop control of pressure and flow. The feedforward compensation mechanism is introduced to improve the dynamic response characteristics of the system. Soft start and soft stop functions are set to prevent pressure shock.

[0078] This embodiment designs a balance state judgment criterion. By monitoring the fluctuation amplitude of pressure and flow, when the fluctuation amplitude within 100 consecutive sampling periods is less than the set threshold, it is judged that the balance state is reached. The threshold value is determined by statistical analysis, and the pressure fluctuation threshold is 0.05MPa and the flow fluctuation threshold is 0.1L / min. A fuzzy judgment mechanism is introduced to comprehensively consider the stability of multiple parameters.

[0079] This embodiment achieves precise control of grouting parameters through innovative model construction and algorithm design. The solution fully considers the slurry solidification effect and establishes a complete parameter correction and control system. The automation level of the grouting system is improved through reasonable communication interface design and control strategy optimization. In actual engineering applications, the solution can effectively avoid pressure over-limit and flow fluctuation, ensure the stability and controllability of the grouting process, and significantly improve the quality and efficiency of grouting construction. At the same time, through precise parameter control, slurry waste is reduced, construction costs are reduced, and it has significant engineering application value.

[0080] From the above description, it can be seen that the tunnel grouting method provided in the embodiment of the present application can construct a grouting balance prediction system by combining a machine learning model and a numerical manifold method. By analyzing the critical value characteristics of the pressure-volume curve and combining the grouting flow rate and rock displacement data for model training, accurate prediction of the grouting state is achieved. The system uses the Bingham fluid model and the fluid-solid coupling model to calculate the optimal grouting parameters, and corrects them through the zero-thickness cohesive unit model to achieve automatic control of the grouting pump. This method effectively solves the shortcomings of traditional technologies in parameter optimization and process control, provides a scientific basis for grouting construction in tunnel engineering, and significantly improves the grouting quality and efficiency.

[0081] In one embodiment of the tunnel grouting method of the present application, see Figure 2 , and can also include the following:

[0082] Step S201: setting a pressure sensor, a flow sensor, a displacement sensor and a volume sensor, performing multi-point time domain sampling on the data collected by the pressure sensor, the flow sensor, the displacement sensor and the volume sensor, performing time sequence alignment and data preprocessing on the sampled data according to timestamp information, and using a wavelet transform algorithm to perform noise reduction on the sampled data to remove abnormal values ​​and redundant values;

[0083] Step S202: Establish a coordinate system using the grouting pressure data after noise reduction as the vertical coordinate and the grouting volume data as the horizontal coordinate, generate a pressure-volume curve in the coordinate system using a cubic spline interpolation algorithm, identify the inflection point and extreme point of the pressure-volume curve based on the rate of change of the slope of the curve, and use the coordinate values ​​of the inflection point and the extreme point as critical value features.

[0084] Optionally, this embodiment innovatively implements a multi-sensor data acquisition system. High-precision pressure sensors are installed at key nodes of the grouting pipeline, using piezoresistive sensors with a range of 0-30MPa, an accuracy level of 0.1%FS, and a response time of less than 5ms. The pressure sensor uses a 4-20mA current output and has good anti-interference performance. Electromagnetic flow sensors are installed at the inlet and outlet of the grouting pipeline, with a measurement range of 0-100L / min and an accuracy of ±0.5%.

[0085] This embodiment designs an innovative displacement monitoring solution. An array of fiber Bragg grating displacement sensors is arranged in the tunnel surrounding rock, with a sensor spacing of 2m, a measurement range of ±50mm, and a resolution of 0.01mm. Wavelength division multiplexing technology is used to achieve multi-point synchronous measurement, and the center wavelength drift is linearly related to the displacement. The volume sensor adopts a dual-range design, with a large range of 0-10m3 for total volume monitoring and a small range of 0-1m3 for precise measurement, and the switching threshold is 0.8m3.

[0086] This embodiment implements an efficient data acquisition strategy. Distributed data acquisition modules are used, each module is equipped with a 32-bit ARM processor, the sampling frequency is adjustable, and the maximum supported is 1kHz. Data transmission uses RS485 bus, Modbus RTU protocol, and baud rate of 115200bps. A multi-level cache mechanism is set, including hardware FIFO cache and software ring cache, to prevent data loss. The watchdog circuit monitors the operating status of the acquisition module to ensure the reliability of data acquisition.

[0087] This embodiment designs a timing alignment algorithm. Each data packet contains a 64-bit timestamp with a time resolution of 1ms. The sliding window method is used for data alignment, with a window length of 100ms and a sliding step of 10ms. Within the window, the data value at the same time is calculated by linear interpolation. A data validity check mechanism is introduced, and when the data missing rate exceeds 20%, the data repair program is automatically started.

[0088] This embodiment innovatively implements the wavelet transform denoising algorithm. The db4 wavelet basis function is selected to perform a 5-layer decomposition of the original signal. In each layer of decomposition, the soft threshold method is used to process the wavelet coefficients, and the threshold value is determined by Stein unbiased risk estimation. The wavelet reconstruction uses an inverse transform algorithm to ensure the energy conservation of the signal. The denoising effect is verified by signal-to-noise ratio analysis, and the processed signal retains the original characteristic information.

[0089] This embodiment designs an outlier detection method. The box plot criterion is used to identify outliers, and data points that exceed 1.5 times the interquartile range are marked as suspicious values. The rationality of the suspicious values ​​is verified by local linear regression analysis, and the confirmed outliers are replaced by local mean. Redundancy value detection is based on the principle of information entropy. When the information gain of adjacent data points is less than the threshold, it is determined to be a redundant point.

[0090] This embodiment implements an innovative coordinate system construction method. Adaptive coordinate transformation is used to dynamically adjust the coordinate axis scale according to the data distribution characteristics. The coordinate origin is selected as the initial grouting state, the horizontal coordinate unit is cubic meters, and the vertical coordinate unit is megapascals. Grid auxiliary lines are introduced, and the grid spacing is automatically adjusted with the data density to improve the readability of the curve.

[0091] This embodiment designs a cubic spline interpolation algorithm. A cubic polynomial function is constructed between adjacent data points to ensure that the first-order derivative and the second-order derivative of the curve at the node are continuous. The pursuit method is used to solve the tridiagonal equations to obtain the coefficients of the spline function. By adjusting the boundary conditions, the properties of the curve at the endpoints are controlled to avoid curve oscillation.

[0092] This embodiment innovatively implements a feature point identification method. The first-order and second-order derivatives of the pressure-volume curve are calculated, and the five-point difference format is used to improve the calculation accuracy. The inflection point position is identified by the change in the derivative sign, and the extreme point is determined by considering the change in curvature. Local feature analysis is introduced to perform a fine search near the feature point to improve positioning accuracy.

[0093] This embodiment achieves accurate monitoring of the grouting process through innovative sensor layout and data processing methods. The solution establishes a complete data acquisition and processing system and ensures data quality through multi-level data processing. Through a reasonable feature extraction method, the key states of the grouting process are accurately identified. In engineering practice, the solution significantly improves the measurement accuracy and data reliability of grouting parameters, providing an accurate data basis for subsequent grouting control. At the same time, through intelligent data processing, manual intervention is reduced and monitoring efficiency is improved.

[0094] In one embodiment of the tunnel grouting method of the present application, see Figure 3 , and can also include the following:

[0095] Step S301: Correlate the critical value features, the grouting flow rate and the rock mass displacement data in time series, construct a feature vector matrix, perform normalization and standardization on the feature vector matrix, divide the processed feature vector matrix into a training set and a validation set, annotate the label data of the grouting equilibrium state in the training set, and generate a training sample set with supervision information;

[0096] Step S302: construct a deep neural network including an input layer, a hidden layer and an output layer, set a recurrent neural network unit and a long short-term memory unit in the hidden layer, input the training sample set into the deep neural network, use the back propagation algorithm and the gradient descent method to optimize the network parameters for training, cross-validate the training results based on the validation set, and obtain a grouting balance prediction model.

[0097] Optionally, this embodiment first implements a data association processing mechanism. The critical value feature includes three sets of coordinate values: initial filling pressure point, crack expansion pressure point, and pressure stability point. Each set of coordinates includes pressure and corresponding volume. The grouting flow rate data is sampled using a sliding window method, with the window length set to 60s and the sliding step length set to 10s. The rock displacement data includes radial displacement and tangential displacement, and the displacement vector in a unified coordinate system is obtained through three-dimensional coordinate transformation. A timestamp alignment algorithm is used to establish a temporal association between feature data.

[0098] This embodiment innovatively designs a method for constructing a feature vector matrix. The number of rows in the matrix corresponds to the number of sampling moments, and the number of columns is the feature dimension. Each row of data contains: 6 critical value features (3 pressure values ​​and 3 volume values), 4 flow features (average value, standard deviation, maximum value, minimum value) and 3 displacement features (radial displacement, tangential displacement, and combined displacement). The frequency domain features of the flow data are extracted by Fourier transform to capture the periodic characteristics of flow fluctuations. The displacement data is decomposed using wavelet to extract multi-scale features.

[0099] This embodiment implements the normalization of feature data. For features of different dimensions, the min-max normalization method is used to map the values ​​to the [0,1] interval. The normalization parameters are obtained through historical data statistics, including the maximum and minimum values ​​of each feature. For outliers, the 3σ criterion is used for identification and processing. The feature data is converted into a standard normal distribution with a mean of 0 and a variance of 1 through Z-score standardization, thereby improving the stability of model training.

[0100] This embodiment designs a data set partitioning strategy. The data set is divided into a training set and a validation set using a stratified sampling method with a ratio of 8:2. The stratification is based on the grouting state category to ensure that the distribution of samples of different states in the training set and the validation set is consistent. The time series characteristics are considered in the partitioning process to maintain the continuity of the data. For the samples in the training set, the grouting equilibrium state labels are marked according to the stability of the grouting pressure and flow rate and the convergence of the rock displacement.

[0101] This embodiment innovatively constructs a deep neural network model. The number of nodes in the input layer is the same as the feature dimension, and a fully connected structure is adopted. The hidden layer contains 4 layers, and the number of nodes in each layer is 128, 256, 256, and 128 respectively. Bidirectional LSTM units are embedded in the second and third hidden layers, and the number of units is 128 to capture the long-term dependency of temporal features. The output layer uses a softmax activation function to output the probability distribution of the grouting state.

[0102] This embodiment realizes the optimized design of recurrent neural network. The LSTM unit includes three control units: forget gate, input gate and output gate, and the gate parameters are calculated by sigmoid function. The forget gate controls the degree of forgetting of historical information, the input gate controls the degree of updating of current information, and the output gate controls the degree of output of information. The long-term dependency information is stored by the memory unit to solve the gradient disappearance problem of traditional RNN.

[0103] This embodiment designs a network training optimization strategy. The loss function uses cross entropy, and an L2 regularization term is added to prevent overfitting. The optimizer uses Adam, the initial learning rate is 0.001, and the cosine annealing strategy is used to dynamically adjust the learning rate. The batch size is set to 64, and the number of training rounds is 200. An early stopping mechanism is introduced during the training process, and training is stopped when the validation set loss does not decrease for 10 consecutive rounds.

[0104] This embodiment implements a cross-validation mechanism. A 5-fold cross-validation is used to evaluate the model performance, and each fold validation uses a different training set and validation set. The prediction ability of the model is comprehensively evaluated by calculating indicators such as accuracy, precision, recall, and F1 score. The confusion matrix is ​​introduced to analyze the prediction deviation and optimize the model structure and parameters in a targeted manner.

[0105] This embodiment realizes the intelligent prediction of grouting equilibrium state through innovative data processing and deep learning schemes. The scheme makes full use of time series features and multi-source data to establish a complete feature extraction and model training system. Through reasonable network structure design and training strategy optimization, the generalization ability and prediction accuracy of the model are improved. In engineering practice, the model can predict the equilibrium state of the grouting process in real time, provide a reliable basis for grouting parameter control, and significantly improve the intelligence level and construction quality of grouting construction.

[0106] In one embodiment of the tunnel grouting method of the present application, see Figure 4 , and can also include the following:

[0107] Step S401: configuring the communication interface parameters of the grouting pump controller, including the communication protocol type, data transmission rate and data format, deploying the grouting balance prediction model in the edge computing device, connecting the edge computing device with the grouting pump controller using industrial Ethernet, establishing a bidirectional data transmission channel based on the TCP / IP protocol, and realizing real-time communication between the grouting balance prediction model and the grouting pump controller;

[0108] Step S402: Setting a data acquisition task in the grouting pump controller, collecting grouting pressure and grouting flow data from the grouting pump according to a preset sampling period, performing digital filtering and signal conditioning on the grouting pressure and grouting flow data, and sending the processed data to the grouting balance prediction model in real time through the data transmission channel.

[0109] Optionally, this embodiment first implements the communication configuration scheme of the grouting pump controller. The controller adopts an industrial-grade programmable controller, equipped with dual Gigabit Ethernet interfaces, and supports multiple industrial communication protocols such as Modbus TCP, EtherNet / IP and PROFINET. The communication interface uses RS485 bus as a backup channel, with an adjustable baud rate range of 9600-115200bps, 8 data bits, 1 stop bit, and optional parity check. The data format adopts the IEEE754 floating point standard and supports 32-bit and 64-bit precision.

[0110] This embodiment innovatively designs an edge computing deployment solution. An edge computing device with AI acceleration function is selected, equipped with an ARM Cortex-A72 processor and a neural network processing unit, and runs a real-time operating system. Through model quantization and pruning optimization, the grouting balance prediction model is compressed and deployed to the edge device. A distributed computing framework is used to support incremental updates and online learning of the model.

[0111] This embodiment implements the industrial Ethernet networking solution. A ring network topology is adopted, and network redundancy is achieved through a managed industrial switch. VLAN is configured to isolate the control network and data network to improve network security. RSTP protocol is used to achieve network self-healing, and the fault recovery time is less than 20ms. The network supports QoS service quality management to ensure real-time transmission of control data.

[0112] This embodiment designs a TCP / IP communication architecture. The C / S mode is used to establish a communication connection. The server is deployed on the edge computing device, and the client runs in the grouting pump controller. The communication adopts a long connection mode, and the connection status is monitored through the heartbeat packet mechanism. The data transmission adopts a custom protocol, which includes a message header and a message body. The message header contains information such as synchronization word, message type, and data length.

[0113] This embodiment implements bidirectional data flow control. The uplink data flow contains grouting parameters and status information, which are uploaded in real time using the push mode. The downlink data flow contains control instructions and parameter settings, which are interacted using the request-response mode. Flow control is achieved through a sliding window mechanism, and the window size is dynamically adjusted according to the network status.

[0114] This embodiment innovatively designs a data acquisition task manager. A multi-task parallel architecture is adopted, and data acquisition, data processing, and data transmission are respectively treated as independent tasks. Task scheduling adopts a priority preemptive strategy, and the data acquisition task has the highest priority. Data exchange between tasks is achieved through shared memory, and a semaphore mechanism is used to ensure data consistency.

[0115] This embodiment implements a high-precision sampling scheme. The grouting pressure is sampled using a 24-bit ADC with a sampling frequency of 1kHz. The signal conditioning circuit includes an instrument amplifier and a precision reference source. The grouting flow rate adopts a dual-redundancy design, using both an electromagnetic flowmeter and a Coriolis flowmeter with a sampling frequency of 500Hz. The sampling clock is calibrated by GPS to ensure that the time synchronization accuracy is better than 1ms.

[0116] This embodiment designs a digital filtering algorithm. A Kalman filter is used for data processing, and the state equation contains two variables: pressure and flow. The covariance matrix of observation noise and process noise is estimated through historical data. An adaptive mechanism is introduced to dynamically adjust the filter parameters to improve the filtering effect. For sudden signal, a median filter is used for preprocessing.

[0117] This embodiment implements the signal conditioning scheme. The analog signal passes through a fourth-order Butterworth low-pass filter with a cutoff frequency of 50Hz. A three-wire wiring method is used to eliminate common-mode interference, and the signal line uses a shielded twisted pair. Zero point and full scale calibration are used to eliminate system errors, and temperature drift is compensated by a thermistor.

[0118] This embodiment innovatively designs a data cache mechanism. A double buffer structure is used to implement asynchronous processing of data acquisition and transmission. The buffer size is dynamically allocated according to the sampling frequency and network delay, and circular storage is supported. Data packets are sorted by timestamps to solve the problem of data disorder caused by network transmission delay.

[0119] This embodiment realizes real-time monitoring of the grouting system through an innovative communication architecture and data acquisition solution. The solution establishes a reliable data transmission channel and improves the system response speed through edge computing. The high-precision data acquisition and processing methods are adopted to ensure the accuracy of the monitoring data. In engineering practice, the solution significantly improves the real-time performance and reliability of the grouting control system and provides high-quality data support for the intelligent prediction of the grouting equilibrium state. At the same time, the system availability and stability are improved through distributed deployment and redundant design.

[0120] In one embodiment of the tunnel grouting method of the present application, see Figure 5 , and can also include the following:

[0121] Step S501: establishing a rock mass fracture network model based on a numerical manifold method, combining a grouting hole generation algorithm with the rock mass fracture network model, constructing a slurry flow path in the rock mass fracture network model using a migration network search algorithm, discretizing the slurry flow path into a finite number of calculation units, applying boundary constraints and initial conditions to each calculation unit, and generating a migration path of the slurry in the fracture network;

[0122] Step S502: Input the migration path as a solution domain into the Bingham fluid model, establish a control equation group for slurry flow, use a fluid-solid coupling iterative algorithm to numerically solve the control equation group, obtain the pressure distribution and flow velocity distribution of each calculation unit under the grouting equilibrium state, and calculate the optimal grouting pressure and the optimal grouting flow rate based on the pressure distribution and the flow velocity distribution.

[0123] Optionally, this embodiment innovatively implements the numerical manifold method to construct a rock mass fracture network model. First, the basic parameters of the fractures are determined through on-site geological survey data, including fracture inclination distribution, strike distribution, trace length distribution and density distribution. The Monte Carlo method is used to generate a three-dimensional random fracture network, the fracture persistence adopts an exponential distribution model, and the fracture direction adopts a Fisher distribution. For the spatial distribution of fractures, the fractal dimension is introduced to describe its self-similar characteristics, and the fractal dimension is calculated by the Box-counting method.

[0124] This embodiment designs an innovative grouting hole generation algorithm. Based on the connectivity analysis of the crack network, the minimum spanning tree algorithm is used to determine the optimal location of the grouting hole. The connectivity of the crack network is calculated by graph theory, and the connectivity is used as the main control parameter for the grouting hole arrangement. Considering the construction constraints, the spacing between grouting holes is set to be no less than 3 meters and the hole depth is no more than 30 meters. After determining the location of the grouting hole, a local encrypted grid is established to improve the calculation accuracy.

[0125] This embodiment implements a migration network search algorithm. The A* search strategy is used to construct possible flow paths of slurry starting from the grouting hole. The heuristic function consists of two parts: the known path cost and the estimated remaining cost. The path cost takes into account the fracture aperture and connectivity, and the estimated cost is based on the spatial distribution characteristics of the fracture network. The search process is optimized through the dynamic programming method to avoid repeated calculations.

[0126] This embodiment designs a sophisticated calculation unit discretization method. The flow path is divided by an unstructured grid, and the unit size is adaptively adjusted according to the crack opening. A local encryption strategy is adopted at the intersection of the cracks to ensure the calculation accuracy. The boundary conditions include: a given pressure or flow at the inlet, a free outflow condition at the outlet, and a no-slip boundary condition on the crack wall. The initial condition assumes that the crack is filled with air and the pressure is atmospheric pressure.

[0127] This embodiment innovatively realizes the construction of the Bingham fluid model. As a non-Newtonian fluid, the slurry has a yield stress characteristic. The control equations include the continuity equation, the momentum equation and the constitutive equation. The constitutive equation adopts a bilinear model, which behaves as a rigid body below the yield stress and exhibits linear rheological properties after exceeding the yield stress. The model parameters are determined by rheological tests, including yield stress, plastic viscosity and shear thinning index.

[0128] This embodiment designs a fluid-solid coupling iterative algorithm. A separation solution strategy is adopted to solve the flow field first in each time step, and then update the fracture deformation. The flow field solution adopts the SIMPLE algorithm, and the velocity field and the pressure field are coupled through the pressure correction equation. The fracture deformation adopts an elastic-plastic model, considering the nonlinear mechanical properties of the rock mass. The change of the fracture opening is calculated by the stress balance equation, and then the flow space is updated.

[0129] This embodiment realizes the stability control of numerical solution. The semi-implicit format is used to discretize the time term, and the time step is determined by the CFL criterion. The finite volume method is used for spatial discretization, and the convection term is processed by the upwind format to ensure the stability of the calculation. The relaxation factor is introduced in the iteration process to control the update amplitude of the variable. The convergence criterion is set: the residual is less than 10^-6 and the mass conservation error is less than 0.1%.

[0130] This embodiment designs an optimal grouting parameter calculation method. Based on the numerical calculation results, the pressure field and velocity field of each calculation unit are extracted. The grouting diffusion range and filling rate are obtained by integral calculation. The optimal grouting pressure is determined by the energy minimum principle, which ensures sufficient diffusion while avoiding rock damage caused by excessive pressure. The optimal grouting flow rate is determined based on the trade-off between the filling time and the slurry diffusion radius.

[0131] This embodiment achieves the optimal design of grouting parameters through an innovative numerical simulation method. The scheme establishes a complete fracture network-slurry flow coupling analysis system, fully considering the nonlinear characteristics of rock mass and slurry. Through reasonable numerical algorithm design, the accuracy and reliability of the calculation results are guaranteed. In engineering applications, the scheme can accurately predict the slurry diffusion law, provide a scientific basis for determining the optimal grouting parameters, and significantly improve the rationality of grouting design and the controllability of construction effects. At the same time, through parameter optimization, slurry waste is reduced and grouting efficiency is improved.

[0132] In one embodiment of the tunnel grouting method of the present application, see Figure 6 , and can also include the following:

[0133] Step S601: establishing a material constitutive equation in a zero-thickness cohesive unit model, substituting rock mass strength parameters and grouting material performance parameters into the material constitutive equation, constructing a bonding reinforcement algorithm that describes the slurry solidification process, numerically solving the bonding reinforcement algorithm using an explicit integration scheme, and obtaining the stress-strain relationship and displacement variation law during the slurry solidification process;

[0134] Step S602: Establish a grouting parameter correction matrix according to the stress-strain relationship and the displacement variation law, substitute the optimal grouting pressure and the optimal grouting flow into the grouting parameter correction matrix for calculation, and use the iterative convergence method to correct the calculation results to obtain the corrected grouting pressure and corrected grouting flow after considering the slurry solidification effect.

[0135] Optionally, this embodiment first implements a material constitutive equation construction method for zero-thickness cohesive units. In the model, the rock-slurry interface is simplified to a zero-thickness unit, and the mechanical behavior of the unit is described by a constitutive equation. Rock strength parameters include elastic modulus E, Poisson's ratio μ, cohesion c, and internal friction angle φ, which are obtained through field rock tests. Grouting material performance parameters include initial viscosity η0, setting time t0, and final setting strength σu, which are determined through indoor slurry tests.

[0136] This embodiment innovatively designs a bonding reinforcement algorithm. The algorithm considers the entire process of the slurry from the flow state to the solidification state, and divides the solidification process into three stages: the initial flow period, the solidification period, and the hardening period. In the initial flow period, the slurry exhibits Bingham fluid characteristics, and its rheological behavior is described by plastic viscosity and yield stress. In the solidification period, the material strength increases exponentially with time, and the strength growth rate is related to the temperature and water-cement ratio. In the hardening period, the slurry shrinkage and creep effects are considered.

[0137] This embodiment implements an explicit integral solution strategy. The time step is determined by the CFL criterion to ensure computational stability. In each time step, the unit node forces and displacements are first calculated, and then the stress state is updated. For nonlinear constitutive relations, the prediction-correction method is used for solution. The sub-relaxation iterative control calculation process is introduced, and the relaxation factor is determined by numerical experiments. In order to improve computational efficiency, a parallel computing framework is used to divide the computational domain into multiple sub-regions.

[0138] This embodiment designs a stress-strain relationship analysis method. A local coordinate system is established at the interface unit, and the stress is decomposed into normal stress and tangential stress. The stress-strain relationship is described by a nonlinear elastic-plastic model, and the strain softening effect is considered. When the stress exceeds the strength limit, the plastic flow criterion is used to calculate the plastic strain. The damage evolution equation is introduced to describe the interface strength degradation process.

[0139] This embodiment innovatively realizes the analysis of displacement variation law. The node displacement is calculated by the principle of virtual work, taking into account the geometric nonlinear effect. The displacement field is interpolated by isoparametric units to ensure deformation continuity. Contact constraints are introduced to deal with interface separation problems, and the penalty function method is used to impose constraints. The rationality of the calculation results is verified by the energy conservation criterion.

[0140] This embodiment designs a parameter correction matrix construction method. The matrix dimension is 6×6, including normal stress component, shear stress component and displacement component. The matrix elements are determined by the stress-strain relationship and the displacement change law, considering the coupling effect between parameters. The weight coefficient is introduced to adjust the influence of each parameter, and the weight value is determined by sensitivity analysis.

[0141] This embodiment implements a grouting parameter correction strategy. The optimal grouting pressure and flow rate are converted into equivalent stress and strain, and transformed through a correction matrix. The correction process takes into account the volume change and strength growth caused by slurry solidification, as well as the feedback effect of rock mass deformation on grouting parameters. The rationality of the correction result is verified by stress balance conditions and continuity conditions.

[0142] This embodiment innovatively designs an iterative convergence algorithm. The Newton-Raphson method is used for nonlinear iteration, and the convergence criteria include two indicators: force balance residual and displacement increment. The iteration step size is determined by line search technology to improve the convergence performance. Adaptive convergence control is introduced to dynamically adjust the convergence threshold according to the iteration process.

[0143] This embodiment implements a method for verifying the correction results. The rationality of the correction parameters is verified through numerical experiments, and the test conditions cover different rock mass conditions and grouting materials. A parameter sensitivity analysis model is established to evaluate the influence of various factors on the correction results. The reliability of the correction method is verified through field test data.

[0144] This embodiment achieves accurate correction of grouting parameters through innovative algorithm design and numerical analysis methods. The scheme fully considers the slurry solidification effect and establishes a complete mechanical analysis system. Through reasonable model simplification and efficient solution strategy, the calculation efficiency and result reliability are improved. In engineering practice, the scheme can accurately predict the mechanical behavior of slurry during solidification, provide a scientific basis for grouting parameter optimization, and significantly improve the grouting effect and engineering quality. At the same time, through parameter adaptive correction, the intelligence level of the grouting process is improved.

[0145] In one embodiment of the tunnel grouting method of the present application, see Figure 7 , and can also include the following:

[0146] Step S701: using an industrial data bus to encapsulate the corrected grouting pressure and the corrected grouting flow rate data according to a preset communication protocol format, sending the encapsulated data to the grouting pump controller in real time through the data transmission channel, establishing a parameter parsing module in the grouting pump controller, and parsing and validating the received data;

[0147] Step S702: Write the parsed grouting pressure and grouting flow data into the parameter register of the grouting pump controller, and adjust the inverter output frequency and servo motor speed of the grouting pump in real time based on the proportional integral differential control algorithm, so that the actual grouting pressure and grouting flow of the grouting pump gradually approach the correction value until the grouting equilibrium state is reached.

[0148] Optionally, this embodiment first implements a networking solution for an industrial data bus. The CAN bus is used as the backbone network, supporting up to 127 nodes, and the communication rate can reach 1Mbps. In order to improve the anti-interference ability, differential signal transmission is adopted, and the signal line adopts a twisted shielded pair. Photoelectric isolation and common mode suppression are introduced at the physical layer to ensure the reliability of data transmission. The bus adopts a distributed arbitration mechanism to achieve orderly data transmission through message priority.

[0149] This embodiment designs an innovative data encapsulation protocol. The protocol frame consists of a frame header, a control field, a data field, and a check field. The frame header contains a synchronization word and a node address, and the control field defines the data type and processing priority. The data field uses a floating point format to encapsulate pressure and flow data, and supports data compression and fragmented transmission. The check field uses a CRC-16 algorithm, which has error detection and correction capabilities.

[0150] This embodiment implements a real-time data transmission mechanism. A master-slave communication architecture is adopted, and the grouting pump controller serves as a slave to receive control data. The communication cycle is set to 10ms, and a synchronous transmission mode is adopted to ensure the real-time nature of the data. The data transmission of multiple controllers is coordinated through a token passing mechanism to avoid data conflicts. Priority arbitration is initiated when the network is congested to ensure that key data is transmitted first.

[0151] This embodiment designs the internal structure of the parameter parsing module. The module includes a data receiving buffer, a protocol parser, and a data verification unit. The buffer adopts a double-ended queue structure to support asynchronous reading and writing of data. The protocol parser implements frame synchronization and data extraction through a state machine, and supports broken frame reorganization and out-of-order processing. Data verification adopts a multiple verification mechanism, including CRC verification, data range check, and timestamp verification.

[0152] This embodiment innovatively implements a data validity verification method. A data quality assessment model is established, which includes three levels: integrity check, consistency check, and rationality check. The integrity check verifies the structural integrity of the data frame, the consistency check compares the logical relationship between related parameters, and the rationality check determines whether the parameters conform to the actual working conditions based on the physical model.

[0153] This embodiment designs a parameter register management mechanism. The register adopts a partition design, including a real-time data area, a historical data area, and a configuration parameter area. The real-time data area adopts a circular buffer structure to support high-speed data updates. The historical data area saves the operating data of the last hour for trend analysis. The configuration parameter area stores PID control parameters and system configuration information, and adopts a power-off protection design.

[0154] This embodiment implements an innovative PID control algorithm. Aiming at the nonlinear characteristics of the grouting process, a fuzzy adaptive PID control strategy is adopted. The proportional coefficient Kp, integral coefficient Ki and differential coefficient Kd are dynamically adjusted through fuzzy reasoning. A fuzzy rule base of pressure error and error change rate is established to achieve online optimization of control parameters. A feedforward compensation mechanism is introduced to improve the dynamic response capability of the system.

[0155] This embodiment designs a frequency converter control scheme. The vector control mode is adopted to achieve precise adjustment of the motor speed. The frequency converter output frequency range is 0-400Hz, and the frequency resolution is 0.01Hz. The double closed-loop adjustment of pressure and flow is achieved through speed closed-loop control. The inner loop adopts speed PI control, and the outer loop adopts pressure-flow compound control. Soft start and soft stop functions are set to prevent water hammer effect.

[0156] This embodiment implements the servo motor control strategy. The motor adopts a permanent magnet synchronous motor with a rated speed of 1500rpm and is equipped with a high-resolution encoder to achieve position feedback. The system lag is eliminated through torque feedforward compensation, and the dynamic response time is less than 20ms. The speed smoothing filter algorithm is introduced to suppress speed fluctuations and improve the stability of grouting pressure.

[0157] This embodiment innovatively designs a method for determining a balanced state. By establishing a multi-dimensional state vector, the stability of the system is comprehensively evaluated. The state vector includes pressure deviation, flow deviation, pressure change rate, and flow change rate. The sliding window method is used to calculate the state evaluation index. When the index continuously meets the stability condition, it is determined that the equilibrium state has been reached.

[0158] This embodiment achieves precise adjustment of grouting parameters through innovative control strategies and communication solutions. The solution establishes a complete data transmission and processing system, and ensures the reliable execution of control instructions through multiple protections. The advanced control algorithm is adopted to achieve smooth adjustment of grouting pressure and flow. In engineering applications, the solution significantly improves the automation level and control accuracy of the grouting process, ensuring the quality and efficiency of grouting construction. At the same time, through intelligent control, human intervention is reduced and the safety of construction is improved.

[0159] In order to effectively solve the deficiencies of traditional technologies in parameter optimization and process control, provide a scientific basis for tunnel engineering grouting construction, and significantly improve the grouting quality and efficiency, the present application provides an embodiment of a tunnel grouting device for realizing all or part of the contents of the tunnel grouting method, see Figure 8 The tunnel grouting device specifically includes the following contents:

[0160] A balance model building module 10 is used to collect grouting pressure, grouting flow, rock displacement and grouting volume data in tunnel engineering, build a pressure-volume curve for the grouting pressure and the grouting volume data, extract a critical value feature from the pressure-volume curve, build a training sample set using the critical value feature, the grouting flow and the rock displacement data, input the training sample set into a preset machine learning model for training, and obtain a grouting balance prediction model;

[0161] A grouting parameter determination module 20 is used to establish a communication connection between the grouting balance prediction model and the grouting pump controller, collect the grouting pressure and grouting flow rate data of the grouting pump in real time, input the grouting pressure and the grouting flow rate data into the grouting balance prediction model, calculate the slurry migration path in the rock mass fracture network based on the numerical manifold method, substitute the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculate the optimal grouting pressure and the optimal grouting flow rate under the grouting equilibrium state;

[0162] The automatic grouting module 30 is used to construct a bonding reinforcement algorithm based on a zero-thickness cohesive unit model, use the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow rate, send the corrected grouting pressure and grouting flow rate data to the grouting pump controller, and drive the grouting pump controller to automatically adjust the grouting pressure and grouting flow rate until a grouting equilibrium state is reached.

[0163] From the above description, it can be seen that the tunnel grouting device provided in the embodiment of the present application can construct a grouting balance prediction system by combining a machine learning model and a numerical manifold method. By analyzing the critical value characteristics of the pressure-volume curve and combining the grouting flow rate and rock displacement data for model training, accurate prediction of the grouting state is achieved. The system uses the Bingham fluid model and the fluid-solid coupling model to calculate the optimal grouting parameters, and corrects them through the zero-thickness cohesive unit model to achieve automatic control of the grouting pump. This method effectively solves the shortcomings of traditional technologies in parameter optimization and process control, provides a scientific basis for tunnel engineering grouting construction, and significantly improves the grouting quality and efficiency.

[0164] From the hardware level, in order to effectively solve the deficiencies of traditional technologies in parameter optimization and process control, provide a scientific basis for tunnel engineering grouting construction, and significantly improve the grouting quality and efficiency, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the tunnel grouting method, and the electronic device specifically includes the following contents:

[0165] Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the tunnel grouting device and related equipment such as the core business system, user terminal and related database; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the tunnel grouting method and the embodiment of the tunnel grouting device in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0166] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0167] In practical applications, part of the tunnel grouting method can be performed on the electronic device side as described above, or all operations can be completed in the client device. The specific selection can be based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0168] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0169] Fig. 9 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig. 9 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig. 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0170] In one embodiment, the tunnel grouting method function may be integrated into the central processor 9100. The central processor 9100 may be configured to perform the following control:

[0171] Step S101: collecting grouting pressure, grouting flow, rock displacement and grouting volume data in the tunnel project, constructing a pressure-volume curve for the grouting pressure and the grouting volume data, extracting a critical value feature from the pressure-volume curve, constructing a training sample set using the critical value feature, the grouting flow and the rock displacement data, inputting the training sample set into a preset machine learning model for training, and obtaining a grouting balance prediction model;

[0172] Step S102: establishing a communication connection between the grouting balance prediction model and the grouting pump controller, collecting the grouting pressure and grouting flow data of the grouting pump in real time, inputting the grouting pressure and the grouting flow data into the grouting balance prediction model, calculating the slurry migration path in the rock mass fracture network based on the numerical manifold method, substituting the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculating the optimal grouting pressure and the optimal grouting flow under the grouting equilibrium state;

[0173] Step S103: construct a bonding reinforcement algorithm based on the zero-thickness cohesive unit model, use the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow, send the corrected grouting pressure and grouting flow data to the grouting pump controller, and drive the grouting pump controller to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

[0174] From the above description, it can be seen that the electronic device provided in the embodiment of the present application constructs a grouting balance prediction system by combining a machine learning model and a numerical manifold method. By analyzing the critical value characteristics of the pressure-volume curve and combining the grouting flow rate and rock displacement data for model training, accurate prediction of the grouting state is achieved. The system uses the Bingham fluid model and the fluid-solid coupling model to calculate the optimal grouting parameters, and corrects them through the zero-thickness cohesive unit model to achieve automatic control of the grouting pump. This method effectively solves the shortcomings of traditional technologies in parameter optimization and process control, provides a scientific basis for grouting construction in tunnel engineering, and significantly improves the quality and efficiency of grouting.

[0175] In another embodiment, the tunnel grouting device can be configured separately from the central processor 9100. For example, the tunnel grouting device can be configured as a chip connected to the central processor 9100, and the tunnel grouting method function is implemented under the control of the central processor.

[0176] like Fig. 9 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig. 9 In addition, the electronic device 9600 may also include Fig. 9 For components not shown, reference may be made to the prior art.

[0177] like Fig. 9 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device, which receives input and controls the operation of various components of the electronic device 9600.

[0178] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0179] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0180] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0181] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0182] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0183] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module, etc. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0184] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the tunnel grouting method in the above embodiments, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the tunnel grouting method in the above embodiments, where the execution subject is a server or a client, are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0185] Step S101: collecting grouting pressure, grouting flow, rock displacement and grouting volume data in the tunnel project, constructing a pressure-volume curve for the grouting pressure and the grouting volume data, extracting a critical value feature from the pressure-volume curve, constructing a training sample set using the critical value feature, the grouting flow and the rock displacement data, inputting the training sample set into a preset machine learning model for training, and obtaining a grouting balance prediction model;

[0186] Step S102: establishing a communication connection between the grouting balance prediction model and the grouting pump controller, collecting the grouting pressure and grouting flow data of the grouting pump in real time, inputting the grouting pressure and the grouting flow data into the grouting balance prediction model, calculating the slurry migration path in the rock mass fracture network based on the numerical manifold method, substituting the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculating the optimal grouting pressure and the optimal grouting flow under the grouting equilibrium state;

[0187] Step S103: construct a bonding reinforcement algorithm based on the zero-thickness cohesive unit model, use the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow, send the corrected grouting pressure and grouting flow data to the grouting pump controller, and drive the grouting pump controller to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

[0188] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application constructs a grouting balance prediction system by combining a machine learning model and a numerical manifold method. By analyzing the critical value characteristics of the pressure-volume curve and combining the grouting flow rate and rock displacement data for model training, accurate prediction of the grouting state is achieved. The system uses the Bingham fluid model and the fluid-solid coupling model to calculate the optimal grouting parameters, and corrects them through the zero-thickness cohesive unit model to achieve automatic control of the grouting pump. This method effectively solves the shortcomings of traditional technologies in parameter optimization and process control, provides a scientific basis for grouting construction in tunnel engineering, and significantly improves the quality and efficiency of grouting.

[0189] The embodiments of the present application also provide a computer program product capable of implementing all the steps of the tunnel grouting method in the above embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the tunnel grouting method are implemented. For example, the computer program / instruction implements the following steps:

[0190] Step S101: collecting grouting pressure, grouting flow, rock displacement and grouting volume data in the tunnel project, constructing a pressure-volume curve for the grouting pressure and the grouting volume data, extracting a critical value feature from the pressure-volume curve, constructing a training sample set using the critical value feature, the grouting flow and the rock displacement data, inputting the training sample set into a preset machine learning model for training, and obtaining a grouting balance prediction model;

[0191] Step S102: establishing a communication connection between the grouting balance prediction model and the grouting pump controller, collecting the grouting pressure and grouting flow data of the grouting pump in real time, inputting the grouting pressure and the grouting flow data into the grouting balance prediction model, calculating the slurry migration path in the rock mass fracture network based on the numerical manifold method, substituting the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculating the optimal grouting pressure and the optimal grouting flow under the grouting equilibrium state;

[0192] Step S103: construct a bonding reinforcement algorithm based on the zero-thickness cohesive unit model, use the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow, send the corrected grouting pressure and grouting flow data to the grouting pump controller, and drive the grouting pump controller to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

[0193] As can be seen from the above description, the computer program product provided in the embodiment of the present application constructs a grouting balance prediction system by combining a machine learning model and a numerical manifold method. By analyzing the critical value characteristics of the pressure-volume curve and combining the grouting flow rate and rock displacement data for model training, accurate prediction of the grouting state is achieved. The system uses the Bingham fluid model and the fluid-solid coupling model to calculate the optimal grouting parameters, and corrects them through the zero-thickness cohesive unit model to achieve automatic control of the grouting pump. This method effectively solves the shortcomings of traditional technologies in parameter optimization and process control, provides a scientific basis for grouting construction in tunnel engineering, and significantly improves the quality and efficiency of grouting.

[0194] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0195] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0196] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0197] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0198] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A tunnel grouting method, characterized in that: The method comprises: The grouting pressure, grouting flow rate, rock displacement and grouting volume data in the tunnel project are collected, a pressure-volume curve is constructed for the grouting pressure and the grouting volume data, a critical value feature is extracted from the pressure-volume curve, a training sample set is constructed using the critical value feature, the grouting flow rate and the rock displacement data, and the training sample set is input into a preset machine learning model for training to obtain a grouting balance prediction model; Establishing a communication connection between the grouting balance prediction model and the grouting pump controller, collecting the grouting pressure and grouting flow rate data of the grouting pump in real time, inputting the grouting pressure and the grouting flow rate data into the grouting balance prediction model, calculating the slurry migration path in the rock mass fracture network based on the numerical manifold method, substituting the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculating the optimal grouting pressure and the optimal grouting flow rate under the grouting equilibrium state; A bonding reinforcement algorithm is constructed based on a zero-thickness cohesive unit model, and the optimal grouting pressure and the optimal grouting flow are corrected using the bonding reinforcement algorithm. The corrected grouting pressure and grouting flow data are sent to the grouting pump controller, and the grouting pump controller is driven to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

2. The tunnel grouting method according to claim 1, characterized in that: The collecting of grouting pressure, grouting flow, rock mass displacement and grouting volume data in the tunnel engineering, constructing a pressure-volume curve for the grouting pressure and the grouting volume data, and extracting critical value features from the pressure-volume curve include: Setting a pressure sensor, a flow sensor, a displacement sensor and a volume sensor, performing multi-point time domain sampling on the data collected by the pressure sensor, the flow sensor, the displacement sensor and the volume sensor, performing time sequence alignment and data preprocessing on the sampled data according to timestamp information, and using a wavelet transform algorithm to perform noise reduction on the sampled data to remove abnormal values ​​and redundant values; A coordinate system is established by taking the grouting pressure data after noise reduction as the ordinate and the grouting volume data as the abscissa, and a pressure-volume curve is generated in the coordinate system using a cubic spline interpolation algorithm. The inflection point and extreme point of the pressure-volume curve are identified based on the rate of change of the slope of the curve, and the coordinate values ​​of the inflection point and the extreme point are used as critical value features.

3. The tunnel grouting method according to claim 1, characterized in that: The method of constructing a training sample set by using the critical value feature, the grouting flow rate and the rock mass displacement data, and inputting the training sample set into a preset machine learning model for training to obtain a grouting balance prediction model includes: The critical value feature, the grouting flow rate and the rock mass displacement data are associated in time series to construct a feature vector matrix, the feature vector matrix is ​​normalized and standardized, the processed feature vector matrix is ​​divided into a training set and a validation set, label data of the grouting equilibrium state are annotated in the training set, and a training sample set with supervision information is generated; A deep neural network including an input layer, a hidden layer and an output layer is constructed, a recurrent neural network unit and a long short-term memory unit are set in the hidden layer, the training sample set is input into the deep neural network, the network parameters are optimized and trained using a back propagation algorithm and a gradient descent method, the training results are cross-validated based on the validation set, and a grouting balance prediction model is obtained.

4. The tunnel grouting method according to claim 1, characterized in that: The method of establishing a communication connection between the grouting balance prediction model and a grouting pump controller, collecting grouting pressure and grouting flow rate data of the grouting pump in real time, and inputting the grouting pressure and grouting flow rate data into the grouting balance prediction model comprises: Configure the communication interface parameters of the grouting pump controller, including the communication protocol type, data transmission rate and data format, deploy the grouting balance prediction model in the edge computing device, connect the edge computing device with the grouting pump controller using industrial Ethernet, establish a bidirectional data transmission channel based on the TCP / IP protocol, and realize real-time communication between the grouting balance prediction model and the grouting pump controller; A data acquisition task is set in the grouting pump controller, and the grouting pressure and grouting flow data are collected from the grouting pump according to a preset sampling period. The grouting pressure and the grouting flow data are digitally filtered and signal conditioned, and the processed data are sent to the grouting balance prediction model in real time through the data transmission channel.

5. The tunnel grouting method according to claim 1, characterized in that: The method calculates the slurry migration path in the rock mass fracture network based on the numerical manifold method, substitutes the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculates the optimal grouting pressure and the optimal grouting flow rate under the grouting equilibrium state, including: A rock mass fracture network model is established based on a numerical manifold method, a grouting hole generation algorithm is combined with the rock mass fracture network model, a slurry flow path is constructed in the rock mass fracture network model using a migration network search algorithm, the slurry flow path is discretized into a finite number of calculation units, boundary constraints and initial conditions are imposed on each calculation unit, and a slurry migration path in the fracture network is generated; The migration path is input into the Bingham fluid model as the solution domain, and a control equation group of slurry flow is established. The fluid-solid coupling iterative algorithm is used to numerically solve the control equation group, and the pressure distribution and flow velocity distribution of each calculation unit under the grouting equilibrium state are obtained. The optimal grouting pressure and the optimal grouting flow rate are calculated based on the pressure distribution and the flow velocity distribution.

6. The tunnel grouting method according to claim 1, characterized in that: The method of constructing a bonding reinforcement algorithm based on a zero-thickness cohesive unit model and using the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow rate includes: A material constitutive equation is established in a zero-thickness cohesive unit model, rock mass strength parameters and grouting material performance parameters are substituted into the material constitutive equation, a bonding reinforcement algorithm describing the slurry solidification process is constructed, and an explicit integration scheme is used to numerically solve the bonding reinforcement algorithm to obtain the stress-strain relationship and displacement change law during the slurry solidification process; A grouting parameter correction matrix is ​​established according to the stress-strain relationship and the displacement variation law, the optimal grouting pressure and the optimal grouting flow rate are substituted into the grouting parameter correction matrix for calculation, and the calculation results are corrected by using an iterative convergence method to obtain a corrected grouting pressure and a corrected grouting flow rate after considering the slurry solidification effect.

7. The tunnel grouting method according to claim 1, characterized in that: The step of sending the corrected grouting pressure and grouting flow rate data to the grouting pump controller and driving the grouting pump controller to automatically adjust the grouting pressure and grouting flow rate until a grouting equilibrium state is reached includes: The corrected grouting pressure and the corrected grouting flow rate data are encapsulated according to a preset communication protocol format using an industrial data bus, and the encapsulated data are sent to the grouting pump controller in real time through the data transmission channel, and a parameter parsing module is established in the grouting pump controller to parse and verify the validity of the received data; The parsed grouting pressure and grouting flow data are written into the parameter register of the grouting pump controller, and the inverter output frequency and servo motor speed of the grouting pump are adjusted in real time based on the proportional integral differential control algorithm, so that the actual grouting pressure and grouting flow of the grouting pump gradually approach the correction value until the grouting equilibrium state is reached.

8. A tunnel grouting device, characterized in that: The device comprises: A balance model building module is used to collect grouting pressure, grouting flow, rock displacement and grouting volume data in tunnel engineering, build a pressure-volume curve for the grouting pressure and the grouting volume data, extract critical value features from the pressure-volume curve, use the critical value features, the grouting flow and the rock displacement data to build a training sample set, input the training sample set into a preset machine learning model for training, and obtain a grouting balance prediction model; A grouting parameter determination module is used to establish a communication connection between the grouting balance prediction model and the grouting pump controller, collect the grouting pressure and grouting flow rate data of the grouting pump in real time, input the grouting pressure and the grouting flow rate data into the grouting balance prediction model, calculate the slurry migration path in the rock mass fracture network based on the numerical manifold method, substitute the slurry migration path into the Bingham fluid model and the fluid-solid coupling model, and calculate the optimal grouting pressure and the optimal grouting flow rate under the grouting equilibrium state; The automatic grouting module is used to construct a bonding reinforcement algorithm based on a zero-thickness cohesive unit model, use the bonding reinforcement algorithm to correct the optimal grouting pressure and the optimal grouting flow, send the corrected grouting pressure and grouting flow data to the grouting pump controller, and drive the grouting pump controller to automatically adjust the grouting pressure and grouting flow until a grouting equilibrium state is reached.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the tunnel grouting method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the tunnel grouting method according to any one of claims 1 to 7 are implemented.

Images