Method and System for Generating Construction Methods of Underground Chambers Based on Multi-Source Data Fusion

Abstract

The present invention provides a method and system for generating an underground chamber construction method based on multi-source data fusion. First, geological exploration data, construction machinery operation data, and environmental monitoring data are collected and subjected to spatio-temporal alignment processing to generate a standardized monitoring data set. Then, multi-source feature extraction is performed on the standardized monitoring data set to obtain a construction state feature set and a geological risk feature set. The input is sent to a reinforcement learning model for multi-objective parameter optimization to obtain a preliminary construction method parameter set, which is input into a digital twin simulation system to simulate the mechanical response. When the preset safety threshold is exceeded, the parameters are iteratively optimized. The optimized construction method parameter set is matched and verified with a preset construction specification knowledge base. When there are non-compliance items, manual correction is performed. Finally, full-cycle risk prediction is performed on the corrected construction method parameter set, and dynamic regulation is carried out according to the construction risk distribution map to generate a final construction method instruction set, realizing scientific and intelligent generation of the construction method.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology. Specifically, it relates to a method and system for generating construction methods for underground chambers based on multi-source data fusion. Background Art

[0002] In the field of underground chamber construction, traditional construction methods such as the drill and blast method, the shield method, and layered excavation and support have long dominated. The above methods highly rely on the empirical judgment of construction personnel. When facing complex and changeable geological conditions, they often show problems such as poor geological adaptability, prominent safety hazards, and low efficiency in process coordination. With the expansion of project scale and the increasing complexity of geological conditions, the limitations of traditional construction methods have become more and more obvious, and it has been difficult to meet the comprehensive requirements of modern underground chamber construction for safety, efficiency, and economy.

[0003] On the other hand, although the advanced prediction and stability monitoring technologies for complex geological conditions can provide certain risk warnings, the above technologies have the problem of lagging data update. In actual construction, geological conditions may change at any time, and the lagging risk warning information cannot timely guide the dynamic adjustment of the construction plan, making the construction face potential safety risks. Summary of the Invention

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present invention, embodiments of the present invention provide a method for generating a construction method for an underground chamber based on multi-source data fusion. The method includes:

[0005] Collect geological exploration data, construction machinery operation data, and environmental monitoring data, perform spatio-temporal alignment processing on the geological exploration data, construction machinery operation data, and environmental monitoring data to generate a standardized monitoring data set;

[0006] Extract multi-source features from the standardized monitoring data set to obtain a construction state feature set and a geological risk feature set, and input the construction state feature set and the geological risk feature set into a reinforcement learning model for multi-objective parameter optimization processing to generate a preliminary construction method parameter set;

[0007] Input the preliminary construction method parameter set into a digital twin simulation system for mechanical response simulation processing. When the result of the mechanical response simulation processing exceeds a preset safety threshold, trigger the iterative optimization processing of the construction method parameters to generate an optimized construction method parameter set;

[0008] Perform compliance matching processing on the optimized construction method parameter set and a preset construction specification knowledge base to generate a construction method compliance verification result. When there are non-compliant items in the construction method compliance verification result, trigger manual correction processing to generate a corrected construction method parameter set;

[0009] Perform full-cycle risk prediction processing on the set of corrected construction method parameters to generate a construction risk distribution map, and perform dynamic regulation processing on the set of corrected construction method parameters according to the construction risk distribution map to generate a final construction method instruction set.

[0010] On the other hand, an embodiment of the present invention further provides a system for generating an underground chamber construction method based on multi-source data fusion, including a processor and a machine-readable storage medium. The machine-readable storage medium is connected to the processor. The machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0011] Based on the above aspects, the embodiment of the present invention realizes the full-chain intelligent decision-making for the generation of the underground chamber construction method through the collaborative drive of multi-source data fusion and intelligent algorithms. First, through spatio-temporal alignment processing, discrete geological, mechanical and environmental data are transformed into a standardized data set with spatio-temporal consistency. Then, a reinforcement learning model is used to optimize multi-objective parameters for multi-dimensional features, breaking through the limitation of empirical dependence in determining traditional construction method parameters, realizing the dynamic adaptation of construction status and geological risks. Further, a virtual mechanical response model is constructed through a digital twin simulation system, and an iterative optimization mechanism is triggered in combination with safety thresholds, significantly improving the reliability and safety of construction method parameters; on this basis, a pre-set construction specification knowledge base is introduced for compliance verification, which not only ensures the standardization of construction method parameters, but also retains the integration channel of expert experience through an artificial correction mechanism; finally, through full-cycle risk prediction and dynamic regulation, the construction risk distribution map is transformed into an executable construction method instruction set, realizing the full-process intelligent decision-making closed-loop from data collection to construction method generation. Thus, not only the efficiency and scientificity of construction method generation are improved, but also a construction method generation system with autonomous optimization ability is constructed through the deep coupling of multi-modal data fusion and intelligent algorithms. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a schematic execution flow diagram of a method for generating an underground chamber construction method based on multi-source data fusion provided by an embodiment of the present invention.

[0013] Figure 2 is a schematic diagram of exemplary hardware and software components of a system for generating an underground chamber construction method based on multi-source data fusion provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] The present invention will be specifically described below in conjunction with the accompanying drawings of the specification. Figure 1It is a schematic flow chart of a method for generating an underground chamber construction method based on multi-source data fusion provided by an embodiment of the present invention. The method for generating an underground chamber construction method based on multi-source data fusion will be introduced in detail below.

[0015] In this embodiment, taking the diversion tunnel project of a pumped storage power station as an example, the specific implementation process of the method for generating an underground chamber construction method based on multi-source data fusion is elaborated in detail.

[0016] Step S110: Collect geological exploration data, construction machinery operation data, and environmental monitoring data, perform spatio-temporal alignment processing on the geological exploration data, construction machinery operation data, and environmental monitoring data, and generate a standardized monitoring data set.

[0017] In this embodiment, in order to comprehensively and accurately obtain data related to underground chamber construction, multi-source data needs to be collected. Specifically, for geological exploration data, an SIR-4000 type ground penetrating radar can be used, with an antenna frequency of 100 MHz. A transverse survey line is arranged every 20 m along the tunnel axis to record parameters such as dielectric constant and wave velocity, and the detection accuracy is ±2%. At the same time, borehole core data and in-situ stress measurement data can also be collected. During the collection of construction machinery operation data, triaxial vibration sensors can be installed on a three-boom rock drill jumbo. For example, its range is ±50 g and the sampling frequency is 1 kHz to obtain drill bit vibration frequency data. At the same time, a Beidou high-precision positioning terminal can be installed, with a horizontal error ≤ 2 cm and an elevation error ≤ 5 cm to obtain the positioning data of the rock drill jumbo. In addition, the load data of the muck truck is also recorded. The collection of environmental monitoring data is carried out by deploying a convergence meter (range 0 - 50 mm, resolution 0.01 mm) and a piezometer (range 0 - 2 MPa, accuracy 0.1% FS), and a set of monitoring points is arranged every 5 m to obtain in-tunnel convergence deformation data, groundwater level change data, and air composition monitoring data.

[0018] Among them, the geological exploration data mainly includes borehole core data, ground penetrating radar scan data, and in-situ stress measurement data. The borehole core data is analyzed by taking out the cores from the boreholes to obtain information such as the physical and mechanical properties and structural characteristics of the rock, providing a basis for judging the strength and stability of the rock. The ground penetrating radar scan data uses ground penetrating radar technology to detect underground geological structures, and can identify the location and scale of geological structures such as faults and folds, as well as the change of the stratigraphic interface. The in-situ stress measurement data is obtained through professional in-situ stress measurement equipment to obtain the stress state of the underground rock mass, including the magnitude and direction of the in-situ stress.

[0019] The operation data of the construction machinery covers the positioning data of the rock drilling jumbo, the bit vibration frequency data, and the muck truck load data. The positioning data of the rock drilling jumbo is obtained through the Beidou high-precision positioning terminal installed on the rock drilling jumbo, which can track the position and movement trajectory of the rock drilling jumbo in real time and reflect the progress of the construction. The bit vibration frequency data is collected through a triaxial vibration sensor, which is closely related to the hardness of the rock and the drilling efficiency, and can be used to judge the working state of the bit and the physical properties of the rock. The muck truck load data records the weight of the rock waste transported by the muck truck each time, indirectly reflecting the excavation progress and efficiency.

[0020] The environmental monitoring data includes the in-hole convergence deformation data, the groundwater level change data, and the air composition monitoring data. The in-hole convergence deformation data is obtained by real-time monitoring of the convergence deformation of the cavern using a convergence meter, which is an important index for measuring the stability of the cavern. The groundwater level change data is obtained through a piezometer, which reflects the dynamic change of the groundwater level and has an important impact on the stability of the cavern and the construction safety. The air composition monitoring data is obtained by real-time monitoring of the air composition such as the oxygen content, carbon monoxide content, and dust concentration in the cavern using air quality monitoring equipment to ensure the health and safety of construction workers.

[0021] Among them, step S110 may include:

[0022] Step S111: Establish a conversion mapping relationship between the geological exploration coordinate system and the construction machinery coordinate system, and perform continuous geological interface generation processing on the discrete exploration points in the geological exploration data to obtain a three-dimensional geological model data set.

[0023] In this embodiment, the data collected by the ground penetrating radar uses the WGS84 coordinate system, while the construction requires a construction coordinate system (the origin is the center point of the cavern entrance). Therefore, the seven-parameter method can be used for coordinate conversion, with translation parameters ΔX = 123.456m, ΔY = 45.678m, and ΔZ = -3.210m. For the discrete exploration points in the collected geological exploration data, the spherical semivariogram model (nugget value 0.1, sill value 1.8, range 25m) can be used for Kriging interpolation, and the fault zone area can be interpolated with increased density (grid size 0.3m × 0.3m) to generate a three-dimensional geological model data set with a grid size of 0.5m × 0.5m, with an error ≤ 3cm.

[0024] Step S112: Perform dynamic trajectory calibration processing on the position coordinate data in the operation data of the construction machinery to generate a construction machinery real-time positioning data set, and perform spatial superposition processing on the construction machinery real-time positioning data set and the three-dimensional geological model data set to generate a mechanical-geological spatial association data set.

[0025] In this embodiment, position coordinate data is obtained through the Beidou high-precision positioning terminal installed on the rock drilling jumbo, and the above-mentioned position coordinate data is dynamically trajectory calibrated to generate a set of real-time positioning data of the construction machinery. Then, this set of real-time positioning data of the construction machinery is spatially superimposed on the previously generated three-dimensional geological model data set. For example, the specific position of the rock drilling jumbo in the three-dimensional geological model is determined, thereby generating a set of mechanical-geological spatial correlation data, from which the operation positions of the construction machinery under different geological conditions can be clearly observed.

[0026] Step S113: Perform timestamp synchronization processing on the environmental monitoring data to generate a synchronized environmental monitoring data sequence, and perform time-axis alignment processing on the synchronized environmental monitoring data sequence and the set of mechanical-geological spatial correlation data to generate a set of spatio-temporal correlation monitoring data.

[0027] In this embodiment, since the environmental monitoring data is collected at different time points, in order to make the environmental monitoring data have temporal consistency, timestamp synchronization processing is performed on the environmental monitoring data to generate a synchronized environmental monitoring data sequence. Then, this synchronized environmental monitoring data sequence is aligned with the set of mechanical-geological spatial correlation data on the time axis. For example, the in-hole convergence deformation data at a certain moment is associated with the position information of the rock drilling jumbo in the geological model at the same moment to generate a set of spatio-temporal correlation monitoring data, which is convenient for comprehensively analyzing the spatio-temporal changes during the construction process.

[0028] Step S114: Perform abnormal data cleaning processing on the set of spatio-temporal correlation monitoring data, and use a preset data integrity verification rule to perform missing value repair processing on the cleaned data to generate the set of standardized monitoring data.

[0029] In this embodiment, the Isolation Forest algorithm (contamination = 0.05) can be used to detect outliers in the set of spatio-temporal correlation monitoring data. For example, some data that significantly deviate from the normal range are regarded as outliers and removed. For the missing values in the cleaned data, the KNN imputation method (k = 5, distance metric = Euclidean distance) is used for repair. Finally, the cleaned and repaired data is converted into JSON format, with fields including {"timestamp":"2023-08-20T14:30:00Z","location":[x,y,z],"rock_RQD":75,"drill_speed":2.4m / min}, and an Apache Kafka cluster (3 nodes, throughput ≥ 100,000 messages / second) is deployed to achieve real-time data stream processing, generating a set of standardized monitoring data.

[0030] Step S120: Perform multi-source feature extraction on the standardized monitoring data set to obtain a construction state feature set and a geological risk feature set, and input the construction state feature set and the geological risk feature set into a reinforcement learning model for multi-objective parameter optimization to generate a preliminary construction method parameter set.

[0031] In this embodiment, multi-source feature extraction can be performed on the previously generated standardized monitoring data set. For example, geological risk features such as rock mass strength and geological structure are extracted from geological exploration data, and construction state features such as rock drilling speed and in-hole convergence deformation rate are extracted from construction machinery operation data and environmental monitoring data, thereby obtaining a construction state feature set and a geological risk feature set, and then inputting them into a reinforcement learning model for multi-objective parameter optimization.

[0032] For example, step S120 may include:

[0033] Step S121: Construct a state space set of the reinforcement learning model, where the state space set includes a rock mass strength feature, a chamber span feature, and a groundwater flow rate feature. The rock mass strength feature is extracted from the geological risk feature set, and the chamber span feature and the groundwater flow rate feature are extracted from the construction state feature set.

[0034] In this embodiment, the state space set of the reinforcement learning model includes multiple features. Among them, the rock mass strength feature is extracted from the geological risk feature set. For example, the uniaxial compressive strength of the rock mass is obtained by comprehensive analysis of borehole core data and in-situ stress measurement data, and the value range is between 20 - 250 MPa. In this project, the uniaxial compressive strength of the rock mass is 65 MPa. The chamber span feature is obtained from the construction design information, and the chamber span in this project is 18 m. The groundwater flow rate feature is extracted from the environmental monitoring data, and the groundwater flow rate in this project is 50 L / min. Thus, the above features and other possible features can be combined into a multi-dimensional state space set.

[0035] Step S122: Define an action space set of the reinforcement learning model, where the action space set includes a blast hole spacing parameter, a charge density parameter, and an anchor bolt spacing parameter. The blast hole spacing parameter and the charge density parameter control the dynamic excavation intensity, and the anchor bolt spacing parameter and the concrete spray layer thickness parameter control the support structure configuration.

[0036] In this embodiment, the action space set of the reinforcement learning model can include multiple parameters. For example, the value range of the blast hole spacing parameter is between 0.8 - 1.2 m, and the value range of the charge density parameter is between 0.3 - 0.6 kg / m. These two parameters jointly control the dynamic excavation intensity. The value range of the bolt spacing parameter is between 1.0 - 1.5 m, and the concrete spray layer thickness parameter is determined according to the actual situation. These two parameters control the support structure configuration, and the output action space is a multi-dimensional parameter vector used to determine the specific parameters of the construction method.

[0037] Step S123: Construct the multi-objective reward function of the reinforcement learning model. The multi-objective reward function consists of a construction efficiency reward term, a safety risk penalty term, and a cost control reward term. Among them, the construction efficiency reward term is calculated based on the single-round footage advance speed and the mechanical utilization rate, the safety risk penalty term is calculated based on the surrounding rock deformation rate and the peak stress of the support structure, and the cost control reward term is calculated based on the material consumption rate and the energy use efficiency.

[0038] In this embodiment, the multi-objective reward function of the reinforcement learning model is a function that comprehensively considers construction efficiency, safety risk, and cost control. The construction efficiency reward term is calculated according to the single-round footage advance speed and the mechanical utilization rate. For example, the faster the single-round footage advance speed and the higher the mechanical utilization rate, the greater the value of the construction efficiency reward term. The safety risk penalty term is calculated according to the surrounding rock deformation rate and the peak stress of the support structure. If the surrounding rock deformation rate exceeds a certain threshold or the peak stress of the support structure is too large, the value of the safety risk penalty term will increase. The cost control reward term is calculated according to the material consumption rate and the energy use efficiency. The lower the material consumption rate and the higher the energy use efficiency, the greater the value of the cost control reward term. By comprehensively considering the above three aspects, a reasonable multi-objective reward function is constructed to guide the reinforcement learning model to find the optimal construction method parameters.

[0039] Step S124: Use the Proximal Policy Optimization algorithm to perform policy network training on the reinforcement learning model, and input the state space set and the action space set into the policy network of the reinforcement learning model to generate a blast parameter gradient set and a support parameter gradient set.

[0040] In this embodiment, the Proximal Policy Optimization algorithm (PPO algorithm, clip_epsilon = 0.2, learning rate α = 0.0003, discount factor γ = 0.99) can be used to perform policy network training on the reinforcement learning model. Input the previously constructed state space set and action space set into the policy network of the reinforcement learning model. Through continuous learning and adjustment, a blast parameter gradient set and a support parameter gradient set are generated. The above blast parameter gradient set and support parameter gradient set reflect how to adjust the blast parameters and support parameters to obtain better rewards.

[0041] Step S125: Verify the convergence of the generated parameter gradient through the value function network of the reinforcement learning model. When the output error of the value function network is less than the preset threshold, output the preliminary construction method parameter set including the optimal blast hole spacing parameter, linear charge density parameter, and support parameter.

[0042] In this embodiment, the value function network of the reinforcement learning model is used to verify the convergence of the generated parameter gradient. By calculating the output error of the value function network, when the output error is less than the preset threshold, it indicates that the reinforcement learning model has converged to a relatively optimal solution. At this time, output the preliminary construction method parameter set including the optimal blast hole spacing parameter (such as 1.05 m in this project), linear charge density parameter (such as 0.45 kg / m in this project), and support parameter (such as bolt spacing of 1.2 m).

[0043] Step S126: When the parameter combination generated by the reinforcement learning model exceeds the execution ability of the construction machinery, trigger the collaborative optimization process between the reinforcement learning model and the preset emergency construction method library, and dynamically correct the conflicting parameters to generate an updated construction method parameter set.

[0044] In this embodiment, if the parameter combination generated by the reinforcement learning model exceeds the execution ability of the construction machinery, for example, the blast hole spacing is too small and the rock drilling jumbo cannot operate according to this parameter. At this time, trigger the collaborative optimization process between the reinforcement learning model and the preset emergency construction method library, and dynamically correct the conflicting parameters according to the preset scheme in the emergency construction method library, such as appropriately increasing the blast hole spacing, to generate an updated construction method parameter set to ensure the smooth progress of the construction.

[0045] Step S130: Input the preliminary construction method parameter set into the digital twin simulation system for mechanical response simulation processing. When the result of the mechanical response simulation processing exceeds the preset safety threshold, trigger the iterative optimization process of the construction method parameters to generate an optimized construction method parameter set.

[0046] In this embodiment, the construction efficiency indicators included in the multi-objective parameter optimization process are the single-round advance rate and the mechanical utilization rate. The single-round advance rate refers to the forward distance of the tunnel excavation in one construction cycle, which directly reflects the construction progress. In this project, by optimizing the blasting parameters and construction technology, the single-round advance rate is increased, thereby accelerating the construction progress of the entire project. The mechanical utilization rate refers to the ratio of the actual working time of the construction machinery to the total available time. Improving the mechanical utilization rate can give full play to the efficiency of the construction machinery and reduce the construction cost. For example, reasonably arranging the operation sequence and time of the rock drilling jumbo and muck truck, reducing the idle time of the equipment, and improving the mechanical utilization rate.

[0047] The safety risk indicators are the surrounding rock deformation rate and the peak stress of the support structure. The surrounding rock deformation rate reflects the deformation of the surrounding rock of the cavern during construction. When the surrounding rock deformation rate exceeds a certain threshold, it may mean that the cavern is at risk of instability. By monitoring the surrounding rock deformation rate in real time, the construction method and support measures can be adjusted in a timely manner to ensure the safety of the cavern. The peak stress of the support structure refers to the maximum stress value generated by the support structure (such as anchor bolts, shotcrete, etc.) when bearing the pressure of the surrounding rock. When the peak stress of the support structure exceeds its design strength, the support structure may be damaged, affecting the stability of the cavern. Therefore, during the multi-objective parameter optimization process, it is necessary to control the surrounding rock deformation rate and the peak stress of the support structure to ensure construction safety.

[0048] The cost control indicators are the material consumption rate and the energy use efficiency. The material consumption rate is the ratio of the actual material consumption to the theoretically calculated material consumption during construction. Reducing the material consumption rate can reduce material waste and construction costs. For example, optimizing blasting parameters can reduce the amount of explosive used, and reasonable design of the support structure can reduce the usage of anchor bolts and concrete. The energy use efficiency refers to the effective utilization degree of energy during construction. Improving the energy use efficiency can reduce the energy consumption cost. For example, using energy-saving construction equipment and optimizing the operating parameters of the equipment can improve the energy use efficiency.

[0049] The digital twin simulation system includes a surrounding rock constitutive model library, a support structure mechanical model library, and a construction machinery behavior model library. The surrounding rock constitutive model library adopts the Mohr-Coulomb criterion and the strain softening model. The Mohr-Coulomb criterion is a commonly used yield criterion for geotechnical materials. It considers the internal friction angle and cohesion of the rock and can better describe the yield and failure behavior of the rock under stress. The strain softening model further considers the characteristic that the strength of the rock decreases with the increase of strain after failure, and more accurately simulates the actual mechanical behavior of the rock. The support structure mechanical model library includes the support effect model of the bolt group and the co-deformation model of the shotcrete. The support effect model of the bolt group considers the interaction and cooperation between the anchor bolts and can accurately calculate the reinforcement effect of the bolt group on the surrounding rock. The co-deformation model of the shotcrete simulates the interaction between the shotcrete and the surrounding rock and considers the deformation and bearing capacity of the shotcrete. The construction machinery behavior model library simulates the operating behavior of construction machinery under different working conditions and provides accurate mechanical parameters for the simulation of the construction process.

[0050] In this project, the digital twin simulation system plays an important role in the optimization and verification of construction method parameters. By inputting the preliminary construction method parameter set into the digital twin simulation system, mechanical response simulation processing is carried out using the surrounding rock constitutive model library, support structure mechanical model library, and construction machinery behavior model library. For example, simulate the stress redistribution and plastic zone expansion of the surrounding rock during the blasting process, as well as the deformation and stress state of the support structure when bearing the surrounding rock pressure. Evaluate the safety and feasibility of the construction method based on the simulation results. When the simulation results exceed the preset safety threshold, trigger the iterative optimization processing of the construction method parameters to ensure that the construction method meets the project requirements.

[0051] In this embodiment, the preliminary construction method parameter set generated previously is input into the digital twin simulation system for mechanical response simulation processing to evaluate the safety and feasibility of the construction method.

[0052] Step S131: Input the blast hole spacing parameter and charge density parameter in the preliminary construction method parameter set into the blast effect simulation module of the virtual chamber three-dimensional grid model, and generate a set of data on the stress redistribution of the surrounding rock after excavation based on the initial in-situ stress field data loaded in the virtual chamber three-dimensional grid model.

[0053] In this embodiment, a geological model (FBX format, number of faces ≥ 100,000) can be imported into Unity3D, and the FLAC3D solver (mesh size 0.5m × 0.5m × 0.5m) is integrated to construct a virtual chamber three-dimensional grid model. Then, input the blast hole spacing parameter (such as 1.05m) and charge density parameter (such as 0.45kg / m) in the preliminary construction method parameter set into the blast effect simulation module of the virtual chamber three-dimensional grid model, and at the same time load the initial in-situ stress field data (such as the horizontal in-situ stress σ1 = 25MPa and the vertical in-situ stress σ3 = 8MPa). By simulating the blasting process, generate a set of data on the stress redistribution of the surrounding rock after excavation, which reflects the change of the surrounding rock stress after blasting.

[0054] Step S132: Input the set of data on the stress redistribution of the surrounding rock into the plastic zone expansion calculation module, judge the plastic deformation range of the surrounding rock according to the rock mass yield criterion, and generate a set of data on the plastic zone expansion depth and distribution pattern.

[0055] In this embodiment, the set of data on the stress redistribution of the surrounding rock generated previously can be input into the plastic zone expansion calculation module, use the Mohr-Coulomb criterion to judge whether the surrounding rock enters the plastic state, determine the plastic deformation range of the surrounding rock according to the judgment result, and generate a set of data on the plastic zone expansion depth and distribution pattern, which can visually display the change of the plastic zone of the surrounding rock after blasting.

[0056] Step S133: Input the bolt spacing parameter and the concrete shotcrete thickness parameter in the preliminary construction method parameter set into the support structure bearing capacity simulation module. Combine the plastic zone expansion depth data to calculate the deformation amount of the support structure, and generate a bolt axial force distribution data set and a shotcrete stress peak data set.

[0057] In this embodiment, the bolt spacing parameter (such as 1.2 m) and the concrete shotcrete thickness parameter in the preliminary construction method parameter set can be input into the support structure bearing capacity simulation module. Combine the plastic zone expansion depth data generated previously to calculate the deformation amount of the support structure when bearing the surrounding rock pressure, and generate a bolt axial force distribution data set and a shotcrete stress peak data set. The above bolt axial force distribution data set and shotcrete stress peak data set can evaluate the bearing capacity of the support structure.

[0058] Step S134: Input the surrounding rock stress redistribution data set, the plastic zone expansion depth data set, the bolt axial force distribution data set, and the shotcrete stress peak data set into the comprehensive safety assessment module. Perform multi-index fusion analysis through the preset deformation threshold judgment rule and the support failure criterion to generate a surrounding rock stability grade evaluation result set.

[0059] In this embodiment, the surrounding rock stress redistribution data set, the plastic zone expansion depth data set, the bolt axial force distribution data set, and the shotcrete stress peak data set can be input into the comprehensive safety assessment module. According to the preset deformation threshold judgment rule (such as when the predicted value of the crown displacement > 50 mm or the side wall convergence > 30 mm is determined to be unstable) and the support failure criterion (such as the bolt axial force exceeding the design value), perform multi-index fusion analysis to generate a surrounding rock stability grade evaluation result set. This set can judge the stability grade of the surrounding rock under the construction method.

[0060] Step S135: When the crown displacement evaluation value or the side wall convergence evaluation value in the surrounding rock stability grade evaluation result set exceeds the preset safety threshold, trigger parameter iterative optimization processing. Adjust the optimization weights of the blasting parameters and the support parameters according to the distribution characteristics of the stress exceeding the standard area, and generate an intermediate optimization parameter set including the hole distance correction parameter, the charge density correction parameter, and the support strength correction parameter.

[0061] In this embodiment, if the roof arch displacement evaluation value or the side wall convergence evaluation value in the surrounding rock stability grade evaluation result set exceeds the preset safety threshold, for example, the predicted roof arch displacement value reaches 52 mm (exceeding the threshold of 50 mm), parameter iterative optimization processing is triggered. According to the distribution characteristics of the stress over-standard area, such as stress concentration at the top of the cavern, the optimization weight of the support parameters is appropriately increased, and the optimization weight of the blasting parameters is reduced, to generate an intermediate optimization parameter set including hole spacing correction parameters (such as increasing the blasting hole spacing to 1.1 m), charge density correction parameters (such as reducing the charge density to 0.4 kg / m), and support strength correction parameters (such as reducing the bolt spacing to 1.1 m).

[0062] Step S136: Re-enter the intermediate optimization parameter set into the blasting effect simulation module and the support structure bearing capacity simulation module for secondary simulation verification. If all the indicators in the surrounding rock stability grade evaluation result set generated by the secondary simulation are lower than the preset safety threshold, mark the intermediate optimization parameter set as the optimized construction method parameter set.

[0063] In this embodiment, re-enter the intermediate optimization parameter set into the blasting effect simulation module and the support structure bearing capacity simulation module for secondary simulation verification. If all the indicators in the surrounding rock stability grade evaluation result set generated by the secondary simulation are lower than the preset safety threshold, such as the predicted roof arch displacement value is 48 mm (lower than the threshold of 50 mm) and the side wall convergence is 28 mm (lower than the threshold of 30 mm), mark the intermediate optimization parameter set as the optimized construction method parameter set, indicating that this parameter set meets the safety requirements.

[0064] Step S137: If there are still indicators exceeding the safety threshold after the secondary simulation, use the gradient descent algorithm to perform multiple rounds of iterative correction on the intermediate optimization parameter set until the safety threshold requirement is met and then output the final optimization result.

[0065] In this embodiment, if there are still indicators exceeding the safety threshold after the secondary simulation, for example, the side wall convergence is 32 mm (exceeding the threshold of 30 mm), use the gradient descent algorithm to perform multiple rounds of iterative correction on the intermediate optimization parameter set. Continuously adjust the blasting parameters and support parameters, such as further increasing the blasting hole spacing, reducing the charge density, increasing the support strength, etc., until all the indicators in the surrounding rock stability grade evaluation result set are lower than the preset safety threshold, and output the final optimization result, that is, the optimized construction method parameter set.

[0066] Step S140: Perform compliance matching processing on the optimized construction method parameter set and the preset construction specification knowledge base to generate a construction method compliance verification result. When there are non-compliant items in the construction method compliance verification result, trigger manual correction processing to generate a corrected construction method parameter set.

[0067] In this embodiment, the optimized construction method parameter set is subjected to compliance matching processing with the preset construction specification knowledge base to ensure that the construction method meets the relevant specification requirements.

[0068] Step S141: Analyze the clause constraint conditions in the preset construction specification knowledge base to generate a set of specification parameter thresholds and a set of construction process requirements.

[0069] In this embodiment, various relevant specifications can be structurally analyzed to extract key clauses, such as "the bolt spacing in Class IV surrounding rock ≤ 1.2 m", etc. The above clause constraint conditions are parsed into a set of specification parameter thresholds and a set of construction process requirements.

[0070] Step S142: Perform key parameter extraction processing on the optimized construction method parameter set to generate a blasting control parameter subset, a support strength parameter subset, and an excavation sequence parameter subset.

[0071] In this embodiment, for the optimized construction method parameter set, it is necessary to extract the parameters that are crucial for the construction compliance verification. Technically speaking, different types of construction operations correspond to different key parameters. Classifying and extracting the above parameters helps to perform targeted matching verification with the specifications in the follow-up. Specifically, when implementing, by analyzing the data structure of the optimized construction method parameter set, classification is carried out according to the meaning and function of the parameters. For example, for the parameters related to blasting control, such as the blast hole spacing and charge density, they are combined into a blasting control parameter subset; for the parameters related to the support strength, such as the bolt spacing and the thickness of the concrete spray layer, they are extracted to form a support strength parameter subset; and for the sequence of the construction process, such as the excavation sequence parameters of different tunnel sections, they constitute the excavation sequence parameter subset. Taking this project as an example, the optimized construction method parameter set contains specific blast hole spacing of 1.1 m, charge density of 0.4 kg / m, bolt spacing of 1.1 m, the thickness of the concrete spray layer, etc. After extraction, the blasting control parameter subset contains a blast hole spacing of 1.1 m and a charge density of 0.4 kg / m, and the support strength parameter subset contains a bolt spacing of 1.1 m and the corresponding thickness of the concrete spray layer and other data.

[0072] Step S143: Compare each item of the blasting control parameter subset with the blasting safety standard in the set of specification parameter thresholds to generate a set of blasting compliance inspection results.

[0073] In this embodiment, comparing the subset of blasting control parameters with the blasting safety standards in the set of specification parameter thresholds is to ensure that the blasting construction complies with safety specifications. Technically, the blasting safety standards in the specifications are formulated based on a large number of engineering practices and safety studies. By comparison, it can be found whether the actual construction parameters are within the safe range. In specific implementation, each parameter in the subset of blasting control parameters is compared with the corresponding standard in the specifications one by one. For example, the specifications stipulate that the blasting hole spacing should be between 0.8 - 1.2 m, and the charge density should be between 0.3 - 0.6 kg / m. For the 1.1 m blasting hole spacing and 0.4 kg / m charge density in the subset of blasting control parameters for this project, they are respectively compared with the specification standards. If the parameter is within the specification standard range, it is marked as compliant; if it exceeds the range, it is marked as non-compliant. Finally, the comparison results of each parameter are combined into a set of blasting compliance inspection results, which clearly reflects whether the blasting construction parameters meet the specification requirements.

[0074] Step S144: Perform strength verification processing on the subset of support strength parameters and the support design requirements in the set of specification parameter thresholds to generate a set of support compliance inspection results.

[0075] In this embodiment, performing strength verification on the subset of support strength parameters is to ensure that the support structure can effectively support the surrounding rock and guarantee construction safety. Technically speaking, the support design requirements are formulated according to the geological conditions of the surrounding rock and engineering mechanics principles. By verifying the subset of support strength parameters, it can be evaluated whether the actual bearing capacity of the support structure meets the requirements. In specific implementation, according to the support design requirements in the specifications, it may involve calculations and comparisons of aspects such as the tensile strength of bolts and the compressive strength of shotcrete layers. Taking the subset of support strength parameters for this project as an example, it includes a bolt spacing of 1.1 m and a shotcrete layer of a certain thickness. According to the specifications, for Class IV surrounding rock, the anchoring force of bolts needs to meet certain numerical requirements, and there are also corresponding standards for the thickness and strength of the shotcrete layer. By performing mechanical calculations and analyses on the above parameters, it is judged whether they meet the specification requirements. If they meet, it is marked as compliant; if not, it is marked as non-compliant, and finally a set of support compliance inspection results is formed.

[0076] Step S145: Perform topological sorting verification processing on the subset of excavation sequence parameters and the process logic rules in the set of construction process requirements to generate a set of process compliance inspection results.

[0077] In this embodiment, the topological sorting verification of the excavation sequence parameter subset is to ensure the rationality and logic of the construction process. Technically speaking, the process logic rules in the construction technology requirements set are formulated based on various factors such as construction safety, efficiency, and quality. Through topological sorting verification, it can be found whether there are conflicts or irrationalities in the excavation sequence. Specifically, when implementing, the excavation sequence parameter subset is regarded as a directed graph, where each excavation step is used as a node in the graph, and the sequence relationship between steps is used as a directed edge. Then, topological sorting is performed on this directed graph to check whether a reasonable sorting result can be obtained. For example, in this project, the excavation sequence may stipulate that the top of the chamber is excavated first, and then the side walls are excavated. If the sequence in the excavation sequence parameter subset does not conform to this rule, conflicts will be found during the topological sorting process. The result of topological sorting is judged. If it conforms to the process logic rules, it is marked as compliant; otherwise, it is marked as non-compliant, and finally, a set of process compliance inspection results is generated.

[0078] Step S146: Perform a comprehensive judgment and processing on the blasting compliance inspection result set, the support compliance inspection result set, and the process compliance inspection result set to generate the construction method compliance verification result.

[0079] In this embodiment, the comprehensive judgment of the three compliance inspection result sets is to comprehensively evaluate the compliance of the construction method. Technically speaking, the compliance of the construction method does not only depend on one aspect, but needs to comprehensively consider multiple links such as blasting, support, and process. Specifically, when implementing, all the marked results in the blasting compliance inspection result set, the support compliance inspection result set, and the process compliance inspection result set are summarized and analyzed. If all the parameters in the three sets are marked as compliant, it is determined that the construction method is compliant; as long as one parameter is marked as non-compliant, it is determined that there are non-compliant items in the construction method. For example, in this project, if all the parameters in the blasting compliance inspection result set are compliant, the bolt spacing in the support compliance inspection result set meets the requirements, but the concrete spraying layer strength does not meet the specifications, and there are conflicts in the excavation sequence in the process compliance inspection result set, then it is comprehensively determined that there are non-compliant items in the construction method, and the corresponding construction method compliance verification result is generated.

[0080] Step S147: When there are non-compliant items in the construction method compliance verification result, trigger the manual correction process to generate a set of corrected construction method parameters.

[0081] In this embodiment, when there are non-compliant items in the verification result of the construction method compliance, manual correction is required. In terms of technical principle, manual correction can combine the experience and professional knowledge of engineers to adjust the parameters that do not meet the specifications to ensure that the construction method meets the requirements. Specifically, a 3D visualization tool is developed based on WebGL, allowing engineers to perform interactive operations on this tool. For example, for the problem that the strength of the concrete spray layer in the subset of support strength parameters does not meet the specifications, the engineer can use this tool to drag and adjust the thickness of the concrete spray layer or modify parameters such as the concrete mix ratio; for the problem of conflicts in the excavation sequence in the set of process compliance inspection results, the engineer can re-plan the excavation sequence. During the adjustment process, the engineer can view the adjusted effect in real time. At the same time, the manually corrected data is added to the reinforcement learning experience pool, and the online learning rate η = 0.01 is set, and the model is incrementally updated every 24 hours so that the model can learn the experience of manual correction and improve the accuracy of generating construction method parameters in the future. After manual correction, a set of corrected construction method parameters is generated.

[0082] Step S150: Perform full-cycle risk prediction processing on the set of corrected construction method parameters to generate a construction risk distribution map, and perform dynamic regulation processing on the set of corrected construction method parameters according to the construction risk distribution map to generate a final construction method instruction set.

[0083] In this embodiment, the full-cycle risk prediction of the set of corrected construction method parameters is to identify in advance the risks that may occur during the construction process, so as to take corresponding measures for prevention. By generating a construction risk distribution map, the risk situation of different construction stages and different regions can be intuitively understood, so as to dynamically regulate the set of corrected construction method parameters to ensure the safe and efficient progress of the construction.

[0084] Step S151: Establish construction stage division rules to divide the construction period into a preparation stage, an excavation stage, a support stage, and a finishing stage, and perform risk factor identification processing on the preparation stage, the excavation stage, the support stage, and the finishing stage to generate a set of stage risk characteristics.

[0085] In this embodiment, the establishment of the construction stage division rules is based on the different characteristics and tasks of the construction process. In terms of technical principles, different construction stages face different types of risks, and dividing the construction period helps to more accurately identify risks. Specifically, when implementing, according to the construction process and time sequence, the construction period is clearly divided into a preparation stage, an excavation stage, a support stage, and a finishing stage. In the preparation stage, the risk factors may include improper equipment commissioning, insufficient material supply, etc.; the risk factors in the excavation stage may be unstable surrounding rock caused by blasting, water inrush, etc.; the risk factors in the support stage may be insecure installation of the support structure, etc.; the risk factors in the finishing stage may be incomplete cleaning work, etc. By analyzing the construction activities and environmental factors in each stage, the corresponding risk factors are identified, and the above risk factors are combined into a stage risk characteristic set. For example, in the preparation stage of this project, it is identified that abnormal vibration sensor data during equipment commissioning may lead to inaccurate acquisition of the drill bit vibration frequency, which is a risk factor; in the excavation stage, according to the geological conditions and blasting parameters, the risk factor of possible water inrush in the fault fracture zone is identified, etc.

[0086] Step S152: Perform Monte Carlo simulation processing on the stage risk characteristic set to generate a risk occurrence probability distribution data set and a risk impact degree evaluation result set for each construction stage.

[0087] The following elaborates on the specific sub-steps in detail:

[0088] Step S1521: Extract the probability distribution type parameters and distribution parameter sets of each risk factor from the stage risk characteristic set to generate risk factor random sampling configuration data

[0089] In this step, it is necessary to deeply analyze each risk factor in the stage risk characteristic set. In terms of technical principles, different risk factors often follow different probability distribution types, which are determined based on their inherent physical characteristics, historical data statistics, and relevant engineering experience.

[0090] For risk factors of geological parameter type, such as the uniaxial compressive strength of rock mass, according to a large amount of geological exploration data and the experience of previous similar projects, it usually follows a normal distribution. Through statistical analysis of the borehole core data of this project, its mean value μ = 65 MPa and standard deviation σ = 8 MPa are determined. Here, the mean value and standard deviation are the distribution parameters for this risk factor to follow a normal distribution.

[0091] For some risk factors related to construction parameters, such as the adjustment range of the blasting hole distance, since its values are random within a certain range and the probability of each value appearing is relatively uniform, it may follow a uniform distribution. Suppose the value range of the adjustment range of the blasting hole distance is between -0.1m and 0.1m, then this value range is the distribution parameter for this risk factor to follow a uniform distribution.

[0092] Sort out and summarize the probability distribution types (such as normal distribution, uniform distribution, etc.) of each risk factor and the corresponding distribution parameters (such as mean value, standard deviation, value range, etc.) to generate risk factor random sampling configuration data, which will be used as the basis for random sampling in the subsequent Monte Carlo simulation.

[0093] Step S1522: Set the sampling times parameter of the Monte Carlo simulator according to the risk factor random sampling configuration data, and perform multiple rounds of independent random sampling to generate a risk scenario simulation data set

[0094] According to the risk factor random sampling configuration data, set the sampling times of the Monte Carlo simulator. The determination of the sampling times needs to comprehensively consider the accuracy of the simulation results and the consumption of computing resources. Generally speaking, the more sampling times, the closer the simulation results are to the real situation, but the computing time and resource requirements will also increase accordingly. In this embodiment, considering the complexity of the project and the requirement for the accuracy of the results, the sampling times are set to 100,000 times.

[0095] The Monte Carlo simulator performs independent random sampling based on the probability distribution type and distribution parameters of each risk factor. Each sampling will obtain a set of value combinations of risk factors, and this value combination represents a possible risk scenario. For example, in a certain sampling, the uniaxial compressive strength of the rock mass is drawn as 68MPa, and the adjustment range of the blasting hole distance is drawn as 0.05m, etc.

[0096] Repeat such independent random sampling 100,000 times, record and sort out the risk scenario data obtained from each sampling to generate a risk scenario simulation data set, which contains a large number of different risk scenario instances, covering various possible combinations of risk factors.

[0097] Step S1523: Detect and process the risk triggering conditions for each risk scenario instance in the risk scenario simulation data set, and count the risk occurrence frequency data set for each construction stage

[0098] After obtaining the risk scenario simulation data set, it is necessary to detect the risk triggering conditions for each risk scenario instance therein. The risk triggering conditions are preset according to the actual situation of the project and relevant safety standards. For example, for the risk of water inrush, when the simulated groundwater level rises to a certain threshold (such as exceeding 2m below the bottom of the cavern), it is determined that this risk scenario triggers the risk of water inrush; for the risk of rockburst, when the stress state of the rock mass meets a specific rockburst criterion (such as the energy release rate ≥ 5×10³ J / m³), it is determined that this risk scenario triggers the risk of rockburst.

[0099] For each construction stage, all risk scenario instances in the risk scenario simulation data set are checked one by one, and the occurrence frequency of each risk is counted. For example, in 100,000 simulations in the excavation stage, the risk of water inrush was triggered 2,000 times, and the risk of rockburst was triggered 1,500 times, etc. The occurrence frequencies of different risks in each construction stage are sorted out to form the risk occurrence frequency data set for each construction stage.

[0100] Step S1524: Calculate the risk occurrence probability distribution data set for each construction stage according to the risk occurrence frequency data set

[0101] According to the risk occurrence frequency data set, calculate the occurrence probability of each risk in each construction stage. The calculation method of the risk occurrence probability is to divide the occurrence frequency of this risk by the total number of sampling times. For example, in the excavation stage, the risk of water inrush occurred 2,000 times, and the total number of sampling times was 100,000 times, then the occurrence probability of the risk of water inrush is 2,000÷100,000 = 2%.

[0102] Such calculations are performed for all risks in each construction stage to obtain the occurrence probability of each risk. The occurrence probabilities of different risks in each construction stage are sorted out and summarized to form the risk occurrence probability distribution data set for each construction stage, which shows the likelihood of different risks occurring in each construction stage.

[0103] Step S1525: Perform a quantitative calculation process on the impact degree of each risk scenario instance in the risk scenario simulation data set to generate the risk impact value sequence for each construction stage

[0104] For each risk scenario instance in the risk scenario simulation data set, it is necessary to perform a quantitative calculation on its risk impact degree. The quantification of the risk impact degree needs to comprehensively consider multiple factors, such as the impact of the risk event on construction progress, cost, quality, and safety, etc.

[0105] Taking the water inrush risk as an example, when the simulated water inrush risk occurs, based on factors such as the water volume of the water inrush, the duration, and the impact on construction equipment and personnel, quantitatively calculate the number of days of delay in the construction progress, the additional costs incurred (such as the investment in drainage equipment, the cost of repairing damaged facilities, etc.), and the potential impact on construction quality and safety. Comprehensively evaluate the above influencing factors to obtain a quantitative risk impact value.

[0106] Perform such a quantification calculation of the degree of influence for each risk scenario instance in the risk scenario simulation data set, record the risk impact value corresponding to each risk scenario instance, and form a risk impact value sequence for each construction stage.

[0107] Step S1526: Perform statistical distribution fitting processing on the risk impact value sequence to generate a risk impact degree evaluation result set for each construction stage.

[0108] After obtaining the risk impact value sequence for each construction stage, it is necessary to perform statistical distribution fitting processing on this risk impact value sequence. The purpose of statistical distribution fitting is to find a suitable probability distribution to describe the distribution of risk impact values, so as to better evaluate the degree of risk impact.

[0109] Common probability distribution types include normal distribution, lognormal distribution, Weibull distribution, etc. Through statistical analysis of the risk impact value sequence, use a suitable fitting method (such as the maximum likelihood estimation method) to determine the probability distribution type and its distribution parameters that are most suitable for this sequence. For example, after analysis, it is found that the water inrush risk impact value sequence of a certain construction stage conforms to the lognormal distribution, and its mean and standard deviation and other distribution parameters are obtained through fitting calculation.

[0110] Perform such statistical distribution fitting processing on all risk impact value sequences of each construction stage, organize and summarize the probability distribution type and distribution parameters corresponding to each risk, and form a risk impact degree evaluation result set for each construction stage, which can be used to predict the probability of different risk impact degrees occurring.

[0111] Step S153: Perform matrix fusion processing on the risk occurrence probability distribution data set and the risk impact degree evaluation result set to generate a three-dimensional risk heat map data set.

[0112] In this embodiment, matrix fusion of the risk occurrence probability distribution data set and the risk impact degree evaluation result set is to comprehensively display the probability and impact degree of risks. In terms of technical principle, through matrix fusion, two pieces of information in different dimensions can be integrated into a three-dimensional space to form a risk heat map, intuitively reflecting the risk status of different regions and different construction stages. When specifically implemented, the risk occurrence probability distribution data set and the risk impact degree evaluation result set are regarded as two matrices, and the elements of each matrix respectively represent the risk occurrence probability and impact degree of different regions or different construction stages. These two matrices are fused through a certain algorithm. For example, the risk occurrence probability is taken as one dimension, the risk impact degree is taken as another dimension, and the construction stage or region is taken as the third dimension to generate a three-dimensional risk heat map data set. In this project, an intuitive three-dimensional graph can be obtained through matrix fusion, where the depth of color indicates the level of risk, and different coordinate positions represent different construction stages and regions.

[0113] Step S154: Perform spatial interpolation processing on the three-dimensional risk heat map data set to generate a continuous risk distribution surface, and perform time-axis mapping processing on the continuous risk distribution surface and the construction progress plan data to generate the construction risk distribution atlas.

[0114] In this embodiment, spatial interpolation processing is performed on the three-dimensional risk heat map data set to obtain a continuous risk distribution surface for more accurately describing the risk distribution. In terms of technical principle, since the three-dimensional risk heat map data set is discrete, spatial interpolation can perform interpolation calculations between discrete data points to obtain a continuous risk distribution. When specifically implemented, a suitable interpolation method, such as Kriging interpolation, is used to process the three-dimensional risk heat map data set to generate a continuous risk distribution surface. Then, time-axis mapping processing is performed on the continuous risk distribution surface and the construction progress plan data to associate the risk distribution with the construction time. For example, in this project, the construction progress plan stipulates the start and end times of each construction stage. The risk information of different construction stages on the continuous risk distribution surface is mapped to the corresponding time points to generate the construction risk distribution atlas, which can clearly display the risk conditions at different times and different regions.

[0115] Step S155: Perform risk area identification processing on the construction risk distribution atlas to generate a key monitoring area coordinate set and a risk level identification set.

[0116] In this embodiment, risk area identification is performed on the construction risk distribution map to determine the areas that need to be monitored with emphasis and the corresponding risk levels. In terms of technical principles, the construction risk distribution map visually shows the distribution of risks. By analyzing the map, areas with higher risks can be identified. Specifically, when implemented, the risk values in the construction risk distribution map are compared according to a preset risk threshold. For example, the risk values are divided into three levels: low, medium, and high. The low-risk threshold is set as the risk value being less than 10%, the medium-risk threshold is 10% - 20%, and the high-risk threshold is greater than 20%. When the risk value of a certain area in the map exceeds the corresponding threshold, that area is identified as a key monitoring area. The coordinate information of the key monitoring area is recorded to generate a set of key monitoring area coordinates. At the same time, the corresponding risk level is determined based on the risk value of this area to generate a set of risk level identifiers. In this project, through the analysis of the construction risk distribution map, it is identified that the risk value of the fault fracture zone area is relatively high. Its coordinates are recorded in the set of key monitoring area coordinates and marked with the risk level.

[0117] Step S156: Adjust the construction machinery operation path planning data according to the set of key monitoring area coordinates to generate a set of avoidance path plans and a set of deceleration control parameters.

[0118] In this embodiment, adjusting the construction machinery operation path planning data according to the set of key monitoring area coordinates is to prevent the construction machinery from entering the risk area and ensure construction safety. In terms of technical principles, the operation path planning of construction machinery is based on the construction site and construction tasks. When the key monitoring area is identified, the original path planning needs to be adjusted. Specifically, when implemented, combined with the terrain of the construction site and the performance of the construction machinery, the operation path of the construction machinery is re-planned according to the set of key monitoring area coordinates. For example, for a rock drilling jumbo, if the key monitoring area is on its original planned path, then a path can be planned to bypass this area to generate a set of avoidance path plans. At the same time, to further ensure safety, the construction machinery needs to be decelerated when approaching the key monitoring area. The deceleration control parameters are determined according to the risk level of the key monitoring area and the type of construction machinery to generate a set of deceleration control parameters. In this project, when the rock drilling jumbo approaches the fault fracture zone, which is the key monitoring area, a path is planned to bypass this area, and the running speed of the rock drilling jumbo is reduced to 50% of the original speed, which is recorded in the set of deceleration control parameters.

[0119] Step S157: Adjust the real-time monitoring frequency parameters according to the set of risk level identifiers to generate a set of enhanced monitoring plans and a set of warning threshold adjustment parameters.

[0120] In this embodiment, adjusting the real-time monitoring frequency parameter according to the risk level identification set is to strengthen the monitoring of the risk area and timely detect potential risks. Technically speaking, areas with different risk levels require different monitoring frequencies, and risk areas need to be monitored more frequently. Specifically, when implementing, the real-time monitoring frequency parameter is adjusted according to the risk levels in the risk level identification set. For example, for risk areas, the monitoring frequency is increased to twice the original; for medium-risk areas, the monitoring frequency is increased to 1.5 times the original; for low-risk areas, the original monitoring frequency is maintained. At the same time, the warning threshold is adjusted according to the risk level. For risk areas, the warning threshold is lowered to detect risks earlier. The adjusted monitoring frequency and warning threshold are recorded, and an enhanced monitoring plan set and a warning threshold adjustment parameter set are generated respectively. In this project, for the risk area of the fault fracture zone, the monitoring frequency of the in-hole convergence deformation is increased from once per hour to once every half hour, and the warning threshold of the convergence deformation is lowered from 5 mm to 3 mm, which are recorded in the enhanced monitoring plan set and the warning threshold adjustment parameter set.

[0121] Step S158: Perform instruction encoding processing on the avoidance path plan set, deceleration control parameter set, enhanced monitoring plan set, and warning threshold adjustment parameter set to generate the final construction method instruction set.

[0122] In this embodiment, performing instruction encoding processing on the avoidance path plan set, deceleration control parameter set, enhanced monitoring plan set, and warning threshold adjustment parameter set is to convert the above information into instructions that can be recognized and executed by construction equipment and monitoring systems. Technically speaking, different construction equipment and monitoring systems have different instruction formats and communication protocols. Through instruction encoding, the above information can be uniformly converted into appropriate instructions. Specifically, when implementing, according to the requirements of construction equipment and monitoring systems, corresponding encoding algorithms are used to encode the information in the above sets. For example, for the avoidance path plan set, the path information is encoded into a series of coordinate points and motion instructions; for the deceleration control parameter set, the deceleration ratio and speed value are encoded into specific digital signals. The encoded instructions are combined together to generate the final construction method instruction set. In this project, through instruction encoding processing, information such as the avoidance path of the rock drilling jumbo, deceleration control parameters, in-hole convergence deformation monitoring frequency, and warning threshold is converted into instructions that can be sent to the rock drilling jumbo control system and monitoring equipment through the MQTT protocol, forming the final construction method instruction set.

[0123] Through the above steps, in this embodiment, through the collaborative drive of multi-source data fusion and intelligent algorithms, the full-chain intelligent decision-making for the generation of underground cavern construction methods is realized. First, through spatio-temporal alignment processing, discrete geological, mechanical, and environmental data are transformed into a standardized data set with spatio-temporal consistency. Then, a reinforcement learning model is used to optimize multi-objective parameters for multi-dimensional features, breaking through the limitation of empirical dependence in determining traditional construction method parameters, and realizing the dynamic adaptation of construction status and geological risks. Further, a virtual mechanical response model is constructed through a digital twin simulation system, and an iterative optimization mechanism is triggered in combination with safety thresholds, significantly improving the reliability and safety of construction method parameters. On this basis, a pre-set construction specification knowledge base is introduced for compliance verification, which not only ensures the standardization of construction method parameters but also retains the integration channel for expert experience through an artificial correction mechanism. Finally, through full-cycle risk prediction and dynamic regulation, the construction risk distribution map is transformed into an executable construction method instruction set, realizing the full-process intelligent decision-making closed-loop from data collection to construction method generation. Thus, not only the efficiency and scientificity of construction method generation are improved, but also a construction method generation system with autonomous optimization ability is constructed through the deep coupling of multi-modal data fusion and intelligent algorithms.

[0124] Figure 2 FIG. shows a schematic diagram of exemplary hardware and software components of an underground cavern construction method generation system 100 based on multi-source data fusion that can implement the inventive concept provided by some embodiments of the present invention. For example, a processor 120 can be used on the underground cavern construction method generation system 100 based on multi-source data fusion and is used to execute the functions in the present invention.

[0125] The underground cavern construction method generation system 100 based on multi-source data fusion can be a general-purpose server or a special-purpose server, both of which can be used to implement the underground cavern construction method generation method based on multi-source data fusion of the present invention. Although only one server is shown in the present invention, for convenience, the functions described in the present invention can be implemented in a distributed manner on multiple similar platforms to balance the processing load.

[0126] For example, the underground chamber construction method generation system 100 based on multi-source data fusion may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and different forms of storage media 140, such as disks, ROMs, or RAMs, or any combination thereof. Exemplarily, the underground chamber construction method generation system 100 based on multi-source data fusion may also include program instructions stored in a ROM, a RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present invention can be implemented according to the above program instructions. The underground chamber construction method generation system 100 based on multi-source data fusion further includes an input / output (I / O) interface 150 between the computer and other input / output devices.

[0127] For ease of illustration, only one processor is described in the underground chamber construction method generation system 100 based on multi-source data fusion. However, it should be noted that the underground chamber construction method generation system 100 in the present invention may also include multiple processors. Therefore, the steps performed by one processor described in the present invention may also be jointly performed or separately performed by multiple processors. For example, if the processor of the underground chamber construction method generation system 100 based on multi-source data fusion performs steps A and B, it should be understood that steps A and B may also be jointly performed by two different processors or separately performed in one processor. For example, the first processor performs step A, the second processor performs step B, or the first processor and the second processor jointly perform steps A and B.

[0128] In addition, an embodiment of the present invention further provides a readable storage medium, in which computer-executable instructions are preset. When the processor executes the computer-executable instructions, the above-mentioned underground chamber construction method generation method based on multi-source data fusion is implemented.

[0129] It should be noted that, in order to simplify the presentation of the disclosure of the present invention and thus help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of the present invention, sometimes multiple features are merged into one embodiment, drawing, or description thereof.

Claims

1. A method for generating a construction method of underground chambers based on multi-source data fusion, characterized in that, The method includes: Collecting geological exploration data, construction machinery operation data, and environmental monitoring data, performing spatio-temporal alignment processing on the geological exploration data, construction machinery operation data, and environmental monitoring data, and generating a standardized monitoring data set; Performing multi-source feature extraction on the standardized monitoring data set to obtain a construction state feature set and a geological risk feature set, inputting the construction state feature set and the geological risk feature set into a reinforcement learning model for multi-objective parameter optimization processing, and generating a preliminary construction method parameter set; Inputting the preliminary construction method parameter set into a digital twin simulation system for mechanical response simulation processing, and triggering construction method parameter iterative optimization processing to generate an optimized construction method parameter set when the result of the mechanical response simulation processing exceeds a preset safety threshold; Performing compliance matching processing on the optimized construction method parameter set and a preset construction specification knowledge base to generate a construction method compliance verification result, and triggering manual correction processing to generate a corrected construction method parameter set when there are non-compliance items in the construction method compliance verification result; Performing full-cycle risk prediction processing on the corrected construction method parameter set to generate a construction risk distribution map, and performing dynamic adjustment processing on the corrected construction method parameter set according to the construction risk distribution map to generate a final construction method instruction set.

2. The method for generating the construction method of underground caverns based on multi-source data fusion according to claim 1, wherein, The performing spatio-temporal alignment processing on the geological exploration data, construction machinery operation data, and environmental monitoring data, and generating a standardized monitoring data set includes: Establishing a conversion mapping relationship between the geological exploration coordinate system and the construction machinery coordinate system, and performing continuous geological interface generation processing on the discrete exploration points in the geological exploration data to obtain a three-dimensional geological model data set; Performing dynamic trajectory calibration processing on the position coordinate data in the construction machinery operation data to generate a construction machinery real-time positioning data set, and performing spatial superposition processing on the construction machinery real-time positioning data set and the three-dimensional geological model data set to generate a machinery-geological space association data set; Performing timestamp synchronization processing on the environmental monitoring data to generate a synchronized environmental monitoring data sequence, and performing time-axis alignment processing on the synchronized environmental monitoring data sequence and the machinery-geological space association data set to generate a spatio-temporal association monitoring data set; Performing abnormal data cleaning processing on the spatio-temporal association monitoring data set, and performing missing value repair processing on the cleaned data using a preset data integrity verification rule to generate the standardized monitoring data set.

3. The method for generating an underground chamber construction method based on multi-source data fusion according to claim 1, characterized in that, The inputting the construction state feature set and the geological risk feature set into a reinforcement learning model for multi-objective parameter optimization processing, and generating a preliminary construction method parameter set includes: Constructing a state space set of the reinforcement learning model, where the state space set includes rock mass strength characteristics, chamber span characteristics, and groundwater flow characteristics, the rock mass strength characteristics are extracted from the geological risk feature set, and the chamber span characteristics and groundwater flow characteristics are extracted from the construction state feature set; Define the action space set of the reinforcement learning model, where the action space set includes the blast hole spacing parameter, the charge density parameter, and the bolt spacing parameter. The blast hole spacing parameter and the charge density parameter control the dynamic excavation intensity, and the bolt spacing parameter and the shotcrete layer thickness parameter control the support structure configuration; Construct the multi-objective reward function of the reinforcement learning model. The multi-objective reward function consists of a construction efficiency reward term, a safety risk penalty term, and a cost control reward term. Among them, the construction efficiency reward term is calculated based on the single-round advance rate and the mechanical utilization rate, the safety risk penalty term is calculated based on the surrounding rock deformation rate and the peak stress of the support structure, and the cost control reward term is calculated based on the material consumption rate and the energy use efficiency; Use the Proximal Policy Optimization algorithm to perform policy network training on the reinforcement learning model, and input the state space set and the action space set into the policy network of the reinforcement learning model to generate a set of blasting parameter gradients and a set of support parameter gradients; Verify the convergence of the generated parameter gradients through the value function network of the reinforcement learning model. When the output error of the value function network is less than the preset threshold, output the preliminary construction method parameter set including the optimal blast hole spacing parameter, the linear charge density parameter, and the support parameter; When the parameter combination generated by the reinforcement learning model exceeds the execution ability of the construction machinery, trigger the collaborative optimization process of the reinforcement learning model and the pre-set emergency construction method library, and dynamically correct the conflicting parameters to generate an updated construction method parameter set.

4. The method for generating an underground chamber construction method based on multi-source data fusion according to claim 1, characterized in that Input the preliminary construction method parameter set into the digital twin simulation system for mechanical response simulation. When the result of the mechanical response simulation exceeds the preset safety threshold, trigger the iterative optimization process of the construction method parameters to generate an optimized construction method parameter set, including: Input the blast hole spacing parameter and the charge density parameter in the preliminary construction method parameter set into the blast effect simulation module of the virtual cavern three-dimensional grid model, and generate a set of data on the redistribution of the surrounding rock stress after excavation based on the initial in-situ stress field data loaded in the virtual cavern three-dimensional grid model; Input the set of data on the redistribution of the surrounding rock stress into the plastic zone expansion calculation module, judge the range of plastic deformation of the surrounding rock according to the rock mass yield criterion, and generate a set of data on the plastic zone expansion depth and distribution pattern; Input the bolt spacing parameter and the shotcrete layer thickness parameter in the preliminary construction method parameter set into the support structure bearing capacity simulation module, calculate the deformation of the support structure in combination with the plastic zone expansion depth data, and generate a set of data on the bolt axial force distribution and a set of data on the peak shotcrete stress; Input the set of data on the redistribution of the surrounding rock stress, the set of data on the plastic zone expansion depth, the set of data on the bolt axial force distribution, and the set of data on the peak shotcrete stress into the comprehensive safety assessment module, and perform multi-index fusion analysis through the preset deformation threshold judgment rule and the support failure criterion to generate a set of evaluation results on the stability level of the surrounding rock; When the roof displacement evaluation value or the side wall convergence evaluation value in the surrounding rock stability grade evaluation result set exceeds the preset safety threshold, trigger parameter iterative optimization processing, adjust the optimization weights of the blasting parameters and support parameters according to the distribution characteristics of the stress exceeding-standard area, and generate an intermediate optimization parameter set including hole spacing correction parameters, charge density correction parameters and support strength correction parameters; Re-input the intermediate optimization parameter set into the blasting effect simulation module and the support structure bearing capacity simulation module for secondary simulation verification. If all indicators in the surrounding rock stability grade evaluation result set generated by the secondary simulation are lower than the preset safety threshold, mark the intermediate optimization parameter set as the optimized construction method parameter set; If there are still indicators exceeding the safety threshold after the secondary simulation, use the gradient descent algorithm to perform multiple rounds of iterative correction on the intermediate optimization parameter set until the final optimization result is output after meeting the safety threshold requirements.

5. The method for generating an underground chamber construction method based on multi-source data fusion according to claim 1, characterized in that The compliance matching process of the optimized construction method parameter set and the preset construction specification knowledge base to generate the construction method compliance verification result includes: Analyze the article constraint conditions in the preset construction specification knowledge base to generate a specification parameter threshold set and a construction process requirement set; Perform key parameter extraction processing on the optimized construction method parameter set to generate a blasting control parameter subset, a support strength parameter subset and an excavation sequence parameter subset; Compare the blasting control parameter subset item by item with the blasting safety standards in the specification parameter threshold set to generate a blasting compliance inspection result set; Perform strength verification processing on the support strength parameter subset and the support design requirements in the specification parameter threshold set to generate a support compliance inspection result set; Perform topological sorting verification processing on the excavation sequence parameter subset and the process logic rules in the construction process requirement set to generate a process compliance inspection result set; Perform comprehensive judgment processing on the blasting compliance inspection result set, the support compliance inspection result set and the process compliance inspection result set to generate the construction method compliance verification result.

6. The method for generating an underground chamber construction method based on multi-source data fusion according to claim 1, wherein The full-cycle risk prediction process of the corrected construction method parameter set to generate a construction risk distribution map includes: Establish a construction stage division rule to divide the construction period into a preparation stage, an excavation stage, a support stage and a finishing stage, and perform risk factor identification processing on the preparation stage, the excavation stage, the support stage and the finishing stage to generate a stage risk characteristic set; Perform Monte Carlo simulation processing on the stage risk characteristic set to generate a risk occurrence probability distribution data set and a risk impact degree evaluation result set for each construction stage; Perform matrix fusion processing on the risk occurrence probability distribution data set and the risk impact degree evaluation result set to generate a three-dimensional risk heat map data set; Perform spatial interpolation processing on the three-dimensional risk heat map data set to generate a continuous risk distribution surface, and perform time-axis mapping processing on the continuous risk distribution surface and the construction progress plan data to generate the construction risk distribution map.

7. The method for generating an underground chamber construction method based on multi-source data fusion according to claim 6, wherein Performing Monte Carlo simulation processing on the set of stage risk characteristics to generate a set of risk occurrence probability distribution data and a set of risk impact degree evaluation results for each construction stage, including: Extracting the probability distribution type parameters and distribution parameter sets of each risk factor from the set of stage risk characteristics to generate risk factor random sampling configuration data; Setting the sampling times parameter of the Monte Carlo simulator according to the risk factor random sampling configuration data, and performing multiple rounds of independent random sampling to generate a set of risk scenario simulation data; Performing risk trigger condition detection processing on each risk scenario instance in the set of risk scenario simulation data, and statistically collecting a set of risk occurrence frequency data for each construction stage; Calculating a set of risk occurrence probability distribution data for each construction stage according to the set of risk occurrence frequency data; Performing impact degree quantification calculation processing on each risk scenario instance in the set of risk scenario simulation data to generate a sequence of risk impact values for each construction stage; Performing statistical distribution fitting processing on the sequence of risk impact values to generate a set of risk impact degree evaluation results for each construction stage.

8. The method for generating the construction method of underground chambers based on multi-source data fusion according to claim 1, characterized in that, The dynamically regulating and controlling the set of corrected construction method parameters according to the construction risk distribution map to generate a final construction method instruction set, including: Performing risk area identification processing on the construction risk distribution map to generate a set of key monitoring area coordinates and a set of risk level identifiers; Adjusting the construction machinery operation path planning data according to the set of key monitoring area coordinates to generate a set of avoidance path schemes and a set of deceleration control parameters; Adjusting the real-time monitoring frequency parameters according to the set of risk level identifiers to generate a set of enhanced monitoring schemes and a set of warning threshold adjustment parameters; Performing instruction encoding processing on the set of avoidance path schemes, the set of deceleration control parameters, the set of enhanced monitoring schemes, and the set of warning threshold adjustment parameters to generate the final construction method instruction set.

9. The method for generating an underground chamber construction method based on multi-source data fusion according to claim 1, wherein The geological exploration data includes borehole core data, geological radar scanning data, and in-situ stress measurement data. The construction machinery operation data includes the positioning data of the rock drilling jumbo, the bit vibration frequency data, and the muck truck load data. The environmental monitoring data includes the in-hole convergence deformation data, the groundwater level change data, and the air composition monitoring data; The construction efficiency indicators included in the multi-objective parameter optimization processing are the single-round advance speed and the mechanical utilization rate. The safety risk indicators are the surrounding rock deformation rate and the peak stress of the support structure. The cost control indicators are the material consumption rate and the energy use efficiency; The digital twin simulation system includes a surrounding rock constitutive model library, a support structure mechanics model library, and a construction machinery behavior model library. The surrounding rock constitutive model library adopts the Mohr-Coulomb criterion and the strain softening model. The support structure mechanics model library includes a bolt group support effect model and a concrete shotcrete co-deformation model.

10. A construction method generation system for underground chambers based on multi-source data fusion, characterized in that, Including a processor and a memory, the memory is connected to the processor. The memory is used to store programs, instructions, or codes. The processor is used to execute the programs, instructions, or codes in the memory to implement the method for generating a construction method for underground caverns based on multi-source data fusion according to any one of claims 1-9 above.

Images