Shaft tunneling receiving section surrounding rock monitoring and measuring method and related equipment

Abstract

This invention belongs to the field of surrounding rock monitoring technology, specifically disclosing a method and related equipment for monitoring and measuring the surrounding rock in the receiving section of a vertical shaft excavation, including the following steps: Step S1: Deploying monitoring sections and collecting surrounding rock deformation and stress response data in real time; Step S2: Optimizing the number and spatial location of monitoring sections based on a genetic algorithm; Step S3: Dynamically optimizing the yellow, orange, and red three-level early warning thresholds based on a genetic algorithm; Step S4: Optimizing the tunneling machine parameters in real time based on a genetic algorithm; Step S5: Establishing a closed-loop feedback mechanism and issuing the optimized tunneling parameters; Step S6: Triggering graded early warnings and executing corresponding safety control measures. This invention can systematically, in real time, and dynamically monitor and measure the surrounding rock in the receiving section of a vertical shaft excavation and the arch of the lower horizontal tunnel.

Description

Technical Field

[0001] This invention relates to the field of surrounding rock monitoring technology, specifically to a method and related equipment for monitoring and measuring the surrounding rock in the receiving section of a vertical shaft excavation. Background Technology

[0002] During the excavation of a vertical shaft SBE (Shaft Boring Excavator), when the excavation reaches the receiving section (i.e., the area adjacent to the lower horizontal tunnel), the thickness of the reserved rock mass between the bottom of the shaft and the lower horizontal tunnel gradually decreases from the initial thickness to zero. During this process, the stress state of the surrounding rock undergoes drastic changes, mainly presenting the following technical challenges:

[0003] First, as the thickness of the reserved rock mass continues to decrease, the self-stabilizing capacity of the surrounding rock significantly declines. The combined effects of the tunneling disturbance caused by the shaft boring machine, the gradual release of deep ground stress, and the plastic zone of the surrounding rock formed by previous blasting operations can easily lead to rockfalls, collapses, or even overall instability of the receiving section, seriously threatening construction safety and equipment safety.

[0004] Secondly, the arch area of ​​the lower horizontal tunnel will be affected by stress redistribution at the moment the vertical shaft is completed. If there is a lack of effective monitoring and measurement methods, it will be impossible to grasp the deformation and stress response characteristics of the surrounding rock of the arch in a timely manner, and there is a risk of arch collapse.

[0005] Third, existing monitoring and measurement methods are mostly static or manual periodic measurements, making it difficult to achieve real-time and dynamic perception of the surrounding rock condition and failing to provide data support for dynamic adjustment of tunneling parameters. The layout of monitoring sections relies heavily on engineering experience, lacking scientific optimization basis, and easily leading to monitoring blind spots or redundant measuring points. Early warning thresholds are usually fixed values, unable to adapt to the dynamic changes in surrounding rock deformation patterns under different geological conditions. The lack of a closed-loop feedback mechanism between tunneling parameter adjustments and monitoring data causes parameter adjustments to lag behind changes in the surrounding rock condition.

[0006] Therefore, how to systematically, in real time, and dynamically monitor and measure the surrounding rock of the vertical shaft receiving section and the arch of the lower horizontal tunnel, and dynamically optimize the tunneling parameters and provide timely warnings of the risk of surrounding rock instability based on the monitoring data, is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] This invention provides a method for monitoring and measuring the surrounding rock of the receiving section of a vertical shaft, with the aim of enabling systematic, real-time, and dynamic monitoring and measurement of the surrounding rock of the receiving section of the vertical shaft and the arch of the lower horizontal tunnel.

[0008] This invention is achieved through the following technical solution: a method for monitoring and measuring the surrounding rock of a vertical shaft excavation receiving section, comprising the following steps:

[0009] Step S1: Set up monitoring sections and collect surrounding rock deformation and stress response data in real time. Set up multiple monitoring sections in the vertical shaft receiving section and the arch area of ​​the lower horizontal tunnel. Each section includes arch settlement measuring points, horizontal convergence measuring points, crack measuring points and chute displacement measuring points, which are used to collect surrounding rock deformation and stress response data in real time.

[0010] Step S2: Optimize the number and spatial location of monitoring sections based on genetic algorithm. Construct a monitoring section optimization model based on genetic algorithm to optimize the number and spatial location of monitoring sections with the goal of maximizing monitoring information entropy and minimizing deployment cost.

[0011] Step S3: Dynamically optimize the yellow, orange, and red three-level early warning thresholds based on genetic algorithm. According to the real-time collected monitoring data, the genetic algorithm is used to dynamically optimize the yellow, orange, and red three-level early warning thresholds. The thresholds include the top arch settlement threshold, horizontal convergence threshold, crack width threshold, and wellhead displacement threshold.

[0012] Step S4: Optimize the tunneling parameters of the tunneling machine in real time based on the genetic algorithm. With the stability of the surrounding rock as the constraint, construct a multi-objective optimization model for the tunneling parameters, and use the genetic algorithm to optimize the thrust, cutterhead speed, step change distance and penetration depth parameters of the tunneling machine in real time.

[0013] Step S5: Establish a closed-loop feedback mechanism and issue optimized tunneling parameters. Establish a closed-loop feedback mechanism between monitoring data and tunneling control commands, and automatically or semi-automatically issue optimized tunneling parameters to the tunneling machine's local control system.

[0014] Step S6: Trigger graded early warning and execute corresponding safety control measures. When any monitoring parameter exceeds the corresponding early warning threshold, the graded early warning response mechanism is triggered, and the tunneling parameters are automatically adjusted or shutdown, reinforcement, and personnel evacuation measures are executed according to the early warning level.

[0015] This invention provides a method for monitoring and measuring the surrounding rock in the receiving section of a vertical shaft excavation project. It achieves comprehensive real-time sensing of surrounding rock deformation and stress by deploying multiple monitoring sections, including those for crown settlement, horizontal convergence, cracks, and chute displacement. Based on a genetic algorithm, the method optimizes the monitoring section layout with the goals of maximizing information entropy and minimizing cost, and dynamically optimizes the yellow, orange, and red warning thresholds to make the warnings adapt to the deformation patterns of the surrounding rock, avoiding false alarms or missed alarms caused by fixed thresholds. Simultaneously, constrained by the stability of the surrounding rock, the method uses a genetic algorithm to optimize the thrust, cutterhead speed, step-change distance, and penetration parameters of the tunneling machine in real time, and automatically sends the optimized parameters to the tunneling machine's onboard control system through a closed-loop feedback mechanism, eliminating parameter adjustment lag. When the monitored parameters exceed the threshold, a graded warning is triggered, and the tunneling parameters are automatically adjusted or shutdown, reinforcement, and personnel evacuation measures are implemented according to the grade. In summary, this invention enables systematic, real-time, and dynamic monitoring and measurement of the surrounding rock and the arch of the lower horizontal tunnel in the vertical shaft excavation receiving section. It achieves comprehensive, real-time, intelligent, and closed-loop monitoring and measurement of the surrounding rock in the vertical shaft excavation receiving section, significantly improving construction safety, tunneling efficiency, and engineering economy.

[0016] Furthermore, the objective function maxF of the monitoring section optimization model constructed in step S2 is:

[0017] ;

[0018] in, To monitor information entropy, For deployment costs, These are the weighting coefficients, and .

[0019] This scheme achieves a quantitative trade-off and scientific optimization between the amount of information and the cost of monitoring section deployment by constructing an objective function that aims to maximize the entropy of monitoring information and minimize the deployment cost, with normalized weight coefficients.

[0020] Furthermore, the process of dynamically optimizing the early warning threshold in step S3 includes:

[0021] Based on real-time monitoring data sequences, with historical data and surrounding rock mechanical parameters as input, and with the goal of minimizing the sum of false alarm rate and missed alarm rate, a genetic algorithm is used to iteratively update the warning thresholds at all levels.

[0022] This scheme achieves adaptive dynamic optimization of the warning threshold by using historical data and surrounding rock mechanical parameters as input, minimizing the sum of false alarm rate and false alarm rate, and iteratively updating the warning threshold using a genetic algorithm, which effectively improves the accuracy and reliability of the warning.

[0023] Furthermore, the optimization objectives of the multi-objective optimization model for tunneling parameters in step S4 include: maximizing tunneling efficiency, minimizing the degree of surrounding rock disturbance, and minimizing the risk of slippage of the support shoe;

[0024] The constraints include: the real-time monitored crown settlement value is not greater than the orange warning threshold, the real-time monitored horizontal convergence value is not greater than the orange warning threshold, the real-time monitored crack width value is not greater than the orange warning threshold, the real-time monitored chute displacement value is not greater than the orange warning threshold, and the cutterhead torque does not exceed the rated value.

[0025] This scheme aims to maximize tunneling efficiency, minimize surrounding rock disturbance and the risk of slippage of the support shoe, and uses the constraint that all four types of monitoring parameters do not exceed the orange warning threshold and the cutterhead torque does not exceed the rated value. This achieves the quantitative constraint of safety boundaries and equipment protection for tunneling parameters in the multi-objective collaborative optimization process.

[0026] Furthermore, the specific steps of step S2 are as follows:

[0027] S2.1 Set the number of iterations for the genetic algorithm and the number of individuals to be solved in each generation of the population;

[0028] S2.2 Establish a monitoring section optimization configuration model, which includes an objective function and overall constraints, wherein the monitoring information entropy , For the first The probability distribution of data collected from each monitoring point; the total constraints include: the number of monitoring sections is not less than 3 and not more than 10, the distance between adjacent monitoring sections is not less than 2 meters and not more than 5 meters, and the distance between the monitoring section and the end point of the receiving section is not less than 1 meter;

[0029] S2.3. The number and spatial coordinates of the monitoring sections are encoded to obtain the encoding value of the individual to be solved. Based on the overall constraint conditions, an initial population is randomly generated from several individuals to be solved. This initial population is called the parent population, and the individuals to be solved in the parent population are called parent individuals. The encoding value of each parent individual includes the number of monitoring sections and the spatial coordinates of each section.

[0030] S2.4 Substitute the encoded values ​​of each individual in the parent population into the simulation model of surrounding rock deformation and stress in the vertical shaft receiving section and the arch area of ​​the lower horizontal tunnel to obtain the information entropy and deployment cost of each monitoring section. The information entropy and deployment cost are the simulation results of the simulation.

[0031] S2.5. Calculate the objective function value of each individual in the parent population based on each simulation result, directly use the objective function value as the fitness value for cross-section optimization, and sort all parent individuals in descending order of fitness value.

[0032] S2.6. Save the top M parent individuals with the largest fitness values ​​in the parent population. Then, select parent individuals from all parent individuals other than the top M parent individuals with the largest fitness values ​​through a roulette wheel and perform crossover and mutation operations to obtain offspring individuals. Calculate the fitness values ​​of the offspring individuals after crossover and mutation, sort them, and reinsert the offspring individuals into the parent population according to their fitness values. Select a set number of individuals to be solved to form a new parent population, and then return to S2.4.

[0033] S2.7 Repeat S2.4 to S2.6 until the number of iterations is reached or the objective function value is within the specified threshold range. The resulting parent population is the feasible solution set, and the parent individuals in the parent population are the feasible individuals.

[0034] This solution deeply integrates genetic algorithms with engineering constraints of the vertical shaft receiving section and simulation models of surrounding rock deformation and stress. It quantifies the amount of monitoring information using the information entropy formula, thereby achieving scientific and intelligent optimization of the number and spatial location of monitoring sections. This reduces deployment costs while ensuring the amount of monitoring information and provides multiple feasible solutions for engineering decision-making.

[0035] Furthermore, the specific steps of step S3 are as follows:

[0036] S3.1 Set the number of iterations for the genetic algorithm and the number of individuals to be solved in each generation of the population;

[0037] S3.2. Establish a dynamic optimization model for the early warning threshold. The dynamic optimization model for the early warning threshold includes an objective function and overall constraints, wherein the objective function minG is:

[0038] ;

[0039] in, For false alarm rate, The false negative rate is defined as follows: the yellow warning threshold is less than the orange warning threshold; the orange warning threshold is less than the red warning threshold; the red warning threshold for the top arch settlement is not greater than 80% of the ultimate displacement of the surrounding rock; the red warning threshold for horizontal convergence is not greater than 80% of the ultimate convergence of the surrounding rock; the red warning threshold for crack width is not greater than 80% of the ultimate crack width of the surrounding rock; and the red warning threshold for wellhead displacement is not greater than 80% of the ultimate displacement of the wellhead.

[0040] S3.3. The top arch settlement threshold, horizontal convergence threshold, crack width threshold and wellhead displacement threshold corresponding to the yellow, orange and red levels are encoded to obtain the encoded value of the individual to be solved. The encoded value of each individual to be solved corresponds to a set of yellow, orange and red level early warning thresholds for top arch settlement, horizontal convergence, crack width and wellhead displacement. Based on the overall constraint conditions, an initial population is randomly generated from several individuals to be solved, and this initial population is called the parent population.

[0041] S3.4 Substitute the encoded values ​​of each individual in the parent population into the early warning simulation model based on historical monitoring data and surrounding rock mechanical parameters to obtain the false alarm rate and the missed alarm rate. The false alarm rate and the missed alarm rate are the simulation results of the simulation.

[0042] S3.5. Calculate the objective function value of each individual in the parent population based on each simulation result, and then calculate and sort the fitness value of each individual in the parent population based on the fitness function optimized according to the warning threshold.

[0043] S3.6. Save the top M parent individuals with the largest fitness values ​​in the parent population. Then, select parent individuals from all parent individuals other than the top M parent individuals with the largest fitness values ​​through a roulette wheel and perform crossover and mutation operations to obtain offspring individuals. Calculate the fitness values ​​of the offspring individuals after crossover and mutation, sort them, and reinsert the offspring individuals into the parent population according to their fitness values. Select a set number of individuals to be solved to form a new parent population, and then return to S3.4.

[0044] S3.7 Repeat S3.4 to S3.6 until the number of iterations is reached or the objective function value is within the specified threshold range. The resulting parent population is the feasible solution set, and the parent individuals in the parent population are the feasible individuals. The encoded values ​​of the feasible individuals correspond to the optimal warning thresholds at each level.

[0045] This scheme aims to minimize the sum of false alarm and false negative rates, and uses a three-level threshold logic progression and a red threshold not exceeding 80% of the limit as constraints. It employs a genetic algorithm to jointly optimize multi-parameter, multi-level early warning thresholds, achieving dynamic self-adaptation, logical consistency, and safety margin assurance in the early warning mechanism, thus significantly improving the accuracy and reliability of early warnings.

[0046] Furthermore, the fitness function for optimizing the early warning threshold is:

[0047] .

[0048] In this scheme, by limiting the fitness function for optimizing the warning threshold to the reciprocal of the objective function, the evolution direction of the genetic algorithm is kept consistent with the optimization objective of minimizing the sum of false alarm rate and false negative rate, and good numerical discrimination and computational efficiency are provided.

[0049] A smart monitoring and measurement system for the surrounding rock of the receiving section during vertical shaft excavation includes:

[0050] The data acquisition module is used to collect real-time data on the top arch settlement, horizontal convergence, crack width, and chute displacement of each monitoring section.

[0051] The cross-section optimization module is used to execute the monitoring cross-section optimization configuration model based on the genetic algorithm. The monitoring cross-section optimization configuration model aims to maximize the monitoring information entropy and minimize the deployment cost, and optimizes the number and spatial location of monitoring cross-sections.

[0052] The threshold dynamic adjustment module is used to execute a dynamic optimization model for early warning thresholds based on a genetic algorithm, taking real-time collected monitoring data, historical monitoring data and surrounding rock mechanical parameters as input. The dynamic optimization model for early warning thresholds aims to minimize the sum of false alarm rate and false alarm rate, and dynamically optimizes the three-level early warning thresholds of yellow, orange and red. The thresholds include the top arch settlement threshold, horizontal convergence threshold, crack width threshold and chute displacement threshold.

[0053] The tunneling parameter optimization module is used to execute a multi-objective optimization model for tunneling parameters based on a genetic algorithm. The multi-objective optimization model for tunneling parameters optimizes the thrust, cutterhead speed, step change distance, and penetration depth parameters of the tunneling machine under the constraint of surrounding rock stability.

[0054] The closed-loop feedback control module is used to establish a closed-loop feedback mechanism between monitoring data and tunneling control commands, and automatically or semi-automatically sends the optimized tunneling parameters to the tunneling machine's local control system.

[0055] The graded early warning module is used to trigger the graded early warning response mechanism when any parameter monitored in real time exceeds the corresponding early warning threshold optimized by the threshold dynamic adjustment module, and automatically adjust the tunneling parameters or implement shutdown, reinforcement and personnel evacuation measures according to the early warning level.

[0056] This solution constructs an intelligent monitoring and measurement system comprising six modules: data acquisition, cross-section optimization, dynamic threshold adjustment, tunneling parameter optimization, closed-loop feedback control, and hierarchical early warning. The cross-section optimization, threshold adjustment, and parameter optimization are all based on genetic algorithms, realizing full-process intelligent monitoring and measurement of the surrounding rock in the vertical shaft tunneling receiving section, multi-source data fusion-driven monitoring, and closed-loop linkage of monitoring and control.

[0057] A monitoring and measurement device for the surrounding rock of a vertical shaft excavation receiving section includes:

[0058] At least one processor;

[0059] At least one memory for storing computer programs;

[0060] At least one data acquisition interface is provided for connecting the top arch settlement sensor, horizontal convergence sensor, crack width sensor and wellhead displacement sensor.

[0061] At least one actuator interface for connecting to the control system of the tunneling machine;

[0062] The memory stores a computer program that can be executed by the at least one processor. When the computer program is executed by the at least one processor, it causes the at least one processor to perform the above-described method for monitoring and measuring the surrounding rock of a vertical shaft tunneling receiving section.

[0063] This solution integrates a processor and memory storing a computer program for monitoring and measuring the surrounding rock of a vertical shaft tunneling receiving section, along with a standardized data acquisition interface and an actuator interface, into a single device. This achieves the hardware transformation of the monitoring and measurement method for the surrounding rock of a vertical shaft tunneling receiving section, providing plug-and-play engineering equipment support for intelligent monitoring driven by genetic algorithms.

[0064] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for monitoring and measuring the surrounding rock of a vertical shaft tunneling receiving section as described above. Attached Figure Description

[0065] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0066] Figure 1 This is a flowchart illustrating an embodiment of a method for monitoring and measuring the surrounding rock in the receiving section of a vertical shaft excavation according to the present invention;

[0067] Figure 2 This is a schematic diagram of an embodiment of a monitoring and measurement device for the surrounding rock of a vertical shaft excavation receiving section according to the present invention.

[0068] The attached diagram shows the markings and corresponding component names:

[0069] 1. Monitoring and measurement equipment; 2. Memory; 3. Processor; 4. Computer program. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0071] Example 1

[0072] This embodiment takes the construction of the receiving section of the SBE tunneling machine in the vertical shaft of a pumped storage power station as the application scenario. The remaining rock thickness of the receiving section is 10m→0m, the surrounding rock is Class III2 and IV, with well-developed joints and broken rock mass, and the arch area of ​​the lower horizontal tunnel is the key monitoring area.

[0073] like Figure 1 As shown, this embodiment provides a method for monitoring and measuring the surrounding rock in the receiving section of a vertical shaft excavation, including the following steps:

[0074] Step S1: Deploy monitoring sections and collect surrounding rock deformation and stress response data in real time. Multiple monitoring sections are deployed in the vertical shaft receiving section and the arch area of ​​the lower horizontal tunnel. Each section includes arch settlement measuring points, horizontal convergence measuring points, crack measuring points, and chute displacement measuring points to collect surrounding rock deformation and stress response data in real time. Specifically:

[0075] Settlement monitoring points for the arch crown: These are located within 3m of the centerline of the arch crown of the lower horizontal tunnel and on both sides. A static level is used to monitor the vertical deformation of the rock mass at the arch crown.

[0076] Horizontal convergence measuring points: arranged symmetrically on the sidewall of the lower horizontal tunnel, and laser convergence meters are used to monitor the horizontal convergence deformation of the rock mass;

[0077] Crack measuring points: These are placed at visible cracks in the surrounding rock, and crack gauges are used to monitor changes in crack width and extension.

[0078] Displacement measuring points at the chute opening: These are arranged in the rock mass surrounding the chute, and displacement gauges are used to monitor the displacement and deformation of the chute opening caused by excavation and slag removal.

[0079] Real-time data on surrounding rock deformation and stress response are collected at each measuring point and transmitted to the monitoring system.

[0080] Step S2: Optimize the number and spatial location of monitoring sections based on genetic algorithms. Construct a monitoring section optimization model based on genetic algorithms. With the objectives of maximizing monitoring information entropy and minimizing deployment costs, optimize the number and spatial location of monitoring sections. Specifically:

[0081] The objective function of the monitoring section optimization model constructed in this embodiment is:

[0082] ;

[0083] in, To monitor information entropy, The deployment cost includes the comprehensive cost of measuring point installation, maintenance, and data acquisition. These are the weighting coefficients, and In this embodiment, we take Prioritize ensuring the integrity of monitoring information; through weighting coefficients and And satisfy The normalization constraint achieves a quantitative trade-off between maximizing monitoring information entropy and minimizing deployment costs, which are two conflicting objectives. The weighting ratio can be flexibly adjusted according to actual engineering needs, avoiding monitoring blind spots or cost waste caused by subjective experience-based deployment. This embodiment uses monitoring information entropy... As one of the optimization objectives, it ensures the acquisition of the maximum amount of information on surrounding rock deformation and stress response with a limited number of monitoring sections, thus solving the problem of insufficient information assessment basis for the layout of measuring points in traditional methods; and considering the layout cost... As one of the optimization objectives, economic indicators such as the number of sensors and installation and maintenance costs are incorporated into the optimization model to minimize costs while ensuring the amount of monitoring information.

[0084] Step S3: Dynamically optimize the yellow, orange, and red warning thresholds based on a genetic algorithm. Based on real-time monitoring data, a genetic algorithm is used to dynamically optimize the yellow, orange, and red warning thresholds. The thresholds include the arch settlement threshold, horizontal convergence threshold, crack width threshold, and chute displacement threshold. In this embodiment, the process of dynamically optimizing the warning thresholds includes:

[0085] Based on real-time monitoring data sequences, and using historical data and surrounding rock mechanical parameters (elastic modulus, Poisson's ratio, ultimate displacement, etc.) as input, a genetic algorithm is used to iteratively update the warning thresholds at all levels with the goal of minimizing the sum of false alarm rate and false alarm rate.

[0086] In this embodiment, the warning threshold is no longer a fixed value, but is dynamically adjusted based on real-time monitoring data sequences and surrounding rock mechanical parameters. This allows the warning threshold to adapt to different geological conditions and the deformation patterns of the surrounding rock at different tunneling stages, solving the technical problem that fixed thresholds are prone to false alarms or missed alarms under complex geological conditions. The optimization objective is to minimize the sum of the false alarm rate and the missed alarm rate, avoiding the problem of optimizing one indicator leading to the deterioration of another, achieving overall optimal warning accuracy, and effectively reducing construction interruption losses caused by false alarms and safety risks caused by missed alarms. Historical monitoring data is used as input for the iterative update of the genetic algorithm, allowing the warning threshold to be continuously optimized as monitoring data accumulates, with increasingly accurate warnings in the later stages of tunneling, realizing the self-learning and continuous evolution of the warning model. Including surrounding rock mechanical parameters (such as elastic modulus, Poisson's ratio, and ground stress) as input variables ensures that the optimization process of the warning threshold fully considers the physical and mechanical properties of the surrounding rock, improving the matching degree between the warning threshold and the actual engineering conditions.

[0087] Step S4: Optimize the tunneling parameters of the tunneling machine in real time based on the genetic algorithm. With the stability of the surrounding rock as a constraint, a multi-objective optimization model for the tunneling parameters is constructed. The genetic algorithm is used to optimize the thrust, cutterhead speed, step change distance and penetration parameters of the tunneling machine in real time. In this embodiment, the optimization objectives of the multi-objective optimization model for the tunneling parameters include: maximizing tunneling efficiency, minimizing the degree of surrounding rock disturbance and minimizing the risk of slippage of the support shoe.

[0088] The constraints include: the real-time monitored crown settlement value is not greater than the orange warning threshold, the real-time monitored horizontal convergence value is not greater than the orange warning threshold, the real-time monitored crack width value is not greater than the orange warning threshold, the real-time monitored chute displacement value is not greater than the orange warning threshold, and the cutterhead torque does not exceed the rated value.

[0089] In this embodiment, the parameters are optimized according to different tunneling stages of the receiving section:

[0090] For sections with remaining rock mass thickness of 10–5 m: thrust 2000–3000 kN, cutterhead speed 3.0 r / min, step change stroke 450–500 mm, penetration 1–3 mm / r;

[0091] For sections with remaining rock mass thickness of 5–3 m: thrust 1800–2000 kN, cutterhead rotation speed 2.0–3.0 r / min, step change stroke 300–400 mm, penetration depth 1–3 mm / r;

[0092] For the remaining rock mass thickness of 3-0m: thrust 1000-1800kN, cutterhead speed 2.0r / min, step change 300mm, penetration 1-2mm / r.

[0093] Through real-time iterative optimization using genetic algorithms, tunneling efficiency is effectively improved and the degree of surrounding rock disturbance is reduced while ensuring the stability of the surrounding rock.

[0094] This embodiment simultaneously optimizes three objectives: tunneling efficiency, surrounding rock disturbance, and the risk of support shoe slippage. It maximizes tunneling speed while ensuring surrounding rock stability, and minimizes surrounding rock disturbance while ensuring equipment safety, achieving a balance between safety, efficiency, and equipment reliability. Hard constraints are imposed, ensuring that four parameters monitored in real-time—roof settlement, horizontal convergence, crack width, and chute displacement—do not exceed the orange warning threshold. This provides clear safety boundaries for optimizing tunneling parameters, ensuring that the optimization process always occurs within a safe range and avoiding the risk of sacrificing safety for efficiency. The cutterhead torque is kept below a certain limit. The rated values ​​serve as constraints to prevent equipment overload damage caused by excessive adjustment of tunneling parameters, thus extending the service life of the tunneling machine and reducing the equipment failure rate. The constraints cover four types of monitoring parameters (arch settlement, horizontal convergence, crack width, and chute displacement), forming a complete correspondence with the monitoring system and avoiding safety blind spots caused by the omission of a certain monitoring parameter. In addition, the orange warning threshold was chosen as the constraint boundary instead of yellow or red. The yellow threshold is too conservative and will excessively limit tunneling efficiency, while the red threshold is too dangerous and will reduce the safety margin. Orange, as an intermediate level, achieves a reasonable balance between safety and efficiency.

[0095] Step S5: Establish a closed-loop feedback mechanism and issue optimized tunneling parameters. Establish a closed-loop feedback mechanism between monitoring data and tunneling control commands, automatically or semi-automatically issuing the optimized tunneling parameters to the tunneling machine's on-board control system. Specifically:

[0096] The data acquisition module transmits real-time monitoring data to the monitoring system, while the cross-section optimization, threshold adjustment, and parameter optimization modules perform real-time calculations.

[0097] The optimized tunneling parameters are automatically sent to the SBE tunneling machine's control system via a closed-loop system, or can be semi-automatically adjusted by on-site technicians, eliminating the lag in parameter adjustment.

[0098] The monitoring data is linked with the slag discharge system. When the displacement of the slag outlet approaches the threshold, the slag discharge speed is automatically adjusted to ensure the stability of the surrounding rock during the slag discharge process.

[0099] Step S6: Trigger tiered early warning and execute corresponding safety control measures. When any monitoring parameter exceeds the corresponding early warning threshold, the tiered early warning response mechanism is triggered, and the tunneling parameters are automatically adjusted or shutdown, reinforcement, and personnel evacuation measures are executed according to the early warning level. Specifically:

[0100] Yellow alert: Parameters are close to the threshold, the surrounding rock is slightly deformed, automatically reduce the tunneling machine's advance speed and cutterhead speed, increase the monitoring frequency, and arrange for dedicated personnel to patrol;

[0101] Orange alert: Parameters exceed the threshold, surrounding rock is significantly deformed, suspend tunneling, tighten the support shoe, monitor in real time 24 hours a day, and analyze the cause of deformation;

[0102] Red Alert: Parameters are severely out of control, and the surrounding rock is close to instability. Stop the machine immediately, implement surrounding rock reinforcement measures, evacuate personnel from the site, and activate the emergency plan.

[0103] In one embodiment, step S2 specifically involves the following steps:

[0104] S2.1 Set the number of iterations for the genetic algorithm and the number of individuals to be solved in each generation. In this embodiment, the number of iterations is 100 and the number of individuals in each generation is 50.

[0105] S2.2 Establish a monitoring section optimization configuration model. The monitoring section optimization configuration model includes an objective function and overall constraints, among which, the monitoring information entropy , For the first The probability distribution of data collected from each monitoring point; the total constraints include: the number of monitoring sections is not less than 3 and not more than 10, the distance between adjacent monitoring sections is not less than 2 meters and not more than 5 meters, and the distance between the monitoring section and the end point of the receiving section is not less than 1 meter;

[0106] S2.3. The number and spatial coordinates of the monitoring sections are encoded to obtain the encoding value of the individual to be solved. Based on the overall constraint conditions, an initial population is randomly generated from several individuals to be solved. This initial population is called the parent population, and the individuals to be solved in the parent population are called parent individuals. The encoding value of each parent individual includes the number of monitoring sections and the spatial coordinates of each section.

[0107] S2.4 Substitute the encoded values ​​of each individual in the parent population into the simulation model of surrounding rock deformation and stress in the vertical shaft receiving section and the arch area of ​​the lower horizontal tunnel to obtain the information entropy and deployment cost of each monitoring section. The information entropy and deployment cost are the simulation results of the simulation.

[0108] S2.5. Calculate the objective function value of each individual in the parent population based on each simulation result, directly use the objective function value as the fitness value for cross-section optimization, and sort all parent individuals in descending order of fitness value.

[0109] S2.6. Save the top M parent individuals with the largest fitness values ​​in the parent population. Then, select parent individuals from all parent individuals other than the top M parent individuals with the largest fitness values ​​through a roulette wheel and perform crossover and mutation operations to obtain offspring individuals. Calculate the fitness values ​​of the offspring individuals after crossover and mutation and sort them. Reinsert the offspring individuals into the parent population according to their fitness values. Select a set number of individuals to be solved to form a new parent population. Then return to S2.4. In this embodiment, M=10.

[0110] S2.7. Repeat S2.4 to S2.6 until the number of iterations is reached or the objective function value is within the specified threshold range. The resulting parent population is the feasible solution set, and the parent individuals in the parent population are the feasible individuals. In this embodiment, the iteration is repeated 100 times to obtain the optimal monitoring section layout scheme: a total of 6 monitoring sections are set up with a spacing of 3m, covering key areas such as the rock fracture zone and the area around the wellhead, eliminating redundant measuring points and reducing the layout cost.

[0111] In one embodiment, step S3 specifically involves the following steps:

[0112] S3.1 Set the number of iterations for the genetic algorithm and the number of individuals to be solved in each generation. In this embodiment, the number of iterations is 80 and the number of individuals in each generation is 40.

[0113] S3.2 Establish a dynamic optimization model for the early warning threshold. The dynamic optimization model for the early warning threshold includes an objective function and overall constraints, wherein the objective function minG is:

[0114] ;

[0115] in, For false alarm rate, The false negative rate is used for the following constraints: the yellow warning threshold is less than the orange warning threshold, the orange warning threshold is less than the red warning threshold, the red warning threshold for the top arch settlement is not greater than 80% of the ultimate displacement of the surrounding rock, the red warning threshold for horizontal convergence is not greater than 80% of the ultimate convergence of the surrounding rock, the red warning threshold for crack width is not greater than 80% of the ultimate crack width of the surrounding rock, and the red warning threshold for the wellhead displacement is not greater than 80% of the ultimate displacement of the wellhead.

[0116] S3.3. The top arch settlement threshold, horizontal convergence threshold, crack width threshold and wellhead displacement threshold corresponding to the yellow, orange and red levels are encoded to obtain the encoded value of the individual to be solved. The encoded value of each individual to be solved corresponds to a set of yellow, orange and red level early warning thresholds for top arch settlement, horizontal convergence, crack width and wellhead displacement. Based on the overall constraint conditions, an initial population is randomly generated from several individuals to be solved, and this initial population is called the parent population.

[0117] S3.4 Substitute the encoded values ​​of each individual in the parent population into the early warning simulation model based on historical monitoring data and surrounding rock mechanical parameters to obtain the false alarm rate and the missed alarm rate. The false alarm rate and the missed alarm rate are the simulation results of the simulation.

[0118] S3.5. Calculate the objective function value of each individual in the parent population based on each simulation result. Then, calculate and sort the fitness value of each individual in the parent population according to the fitness function optimized by the warning threshold. The fitness function optimized by the warning threshold is:

[0119] ;

[0120] S3.6. Save the top M parent individuals with the largest fitness values ​​in the parent population. Then, select parent individuals from all parent individuals other than the top M parent individuals with the largest fitness values ​​through a roulette wheel and perform crossover and mutation operations to obtain offspring individuals. Calculate the fitness values ​​of the offspring individuals after crossover and mutation, sort them, and reinsert the offspring individuals into the parent population according to their fitness values. Select a set number of individuals to be solved to form a new parent population, and then return to S3.4. In this embodiment, M=8.

[0121] S3.7 Repeat S3.4 to S3.6 until the number of iterations is reached or the objective function value is within the specified threshold range. The resulting parent population is the feasible solution set, and the parent individuals in the parent population are the feasible individuals. The coded values ​​in the feasible individuals correspond to the optimal warning thresholds at each level. In this embodiment, the iteration is repeated up to 80 times to obtain the optimal warning thresholds. For example, the yellow warning threshold for crown settlement is 2mm / d, the orange threshold is 3mm / d, and the red threshold is 4mm / d. The horizontal convergence thresholds are 1.5mm / d for yellow, 2.5mm / d for orange, and 3.5mm / d for red, achieving dynamic matching between the thresholds and the surrounding rock condition. The false alarm rate and the missed alarm rate are both reduced to below 5%.

[0122] The present invention provides a method for monitoring and measuring the surrounding rock in the receiving section of a vertical shaft excavation, which has the following technical advantages:

[0123] By setting up multiple monitoring sections, including top arch settlement measuring points, horizontal convergence measuring points, crack measuring points, and chute displacement measuring points, in the vertical shaft receiving section and the lower horizontal tunnel arch area through step S1, comprehensive and real-time acquisition of surrounding rock deformation and stress response data is realized, solving the problems of monitoring blind spots and data lag in the existing technology.

[0124] By constructing a monitoring section optimization model based on a genetic algorithm in step S2, with the goal of maximizing monitoring information entropy and minimizing deployment costs, intelligent optimization of the number and spatial location of monitoring sections is achieved. This reduces the redundancy of measuring points and lowers monitoring costs while ensuring the amount of monitoring information.

[0125] In step S3, a genetic algorithm is used to dynamically optimize the yellow, orange, and red three-level early warning thresholds (including four types of thresholds: top arch settlement, horizontal convergence, crack width, and wellhead displacement) based on real-time collected monitoring data. This allows the early warning thresholds to adapt to changes in the deformation law of the surrounding rock, avoiding false alarms or missed alarms caused by fixed thresholds, and improving the accuracy and timeliness of early warnings.

[0126] By using the surrounding rock stability as a constraint in step S4, a multi-objective optimization model for tunneling parameters is constructed. A genetic algorithm is then used to optimize the thrust, cutterhead speed, step change distance, and penetration depth parameters of the tunneling machine in real time. This achieves adaptive matching between tunneling parameters and the surrounding rock condition, thereby improving tunneling efficiency while ensuring the stability of the surrounding rock.

[0127] By establishing a closed-loop feedback mechanism between monitoring data and tunneling control commands in step S5, the optimized tunneling parameters are automatically or semi-automatically sent to the tunneling machine control system. This eliminates the time delay in parameter adjustment due to changes in the surrounding rock condition, which is common in traditional methods, and enables real-time dynamic control of the tunneling process.

[0128] Step S6 triggers a tiered early warning response mechanism when any monitored parameter exceeds the corresponding early warning threshold. Based on the early warning level, the tunneling parameters are automatically adjusted or measures such as shutdown, reinforcement, and personnel evacuation are implemented, forming a complete closed loop from monitoring and perception, threshold early warning to safety handling. This effectively reduces the safety risks of rockfall, collapse, and overall instability in the receiving section.

[0129] In summary, this invention achieves comprehensive, real-time, intelligent, and closed-loop monitoring and measurement of surrounding rock in the high-risk construction phase of the shaft excavation receiving section, significantly improving construction safety, tunneling efficiency, and engineering economy.

[0130] Example 2

[0131] This embodiment discloses an intelligent monitoring and measurement system for the surrounding rock of the receiving section during vertical shaft excavation, including:

[0132] The data acquisition module is used to collect data on the top arch settlement, horizontal convergence, crack width, and chute displacement of each monitoring section in real time. In this embodiment, the data acquisition module is connected to the top arch settlement sensor, horizontal convergence sensor, crack width sensor, and chute displacement sensor to collect four types of monitoring data in real time and transmit them to the system main control unit. The system main control unit refers to the central processing unit, host computer, or main control unit in the prior art, which is used to receive data, run algorithms, optimize parameters, and issue instructions.

[0133] In this embodiment, the arch settlement sensor is used to monitor the vertical settlement deformation of the surrounding rock of the arch in the vertical shaft receiving section and the lower horizontal tunnel in real time, obtain the settlement rate and cumulative settlement value of the arch rock mass, and determine whether the surrounding rock of the arch has become loose, voided, or unstable, providing data basis for the stability evaluation of the arch; the horizontal convergence sensor is used to monitor the relative horizontal displacement of the surrounding rock on both sides of the tunnel / cavity in real time, obtain the deformation rate and cumulative convergence value of the surrounding rock squeezing and converging into the tunnel, and determine whether the surrounding rock has undergone squeezing deformation, excessive convergence, or local instability, providing data basis for the overall stability of the cavity. The system provides data for the assessment; the crack width sensor is used to monitor the opening, expansion rate and cumulative width of cracks on the surrounding rock surface in real time, identify whether the cracks are continuously expanding, and determine whether the surrounding rock has experienced tensile failure, shear failure or structural slippage, providing a basis for early warning of surrounding rock damage risk; the wellhead displacement sensor is used to monitor the displacement changes of the rock mass around the wellhead in real time, obtain the offset and deformation rate of the wellhead under the action of tunneling disturbance and slag discharge impact, determine whether the wellhead structure is stable, and prevent the risk of wellhead blockage and collapse due to surrounding rock deformation;

[0134] The cross-section optimization module is used to execute the monitoring cross-section optimization configuration model based on the genetic algorithm. The monitoring cross-section optimization configuration model aims to maximize the monitoring information entropy and minimize the deployment cost, and optimizes the number and spatial location of monitoring cross-sections.

[0135] The threshold dynamic adjustment module is used to execute a dynamic optimization model for early warning thresholds based on a genetic algorithm, taking real-time collected monitoring data, historical monitoring data, and surrounding rock mechanical parameters as input. The dynamic optimization model for early warning thresholds aims to minimize the sum of false alarm rate and false alarm rate, and dynamically optimizes the three levels of early warning thresholds: yellow, orange, and red. The thresholds include the crown settlement threshold, horizontal convergence threshold, crack width threshold, and chute displacement threshold. Each sensor uploads the collected surrounding rock deformation data to the system's main control unit, providing basic data for cross-section optimization, threshold optimization, and tunneling parameter optimization.

[0136] The tunneling parameter optimization module is used to execute a multi-objective optimization model for tunneling parameters based on a genetic algorithm. The multi-objective optimization model optimizes the thrust, cutterhead speed, step change distance, and penetration depth parameters of the tunneling machine under the constraint of surrounding rock stability.

[0137] The closed-loop feedback control module is used to establish a closed-loop feedback mechanism between monitoring data and tunneling control commands, and automatically or semi-automatically sends the optimized tunneling parameters to the tunneling machine's local control system.

[0138] The graded early warning module is used to trigger the graded early warning response mechanism when any parameter monitored in real time exceeds the corresponding early warning threshold optimized by the threshold dynamic adjustment module, and automatically adjust the tunneling parameters or implement shutdown, reinforcement and personnel evacuation measures according to the early warning level.

[0139] In this embodiment, the six modules work together to form a complete intelligent monitoring chain from data acquisition, cross-section optimization, threshold adjustment, parameter optimization, closed-loop control to hierarchical early warning, realizing full-process automation and intelligence of monitoring and measurement of surrounding rock in the vertical shaft tunneling receiving section.

[0140] The threshold dynamic adjustment module takes real-time collected monitoring data, historical monitoring data and surrounding rock mechanical parameters as inputs, realizes the fusion drive of multi-source data, and makes the dynamic optimization of the warning threshold fully integrate real-time status, historical patterns and physical and mechanical properties.

[0141] The closed-loop feedback control module automatically or semi-automatically sends the optimized tunneling parameters to the tunneling machine control system, and the graded early warning module triggers a response mechanism when the monitored parameters exceed the threshold. Together, they form a complete closed loop of monitoring, optimization, control, and re-monitoring, eliminating the time delay in parameter adjustment that lags behind changes in the surrounding rock condition in traditional methods.

[0142] The tiered early warning module explicitly references the corresponding early warning thresholds optimized by the threshold dynamic adjustment module, forming a logical connection with the threshold dynamic adjustment module. This ensures that the thresholds used for early warning are the latest optimized dynamic thresholds rather than fixed thresholds, thus improving the accuracy of early warnings.

[0143] The graded early warning module automatically adjusts tunneling parameters or implements shutdown, reinforcement, and personnel evacuation measures based on the early warning level, achieving a safety closed loop from early warning to response. It adopts differentiated response strategies according to risk levels, maintaining tunneling through parameter adjustments during low-risk situations and decisively shutting down and activating emergency plans during high-risk situations. The six modules have clearly defined functions and work in tandem, allowing for deployment as a whole system at the shaft tunneling construction site, or allowing selected modules to operate independently according to actual needs, demonstrating good engineering applicability and scalability.

[0144] Example 3

[0145] like Figure 2 As shown, this embodiment discloses a monitoring and measurement device 1 for the surrounding rock of the shaft excavation receiving section, comprising:

[0146] At least one processor 3 is used to execute computer programs to implement all the operational logic of the above-mentioned monitoring and measurement methods;

[0147] At least one memory 2 is used to store computer programs, monitoring data, optimization model parameters, etc.

[0148] At least one data acquisition interface is provided for connecting the top arch settlement sensor, horizontal convergence sensor, crack width sensor and wellhead displacement sensor.

[0149] At least one actuator interface is provided for connecting to the tunneling machine's local control system to issue optimized tunneling parameters and control commands.

[0150] The memory 2 stores a computer program 4 that can be executed by at least one processor. When the computer program 4 is executed by at least one processor 3, it causes at least one processor 3 to perform a method for monitoring and measuring the surrounding rock of a vertical shaft tunneling receiving section as described in the above embodiment.

[0151] Example 4

[0152] This embodiment also discloses a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the steps of a method for monitoring and measuring the surrounding rock of a vertical shaft tunneling receiving section as described in the above embodiment.

[0153] It should be noted that the above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for monitoring and measuring surrounding rock of a receiving section of a shaft sinking, characterized in that, The method comprises the following steps: Step S1: laying out monitoring sections and collecting surrounding rock deformation and stress response data in real time, laying out a plurality of monitoring sections in the shaft receiving section and the top arch region of the horizontal tunnel, each section comprising a top arch settlement measuring point, a horizontal convergence measuring point, a crack measuring point and a chute opening displacement measuring point, for collecting surrounding rock deformation and stress response data in real time; Step S2: optimizing the number and spatial position of the monitoring sections based on a genetic algorithm, constructing a monitoring section optimization model based on the genetic algorithm, and optimizing the number and spatial position of the monitoring sections to maximize monitoring information entropy and minimize layout cost; Step S3: dynamically optimizing yellow, orange and red three-level early warning thresholds based on the genetic algorithm, dynamically optimizing the yellow, orange and red three-level early warning thresholds by using the genetic algorithm according to the monitoring data collected in real time, the thresholds comprising a top arch settlement threshold, a horizontal convergence threshold, a crack width threshold and a chute opening displacement threshold; Step S4: optimizing tunneling machine tunneling parameters in real time based on the genetic algorithm, constructing a tunneling parameter multi-objective optimization model with surrounding rock stability as a constraint condition, and optimizing the thrust, cutter head speed, step change stroke and penetration degree parameters of the tunneling machine in real time by using the genetic algorithm; Step S5: establishing a closed-loop feedback mechanism and issuing the optimized tunneling parameters, establishing a closed-loop feedback mechanism between the monitoring data and the tunneling control instructions, and automatically or semi-automatically issuing the optimized tunneling parameters to the tunneling machine local control system; Step S6: triggering a hierarchical early warning and executing corresponding safety control measures, triggering a hierarchical early warning response mechanism when any monitoring parameter exceeds the corresponding early warning threshold, and automatically adjusting the tunneling parameters or executing a shutdown, reinforcement or personnel evacuation measure according to the early warning level.

2. The method according to claim 1, wherein, The objective function of the monitoring section optimization model constructed in step S2 is: ； wherein, for monitoring information entropy, for laying cost, for weight coefficient, and .

3. The method for monitoring and measuring surrounding rock of a receiving section of a shaft sinking according to claim 1, characterized in that, The process of dynamically optimizing the early warning thresholds in step S3 comprises: According to the real-time monitoring data sequence, taking historical data and surrounding rock mechanical parameters as inputs and minimizing the sum of false alarm rate and missed alarm rate as the target, the genetic algorithm is used to iteratively update the early warning thresholds of each level.

4. The method according to claim 1, wherein, The optimization targets of the tunneling parameter multi-objective optimization model in step S4 include maximizing tunneling efficiency, minimizing surrounding rock disturbance degree and minimizing support shoe slip risk; The constraint conditions include that the real-time monitored top arch settlement value is not greater than the orange early warning threshold, the real-time monitored horizontal convergence value is not greater than the orange early warning threshold, the real-time monitored crack width value is not greater than the orange early warning threshold, the real-time monitored chute opening displacement value is not greater than the orange early warning threshold, and the cutter head torque is not greater than the rated value.

5. The method according to claim 2, wherein, The specific steps of step S2 are: S2.1, setting the number of iterations of the genetic algorithm and the number of individuals to be solved in each generation of the population; S2.2, a monitoring section optimization configuration model is established, the monitoring section optimization configuration model comprising a target function and total constraint conditions, wherein the monitoring information entropy , is the probability distribution of data collected by the i th monitoring point; the total constraint conditions comprise that the number of monitoring sections is not less than 3 and not more than 10, the distance between adjacent monitoring sections is not less than 2 meters and not more than 5 meters, and the distance from the monitoring section to the terminal point of the receiving section is not less than 1 meter. S2.3, encoding the number and spatial position coordinates of the monitoring sections to obtain the encoding value of the individual to be solved, and randomly generating an initial population formed by a plurality of individuals to be solved based on the total constraint condition, letting the initial population be the parent population, letting the individuals to be solved in the parent population be parent individuals, and the encoding value of each parent individual comprising the number of monitoring sections and the spatial position coordinates of each section; S2.4 Substitute the encoded values ​​of each individual in the parent population into the simulation model of surrounding rock deformation and stress in the vertical shaft receiving section and the arch area of ​​the lower horizontal tunnel to obtain the information entropy and deployment cost of each monitoring section. The information entropy and deployment cost are the simulation results of the simulation. S2.

5. Calculate the objective function value of each individual in the parent population based on each simulation result, directly use the objective function value as the fitness value for cross-section optimization, and sort all parent individuals in descending order of fitness value. S2.

6. Save the top M parent individuals with the largest fitness values ​​in the parent population. Then, select parent individuals from all parent individuals other than the top M parent individuals with the largest fitness values ​​through a roulette wheel and perform crossover and mutation operations to obtain offspring individuals. Calculate the fitness values ​​of the offspring individuals after crossover and mutation, sort them, and reinsert the offspring individuals into the parent population according to their fitness values. Select a set number of individuals to be solved to form a new parent population, and then return to S2.

4. S2.7 Repeat S2.4 to S2.6 until the number of iterations is reached or the objective function value is within the specified threshold range. The resulting parent population is the feasible solution set, and the parent individuals in the parent population are the feasible individuals.

6. The method according to claim 3, wherein, The specific steps of step S3 are as follows: S3.1 Set the number of iterations for the genetic algorithm and the number of individuals to be solved in each generation of the population; S3.

2. Establish a dynamic optimization model for the early warning threshold. The dynamic optimization model for the early warning threshold includes an objective function and overall constraints, wherein the objective function minG is: ； wherein, is the false alarm rate, is the missed alarm rate; the total constraint conditions include: the yellow pre-warning threshold is less than the orange pre-warning threshold, the orange pre-warning threshold is less than the red pre-warning threshold, the red pre-warning threshold of the crown settlement is not greater than 80% of the limit displacement of the surrounding rock, the red pre-warning threshold of the horizontal convergence is not greater than 80% of the limit convergence of the surrounding rock, the red pre-warning threshold of the crack width is not greater than 80% of the limit crack width of the surrounding rock, and the red pre-warning threshold of the displacement of the chute opening is not greater than 80% of the limit displacement of the chute opening. S3.

3. The top arch settlement threshold, horizontal convergence threshold, crack width threshold and wellhead displacement threshold corresponding to the yellow, orange and red levels are encoded to obtain the encoded value of the individual to be solved. The encoded value of each individual to be solved corresponds to a set of yellow, orange and red level early warning thresholds for top arch settlement, horizontal convergence, crack width and wellhead displacement. Based on the overall constraint conditions, an initial population is randomly generated from several individuals to be solved, and this initial population is called the parent population. S3.4 Substitute the encoded values ​​of each individual in the parent population into the early warning simulation model based on historical monitoring data and surrounding rock mechanical parameters to obtain the false alarm rate and the missed alarm rate. The false alarm rate and the missed alarm rate are the simulation results of the simulation. S3.

5. Calculate the objective function value of each individual in the parent population based on each simulation result, and then calculate and sort the fitness value of each individual in the parent population based on the fitness function optimized according to the warning threshold. S3.

6. Save the top M parent individuals with the largest fitness values ​​in the parent population. Then, select parent individuals from all parent individuals other than the top M parent individuals with the largest fitness values ​​through a roulette wheel and perform crossover and mutation operations to obtain offspring individuals. Calculate the fitness values ​​of the offspring individuals after crossover and mutation, sort them, and reinsert the offspring individuals into the parent population according to their fitness values. Select a set number of individuals to be solved to form a new parent population, and then return to S3.

4. S3.7 Repeat S3.4 to S3.6 until the number of iterations is reached or the objective function value is within the specified threshold range. The resulting parent population is the feasible solution set, and the parent individuals in the parent population are the feasible individuals. The encoded values ​​of the feasible individuals correspond to the optimal warning thresholds at each level.

7. The method according to claim 6, wherein, The fitness function for optimizing the early warning threshold is: 。 8. A shaft tunneling receiving section surrounding rock intelligent monitoring measurement system, characterized in that, include: The data acquisition module is used to collect real-time data on the top arch settlement, horizontal convergence, crack width, and chute displacement of each monitoring section. The cross-section optimization module is used to execute the monitoring cross-section optimization configuration model based on the genetic algorithm. The monitoring cross-section optimization configuration model aims to maximize the monitoring information entropy and minimize the deployment cost, and optimizes the number and spatial location of monitoring cross-sections. The threshold dynamic adjustment module is used to execute a dynamic optimization model for early warning thresholds based on a genetic algorithm, taking real-time collected monitoring data, historical monitoring data and surrounding rock mechanical parameters as input. The dynamic optimization model for early warning thresholds aims to minimize the sum of false alarm rate and false alarm rate, and dynamically optimizes the three-level early warning thresholds of yellow, orange and red. The thresholds include the top arch settlement threshold, horizontal convergence threshold, crack width threshold and chute displacement threshold. The tunneling parameter optimization module is used to execute a multi-objective optimization model for tunneling parameters based on a genetic algorithm. The multi-objective optimization model for tunneling parameters optimizes the thrust, cutterhead speed, step change distance, and penetration depth parameters of the tunneling machine under the constraint of surrounding rock stability. The closed-loop feedback control module is used to establish a closed-loop feedback mechanism between monitoring data and tunneling control commands, and automatically or semi-automatically sends the optimized tunneling parameters to the tunneling machine's local control system. The graded early warning module is used to trigger the graded early warning response mechanism when any parameter monitored in real time exceeds the corresponding early warning threshold optimized by the threshold dynamic adjustment module, and automatically adjust the tunneling parameters or implement shutdown, reinforcement and personnel evacuation measures according to the early warning level.

9. A shaft tunneling receiving section surrounding rock monitoring and measuring equipment, characterized by, include: At least one processor; At least one memory for storing computer programs; At least one data acquisition interface is provided for connecting the top arch settlement sensor, horizontal convergence sensor, crack width sensor and wellhead displacement sensor. At least one actuator interface for connecting to the tunneling machine's local control system; The memory stores a computer program that can be executed by the at least one processor. When the computer program is executed by the at least one processor, it causes the at least one processor to perform a method for monitoring and measuring the surrounding rock of a vertical shaft excavation receiving section according to any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of a method for monitoring and measuring the surrounding rock of a vertical shaft tunneling receiving section as described in any one of claims 1 to 7.

Images