Rock mass blasting optimization technology based on three-dimensional fracture network model and multi-field coupling analysis

Abstract

The invention provides a rock mass blasting optimization technology based on a three-dimensional fracture network model and multi-field coupling analysis, which comprises the following steps: inputting an initial blasting load according to a discretized rock mass structure, simulating a stress propagation process by adopting a finite element method, calculating a crack propagation path, and determining rock mass response distribution under multi-field coupling; if the vibration influence in the rock mass response distribution exceeds a preset threshold value, blasting parameters are adjusted through a neural network algorithm to optimize a crack propagation path, and an adjusted blasting parameter set is obtained; according to the corrected blasting parameter set, a multi-field coupling process is simulated, an overall effect evaluation score is calculated, and a comprehensive blasting effect index is obtained; and determining a final blasting scheme and outputting a simulation result report through matching analysis of the comprehensive blasting effect indexes and geological features.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, and particularly relates to a rock mass blasting optimization technology based on a three-dimensional fracture network model and multi-field coupling analysis. BACKGROUND

[0002] Pre-splitting blasting technology is crucial in mining, tunneling, and slope protection, directly affecting the safety, efficiency, and environmental impact of projects.

[0003] This technology precisely controls blasting energy, generating smooth fracture surfaces, reducing damage to surrounding rock, and ensuring structural stability.

[0004] However, current pre-splitting blasting methods face many challenges in practical applications, making it difficult to meet the high requirements for precision and safety in complex geological conditions.

[0005] Traditional methods often rely on empirical formulas or simple static calculations, lacking adaptability to complex geological environments, especially when faced with variable geological characteristics and dynamic blasting responses, making it difficult to achieve precise parameter adjustment and comprehensive effect evaluation.

[0006] One significant limitation of existing methods is that the design and optimization process of blasting parameters cannot fully consider the complex fracture distribution within the rock mass.

[0007] In actual engineering, the fracture network of rock mass often presents a three-dimensional irregular distribution, and traditional methods are usually based on simplified two-dimensional assumptions, ignoring the influence of fractures on blasting stress wave propagation and crack extension.

[0008] This leads to inaccurate prediction of fracture smoothness and vibration impact in blasting design.

[0009] For example, in a certain mine blasting, due to insufficient consideration of three-dimensional fractures, the fracture surface after blasting is uneven, and excessive vibration occurs in local areas, threatening the stability of the slope.

[0010] The deeper challenge lies in the fact that the blasting process involves the interaction of multiple physical fields, such as stress waves, detonation gases, and crack propagation, which differ greatly in time and spatial scales.

[0011] Stress waves propagate at extremely high speeds, while crack propagation speeds are much lower, making it difficult to coordinate simulation in a single model.

[0012] The complexity of this multi-scale coupling makes it difficult for existing technologies to accurately predict the actual impact of blasting on rock mass.

[0013] In addition, the evaluation of blasting effect often relies on a single monitoring method, such as vibration velocity measurement, lacks comprehensive analysis of geological features, simulation results and field data, and is difficult to fully reflect the safety and effect of blasting.

[0014] Therefore, how to accurately simulate the blasting process of multi-physical field coupling in a complex three-dimensional fracture network and comprehensively evaluate the blasting effect through multi-source data has become a key problem that needs to be solved in pre-splitting blasting technology. SUMMARY

[0015] The application provides a rock mass blasting optimization technology based on a three-dimensional fracture network model and multi-field coupling analysis, mainly comprising: The three-dimensional fracture network model is constructed by scanning equipment to collect rock mass geological data, and the three-dimensional fracture network model is meshed by using a finite element method to obtain a discretized rock mass structure containing fracture distribution details. The initial blasting load is input according to the discretized rock mass structure, the stress propagation process is simulated by using the finite element method, and the crack propagation path is calculated to determine the rock mass response distribution under multi-field coupling. If the vibration influence in the rock mass response distribution exceeds a preset threshold, the blasting parameters are adjusted by a neural network algorithm to optimize the crack propagation path to obtain an adjusted blasting parameter set. For the adjusted blasting parameter set, the dynamic response is re-simulated by using the finite element method, and a crack surface flatness index is extracted to determine whether the crack surface flatness index meets the geological feature requirements. If the crack surface flatness index does not meet the geological feature requirements, additional vibration influence data is obtained from the dynamic response and input into the neural network algorithm, and the parameter adjustment is further optimized to generate a corrected blasting parameter set. According to the corrected blasting parameter set, the multi-field coupling process is simulated and an overall effect evaluation score is calculated to obtain a comprehensive blasting effect index. Through matching analysis of the comprehensive blasting effect index and the geological features, a final blasting scheme is determined and a simulation result report is output.

[0016] Further, the rock mass response distribution under multi-field coupling is determined.

[0017] Further, the adjusted blasting parameter set is obtained.

[0018] Further, it is determined whether the crack surface flatness index meets the geological feature requirements.

[0019] Further, the parameter adjustment is further optimized to generate a corrected blasting parameter set.

[0020] Further, the comprehensive blasting effect index is obtained.

[0021] The technical scheme provided by the embodiment of the application can include the following beneficial effects: The application discloses a rock mass blasting optimization technology based on a three-dimensional fracture network model and multi-field coupling analysis, and aims at a business scene problem that a traditional blasting scheme is difficult to balance crack propagation accurate control and geological feature adaptation, and proposes a comprehensive solution set of data collection, model construction, parameter optimization and effect evaluation. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the application will be described below clearly and in detail. The described embodiments are only some of the embodiments of the application.

[0023] The rock mass blasting optimization technology based on a three-dimensional fracture network model and multi-field coupling analysis specifically can include the following steps. The rock mass fracture analysis and blasting control method based on rock mass geological data and blasting parameter optimization provided by the embodiments of the application aims to realize accurate analysis of rock mass structure and optimized control of blasting effect through multi-field coupling simulation and intelligent parameter adjustment. The technical solutions of the application will be described in detail in combination with specific embodiments, so that the purposes, technical solutions and advantages of the application are more clear.

[0024] S1, rock mass geological data is collected through a scanning device and a three-dimensional fracture network model is constructed, a finite element method is used to divide a grid of the three-dimensional fracture network model, and a discretized rock mass structure containing fracture distribution details is obtained. This step aims to obtain rock mass data from an actual geological environment, and through modeling and grid division technology, a complex rock mass structure is converted into a calculable discretized model, laying a foundation for subsequent stress analysis and blasting simulation. The specific implementation process includes multiple sub-steps, and the whole process from data collection to model construction is gradually completed.

[0025] S11, rock mass geological data is obtained through device scanning technology, and an original geological data set is generated. In an embodiment, a high-precision geological scanning device is used to collect rock mass geological data, For example, a laser scanner or acoustic detector is used to scan the target rock mass area comprehensively. These devices can capture the geometry of the rock mass surface and the distribution of internal cracks. During the scanning process, the device records the three-dimensional coordinate point cloud data of the rock mass surface, and preliminarily judges the depth and direction of the internal cracks through acoustic reflection or laser echo analysis. After scanning is completed, all collected data is integrated into an original geological data set containing point cloud information and preliminary crack characteristics. This data set usually contains a large number of noise points and redundant information, which needs to be further processed to improve data quality.

[0026] For example, when scanning the construction area of a mountain tunnel, a laser scanner is placed at multiple angles to scan the rock mass surface point by point, obtaining point cloud data covering the entire construction area. At the same time, the acoustic detector records the propagation time and reflection intensity of the signal inside the rock mass by emitting acoustic signals, preliminarily identifying the distribution of several main cracks. These data are integrated into an original geological data set, providing a basis for subsequent modeling. Through joint scanning of multiple devices, the coverage and accuracy of the data can be effectively improved, ensuring that the subsequent model can truly reflect the actual state of the rock mass.

[0027] S12, using a data processing algorithm to clean and format the original geological data set to obtain standardized geological data. In one possible implementation, in view of the problems of noise points, missing values and inconsistent formats existing in the original geological data set, data cleaning technology is used for preprocessing.

[0028] Specifically, first, the abnormal points in the point cloud data are removed by filtering algorithm, such as isolated points caused by device jitter or environmental interference; second, the missing data area is interpolated, and the missing part is estimated using the data of the surrounding points; finally, the data collected by different devices are converted into a unified format, such as converting point cloud data and acoustic data into a standard data structure in a three-dimensional coordinate system. The processed data is arranged as standardized geological data, which is convenient for subsequent modeling software to read and use directly.

[0029] For example, when processing the original geological data set of the above-mentioned tunnel construction area, it is found that part of the point cloud data has missing areas due to device shielding. Through the interpolation method based on neighborhood points, the geometry of the missing area is estimated, and the noise points caused by light interference are filtered out. The finally generated standardized geological data is stored in a unified three-dimensional grid point format, and each data point contains coordinate information and preliminary crack attributes. This processing method can significantly improve the usability of the data and provide reliable data support for building an accurate three-dimensional model.

[0030] S13. Based on standardized geological data, an initial three-dimensional fracture network model is constructed using three-dimensional modeling technology to determine the fracture network framework. In one embodiment, after importing standardized geological data using three-dimensional modeling software, an initial three-dimensional model of the rock mass is constructed through surface fitting and fracture identification algorithms.

[0031] Specifically, the geometric surface of the rock mass is first generated based on point cloud data. Then, the location of internal fractures reflected in the acoustic data is combined to draw a fracture distribution network. The fracture network framework consists of multiple nodes and connecting lines. Nodes represent fracture intersections, and connecting lines represent fracture extension paths. In the initial model construction process, the distribution characteristics of the main fractures are given priority, while the influence of secondary fractures is temporarily ignored to simplify computational complexity.

[0032] For example, during the modeling of the tunnel construction area, based on standardized geological data, three large-scale fractures and several small fractures penetrating the rock mass were identified. Using surface fitting technology, the rock mass surface was constructed as a continuous geometry, and the main fractures were then embedded into the model as lines, forming an initial fracture network framework. This framework can intuitively reflect the distribution pattern of fractures in the rock mass, providing a basic structure for subsequent optimization.

[0033] S14. If the node density of the fracture network framework is lower than a preset threshold, the modeling parameters are adjusted through model accuracy optimization to generate an optimized 3D fracture network model. One possible implementation involves checking the node density of the fracture network framework in the initial model, i.e., the number of fracture intersections per unit volume. If the node density is lower than the preset threshold, it indicates that the model's description of the fracture distribution is not detailed enough, which may affect the accuracy of subsequent analysis. In this case, the modeling parameters are adjusted... For example, the identification weight of secondary fractures can be increased or the sampling density of point cloud data can be improved to regenerate the model. The optimized model shows significant improvements in node density and fracture distribution details, and can more realistically reflect the complex structure of the rock mass.

[0034] For example, in the model optimization of the aforementioned tunnel construction area, it was found that the node density of the initial model was insufficient to reflect the distribution characteristics of small cracks. By increasing the sampling density of the point cloud data and readjusting the parameters of the crack identification algorithm, more small cracks were incorporated into the model. In the optimized 3D crack network model, the node density was increased by approximately two times, and the crack distribution more closely resembled the actual geological conditions. This optimization method effectively improves the model's accuracy, providing a more reliable foundation for subsequent mesh generation and stress analysis.

[0035] S15. The optimized three-dimensional fracture network model is meshed using the finite element method to obtain a discretized rock mass mesh structure. In one embodiment, the optimized three-dimensional fracture network model is imported into finite element analysis software, and the continuous rock mass model is discretized into a mesh structure composed of multiple small units through mesh generation technology.

[0036] Specifically, during mesh generation, the mesh is preferentially refined near fractures to capture stress concentration in the fracture region, while a sparser mesh is used in areas far from fractures to reduce computational cost. The resulting discretized rock mass mesh structure contains detailed information about the fracture distribution, and each mesh cell is assigned corresponding geometric and material properties.

[0037] For example, in the model of the tunnel construction area, a high-density triangular mesh was used near the main fractures in the optimized 3D fracture network model, while a larger quadrilateral mesh was used in the uniform regions inside the rock mass. The resulting discretized rock mass mesh structure contains tens of thousands of mesh elements, each of which records its location, shape, and preliminary material parameters. This meshing method can control the consumption of computational resources while ensuring computational accuracy, providing a workable model for subsequent stress simulation.

[0038] S16. By analyzing the fracture distribution characteristics, extract the fracture distribution parameters from the discretized rock mass mesh structure to generate a fracture distribution description. In one possible implementation, for the discretized rock mass mesh structure, analyze the fracture distribution characteristics, including the fracture length, width, density, and spatial distribution pattern.

[0039] Specifically, by statistically analyzing the crack-related data within the grid cells, crack distribution parameters are extracted. For example, the average spacing and main orientation of fractures. These parameters are compiled into a fracture distribution description for subsequent rock mass structure analysis. Through this analysis, a more comprehensive understanding of the fracture distribution in the rock mass can be obtained, providing data support for stability assessment.

[0040] For example, in the analysis of the aforementioned tunnel construction area, statistical analysis of the fracture distribution characteristics of the discretized rock mass grid structure revealed that major fractures are mostly distributed longitudinally along the rock mass with relatively small average spacing, while smaller fractures exhibit a random distribution. The generated fracture distribution description meticulously records these characteristics, providing crucial information for subsequent calculations of rock mass stability. This detailed analytical approach helps to reveal potential risk points in the rock mass structure.

[0041] S17. Based on the description of fracture distribution and combined with rock mass structure analysis, calculate the stability parameters of the rock mass structure and determine the characteristics of the rock mass structure. In one embodiment, based on the description of fracture distribution and combined with the geometric morphology and material properties of the rock mass, a mechanical analysis method is used to calculate the stability parameters of the rock mass structure.

[0042] Specifically, stability parameters include the overall strength of the rock mass, the local stability of fractured regions, and the possible location of slip surfaces. By comprehensively analyzing these parameters, the characteristics of the rock mass structure are determined. For example, whether it is prone to local collapse or overall instability. The final rock mass structure characteristics provide an important reference for subsequent blasting load design.

[0043] For example, in the stability analysis of the tunnel construction area, combining the description of fracture distribution and rock mass material parameters, it was found that the overall strength of the rock mass was high, but there was a risk of local instability near the main fractures. The calculated stability parameters determined that the rock mass structure was characterized by a local fracture-dominated instability mode. This analysis provided guidance for the subsequent design of the location and intensity of blasting loads, avoiding the potential risk of rock mass collapse.

[0044] S2. Based on the initial blasting load input from the discretized rock mass structure, the stress propagation process is simulated using the finite element method (FEM) to calculate the crack propagation path and determine the rock mass response distribution under multi-field coupling. This step aims to analyze the propagation law of the blasting load in the rock mass and the dynamic process of crack propagation through finite element simulation technology, and, combined with the multi-field coupling effect, comprehensively evaluate the response state of the rock mass, providing a basis for optimizing blasting parameters. The specific implementation process includes multiple sub-steps, gradually completing the entire process from load application to response distribution determination.

[0045] S21. Obtain the geometric data and material parameters of the discretized rock mass structure, construct a three-dimensional finite element mesh model, determine the application location and intensity of the initial blasting load, and obtain the discretized rock mass model. In one embodiment, geometric data, including the shape and position information of the mesh elements, is extracted from the discretized rock mass mesh structure generated in the preceding steps. Simultaneously, based on the rock mass material test results, corresponding material parameters are assigned to each mesh element. For example, the elastic modulus and Poisson's ratio. Then, the application location of the initial blasting load is determined in the model, typically in areas with dense rock fractures or key construction points, and the intensity of the blasting load is initially set according to engineering requirements. The final discretized rock mass model contains complete geometric and material information and can be directly used for finite element simulation.

[0046] For example, in the preparation for blasting simulation in the tunnel construction area, the geometric data of all grid elements were extracted from the discretized rock mass grid structure, and corresponding material parameters were set for each element based on the results of on-site sampling tests. The initial blasting load was applied near the intersection of major fractures, and the intensity was initially set to a medium level to avoid excessive damage to the surrounding rock mass. The completed discretized rock mass model can realistically reflect the mechanical properties of the rock mass, providing a reliable foundation for subsequent simulations.

[0047] S22. Apply an initial blasting load to the discretized rock mass model using the finite element method to simulate the stress propagation process, calculate the stress components of each element, and obtain stress distribution data. In one possible implementation, the initial blasting load is applied to a specified location on the discretized rock mass model, and the load propagation process in the rock mass is simulated using the finite element method.

[0048] Specifically, the blasting load acts on the model as an instantaneous impact force, triggering stress waves to propagate between grid elements. During the calculation, the stress state of each grid element is analyzed, obtaining the stress components of each element, including tensile stress, compressive stress, and shear stress. The final stress distribution data reflects the propagation law and influence range of the blasting load in the rock mass.

[0049] For example, in the simulation of the tunnel construction area mentioned above, the initial blasting load was applied to the intersection of the main fractures, and the process of stress wave propagation from the application point to the surrounding mesh elements was calculated using the finite element method. The results showed that the mesh elements near the fractures were subjected to higher shear stress, while the areas far from the fractures were mainly affected by compressive stress. The generated stress distribution data clearly demonstrated the differences in the impact of the blasting load on different areas of the rock mass, providing a direct basis for subsequent crack propagation analysis.

[0050] S23. Based on the stress distribution data and fracture mechanics criteria, calculate the critical stress value for crack propagation, determine the crack propagation direction, and obtain the crack propagation path. In one embodiment, based on the aforementioned stress distribution data, analyze the stress concentration phenomenon in the fracture region of the rock mass, and use fracture mechanics criteria to determine whether the crack will propagate.

[0051] Specifically, the stress intensity factor at the crack tip is calculated and compared with the critical value of the rock mass material. If it exceeds the critical value, the crack will propagate along the direction of maximum stress concentration. By analyzing the stress state at the crack tip point by point, the direction and path of crack propagation are determined, and finally, a complete crack propagation path is generated.

[0052] For example, in crack propagation analysis in tunnel construction areas, based on stress distribution data, it was found that the stress intensity factor at the tip of the main crack exceeded the critical value of the rock mass material, causing the crack to propagate further along the crack direction. By calculating the crack propagation direction, it was determined that the crack mainly extends along the longitudinal crack and branches at the intersection point. The generated crack propagation path provides an important reference for subsequent multi-field coupling analysis, avoiding potential threats to construction safety from crack propagation.

[0053] S24. Using a multi-field coupling model, combined with crack propagation path and stress distribution data, the coupling effect of temperature field and seepage field on crack propagation is calculated to obtain multi-field coupling response data. In one possible implementation, the influence of temperature changes and groundwater seepage on rock mass stress during blasting is considered, and a multi-field coupling model is constructed for simulation.

[0054] Specifically, the temperature field is determined by the heat generated by the blast and changes in ambient temperature, while the seepage field is calculated based on the water flow distribution and permeability within the rock fractures. Combining the aforementioned stress distribution data and crack propagation paths, the coupling effect of the temperature field and seepage field on crack propagation is analyzed. For example, increased temperature may reduce the strength of rock materials, while seepage may increase stress concentration at fracture tips. The resulting multi-field coupled response data reflects the combined influence of multiple physical fields on the rock mass response.

[0055] For example, in a multi-field coupled analysis of the tunnel construction area, the impact of the instantaneous high temperature generated during blasting on the rock mass material was simulated, revealing that the rock mass strength near the fracture decreased with increasing temperature. Simultaneously, by combining the groundwater seepage distribution within the fracture, the amplification effect of seepage on the stress at the fracture tip was calculated. The generated multi-field coupled response data showed that the crack propagation rate accelerated under the combined effects of temperature and seepage. This comprehensive analysis method can more realistically reflect the actual impact of blasting on the rock mass.

[0056] S25. If the stress value in the multi-field coupled response data exceeds a preset rock mass strength threshold, the crack propagation path is adjusted based on linear elastic fracture mechanics to obtain an updated crack propagation path. In one embodiment, the stress value of each grid cell in the multi-field coupled response data is checked. If the stress value in some areas exceeds the rock mass material strength threshold, it indicates that the current crack propagation path may lead to rock mass instability. At this time, based on the principle of linear elastic fracture mechanics, the stress distribution at the crack tip is recalculated, and the crack propagation direction is adjusted. For example, avoiding high-stress areas or changing the propagation angle. The adjusted crack propagation path can effectively reduce the risk of rock mass instability.

[0057] For example, in the analysis of the aforementioned tunnel construction area, it was found that the stress value at the fracture intersection point in the multi-field coupled response data far exceeded the rock mass strength threshold. By adjusting the crack propagation direction to avoid the high-stress area at the intersection point, a safer crack propagation path was regenerated. The updated path, while ensuring crack propagation, significantly reduced the possibility of local rock mass instability, providing a more reliable basis for subsequent blasting parameter adjustments.

[0058] S26. For the updated crack propagation path, the stress distribution of the rock mass is recalculated using the finite element method to determine the rock mass response distribution under multi-field coupling. In one possible implementation, the updated crack propagation path is re-imported into the finite element model, and the stress distribution of each mesh element in the rock mass is recalculated in conjunction with the multi-field coupling conditions.

[0059] Specifically, the stress changes as the crack propagates along the new path are simulated, while the continuous effects of the temperature and seepage fields are considered. The final determined rock mass response distribution reflects the rock mass mechanical state under the adjusted crack path, providing comprehensive data support for the evaluation of blasting effectiveness.

[0060] For example, in the recalculation of the tunnel construction area, based on the updated crack propagation path, the propagation law of stress waves during crack propagation was simulated. It was found that the stress distribution under the new path was more uniform, and the high-stress area near the crack was effectively alleviated. Combining the coupling effect of the temperature field and seepage field, the generated rock mass response distribution showed an improvement in the overall stability of the rock mass. This recalculation method ensures the safety and controllability of the blasting process.

[0061] S27. By comparing the initial stress distribution and the updated rock mass response distribution, the calculation accuracy is verified, and the final rock mass response distribution is obtained. In one embodiment, the stress distribution under the initial blasting load is compared with the updated rock mass response distribution to analyze the difference in stress values ​​in key areas. If the difference is within an acceptable range, the calculation accuracy meets the requirements, and the updated rock mass response distribution can be used as the final result; if the difference is large, the model parameters or crack propagation path need to be further adjusted until the accuracy meets the standard. The final generated rock mass response distribution provides a reliable mechanical basis for subsequent blasting parameter optimization.

[0062] For example, in the accuracy verification of the tunnel construction area, by comparing the initial stress distribution and the updated rock mass response distribution, it was found that the stress values ​​in key areas near the cracks were relatively similar, indicating that the calculation results have high reliability. The finally determined rock mass response distribution clearly reflects the comprehensive influence of blasting load and multi-field coupling effects on the rock mass, laying a solid foundation for subsequent adjustments to blasting parameters.

[0063] S3. If the vibration influence in the rock mass response distribution exceeds a preset threshold, the blasting parameters are adjusted using a neural network algorithm to optimize the crack propagation path, resulting in an adjusted set of blasting parameters. This step aims to optimize the blasting parameters using intelligent algorithms, reducing the impact of vibration on the rock mass and surrounding environment during blasting, and ensuring that the crack propagation path meets expectations. The specific implementation process includes multiple sub-steps, gradually completing the entire process from vibration data acquisition to parameter optimization.

[0064] S31. Acquire real-time monitoring data from on-site sensors to generate a rock mass vibration response dataset. In one embodiment, during blasting simulation or actual blasting tests, vibration sensors positioned at key locations within the rock mass are used to collect vibration data triggered by the blast in real time. These sensors can record information such as the frequency, amplitude, and propagation velocity of the vibration waves. The collected data is integrated into a rock mass vibration response dataset, which includes the vibration characteristics of various regions of the rock mass during the blasting process, providing a direct basis for subsequent analysis.

[0065] For example, in vibration monitoring in tunnel construction areas, multiple high-sensitivity vibration sensors were deployed on the rock surface and near fissures to record the propagation of vibration waves during blasting tests in real time. The generated rock vibration response dataset showed that the vibration amplitude near fissures was significantly higher than in other areas, reflecting the concentrated impact of blasting on the fissure region. This real-time monitoring method can comprehensively capture the vibration characteristics induced by blasting, providing important data support for parameter optimization.

[0066] S32. If the values ​​in the rock mass vibration response dataset exceed a preset threshold, vibration features are extracted using a response analysis model to obtain a vibration feature set. In one possible implementation, the vibration amplitude and frequency values ​​in the rock mass vibration response dataset are checked. If values ​​in some areas exceed the preset threshold, it indicates that the blasting vibration may have an adverse impact on rock mass stability or the surrounding environment. In this case, key features are extracted from the data using vibration response analysis technology. For example, the dominant frequency, peak amplitude, and decay rate of the vibration wave. These characteristics are compiled into a vibration feature set for subsequent intelligent analysis.

[0067] For example, in the vibration analysis of the aforementioned tunnel construction area, it was found that the vibration amplitude near the crack exceeded the preset threshold, potentially threatening construction safety. By analyzing the vibration response data, the dominant frequency and peak amplitude of the vibration wave were extracted, revealing that the vibration was mainly concentrated in the low-frequency range and decayed slowly near the crack. The generated vibration feature set provided targeted data support for subsequent adjustments to blasting parameters, avoiding potential risks caused by excessive vibration.

[0068] S33. The vibration feature set is processed by a neural network algorithm to predict the crack propagation path, resulting in a set of predicted crack paths. In one embodiment, the vibration feature set is input into a pre-trained neural network model, and the model analyzes the correlation between the vibration features and the crack propagation path to predict the possible propagation direction and range of the crack during the blasting process.

[0069] Specifically, the neural network model establishes a mapping relationship between vibration parameters and crack propagation paths by learning from historical blasting data and vibration characteristics. The resulting set of predicted crack paths contains a variety of possible propagation paths, providing diverse references for parameter optimization.

[0070] For example, in predictive analysis of tunnel construction areas, vibration feature sets are input into a neural network model to predict that cracks may propagate longitudinally along the main cracks, while branching may occur at intersection points. The generated set of predicted crack paths includes three main paths, each corresponding to a different degree of vibration influence. This prediction method can identify potential risks of crack propagation in advance, providing a scientific basis for adjusting blasting parameters.

[0071] S34. Based on the predicted crack path set, optimize the blasting parameters using a parameter adjustment strategy to obtain a preliminary blasting parameter set. In one possible implementation, based on the predicted crack path set, analyze the impact of different paths on rock mass stability and construction safety, and adjust the blasting parameters to optimize the crack propagation path.

[0072] Specifically, the blasting parameters include blasting load intensity, application location, and initiation time. By reducing the load intensity or adjusting the application location, the crack propagation direction is controlled, preventing the path from deviating from the expected direction. The final preliminary blasting parameter set contains multiple parameter combinations for subsequent verification.

[0073] For example, in the parameter optimization of the aforementioned tunnel construction area, a strategy of reducing the intensity of the blasting load and adjusting its application position was chosen for the predicted crack path set, making the crack propagation path closer to the expected direction. The generated preliminary blasting parameter set included three parameter combinations, each corresponding to a different crack propagation control effect. This optimization method can significantly reduce the adverse effects of vibration on the rock mass while ensuring construction effectiveness.

[0074] S35. The preliminary blasting parameter set is verified by the blasting control system to determine whether the parameters meet the crack path constraints, thus obtaining the verified blasting parameter set. In one embodiment, the preliminary blasting parameter set is input into the blasting control system, and the influence of each parameter combination on the crack propagation path is verified through simulation or experiment.

[0075] Specifically, the blasting control system operates based on preset crack path constraints. For example, the path must not deviate from the main fracture or cause local instability, and the validity of each parameter combination is verified. The final set of verified blasting parameters eliminates parameter combinations that do not meet the constraints and retains feasible parameter schemes.

[0076] For example, in parameter verification within a tunnel construction area, simulations of the initial blasting parameter set using the blasting control system revealed that one particular parameter combination could cause crack propagation paths to deviate from the expected direction, leading to local instability risks. By eliminating this combination, the resulting verified blasting parameter set included two feasible solutions. This verification method ensures the practicality and safety of the blasting parameters.

[0077] S36. Obtain an optimized parameter set from the verified blasting parameter set and generate blasting execution instructions. In one possible implementation, select the parameter combination that best controls the crack propagation path from the verified blasting parameter set as the optimized parameter set.

[0078] Specifically, the final parameters are determined by comprehensively considering factors such as vibration effects, crack propagation direction, and rock mass stability. Subsequently, based on the optimized parameter set, blasting execution instructions are generated, including specific blasting load intensity, application location, and detonation time, for use in actual blasting operations.

[0079] For example, in the parameter selection for the aforementioned tunnel construction area, the parameter combination with the least vibration impact and the crack propagation path closest to the expectation was chosen from the verified blasting parameter set as the optimized parameter set. The generated blasting execution command records in detail the application location and intensity of the blasting load, ensuring that the blasting operation can precisely control crack propagation. This selection method can effectively improve the controllability of the blasting process.

[0080] S37. By applying blasting execution commands through the blasting control system, the blasting operation is adjusted to obtain an optimized crack propagation path. In one embodiment, the blasting execution commands are input into the blasting control system, and the specific implementation method of the actual blasting operation is adjusted according to the parameter settings in the commands.

[0081] Specifically, the blasting control system adjusts the position and load intensity of the blasting device according to instructions to ensure that the blasting process meets expectations. The adjusted blasting operation can effectively control the crack propagation path, making it closer to the predicted optimal path, and ultimately generating an optimized crack propagation path.

[0082] For example, in blasting adjustments within a tunnel construction area, the blasting control system executes generated instructions to place the blasting device in a designated location and set the load intensity according to the instructions. The adjusted blasting operation makes the crack propagation path more consistent with expectations, avoiding excessive vibration that could affect the surrounding rock mass. The generated optimized crack propagation path provides a reliable basis for subsequent blasting effect evaluation, ensuring the safety and efficiency of the construction process.

[0083] S4. For the adjusted blasting parameter set, the dynamic response is re-simulated using the finite element method, and the crack surface smoothness index is extracted to determine whether the crack surface smoothness index meets the geological characteristics requirements. This step aims to evaluate the impact of the adjusted blasting parameters on the crack surface quality through dynamic response simulation, and further optimize the parameters according to the geological characteristics requirements to ensure that the blasting effect meets expectations. The specific implementation process includes multiple sub-steps, gradually completing the entire process from dynamic response simulation to crack surface quality assessment.

[0084] S41. The finite element method is used to simulate the dynamic response of the blasting parameter set and obtain the crack surface smoothness index data. In one embodiment, the adjusted blasting parameter set is imported into the finite element analysis software, and the dynamic response during the blasting process is re-simulated according to the blasting load intensity and application location specified in the parameters.

[0085] Specifically, the simulation focuses on the entire process of stress wave propagation, crack propagation, and crack formation induced by the blast. By analyzing the geometry of the crack surface after formation, data on crack surface smoothness indicators are extracted, including information on crack surface smoothness, roughness, and geometric deviation. These data reflect the direct impact of blasting parameters on crack surface quality, providing a basis for subsequent evaluation.

[0086] For example, in the dynamic response simulation of the tunnel construction area, based on the adjusted blasting parameter set, the entire process of crack propagation to crack surface formation under blasting load was simulated. The results showed that the crack surface formed a relatively regular geometric shape near the main cracks, but some roughness existed in local areas. The extracted crack surface smoothness index data recorded the smoothness and deviation of each area of ​​the crack surface in detail, providing comprehensive data support for subsequent quality assessment.

[0087] S42. Extract crack surface smoothness features from crack surface smoothness index data to generate crack surface quality assessment results. In one possible implementation, analyze the smoothness features of the crack surface based on the crack surface smoothness index data. For example, the overall smoothness of the crack surface, the distribution of local protrusions or depressions, etc.

[0088] Specifically, by statistically analyzing the geometric deviation values ​​of each region of the crack surface, the average value and standard deviation of the crack surface smoothness are calculated to comprehensively assess the quality status of the crack surface. The final crack surface quality assessment result presents the smoothness of the crack surface in a quantitative form, providing an intuitive basis for subsequent parameter optimization.

[0089] For example, in the assessment of the aforementioned tunnel construction area, crack surface smoothness characteristics were extracted from the crack surface smoothness index data. It was found that the overall crack surface smoothness was high, but local bulges existed near the main crack intersection points. By calculating the average smoothness and deviation distribution, the generated crack surface quality assessment results showed that the crack surface quality basically met the requirements, but some local areas still needed improvement. This assessment method can accurately identify deficiencies in crack surface quality.

[0090] S43. If the deviation between the crack surface quality assessment result and the geological feature requirements exceeds a preset threshold, the blasting parameter set is optimized and adjusted to generate an optimized blasting parameter set. In one embodiment, the crack surface quality assessment result is compared with the preset geological feature requirements. If the deviation exceeds the preset threshold, it indicates that the current blasting parameters fail to meet the expected target for crack surface smoothness. In this case, the blasting parameters are adjusted... For example, reducing the blast load intensity or changing the application location can optimize the crack formation process. The adjusted parameters are compiled into an optimized set of blast parameters for subsequent resimulation.

[0091] For example, in the optimization of parameters in the tunnel construction area, it was found that the results of the crack surface quality assessment deviated significantly from the geological characteristics required, especially the insufficient smoothness near the crack intersection points. By reducing the blasting load intensity and adjusting the application position, the resulting optimized blasting parameter set aims to reduce local protrusions and improve the overall smoothness of the crack surface. This optimization method can effectively improve the crack surface quality and meet construction requirements.

[0092] S44. For the optimized blasting parameter set, the finite element method is used again to simulate the dynamic response and obtain new crack surface smoothness index data. In one possible implementation, the optimized blasting parameter set is re-imported into the finite element analysis software to simulate the dynamic response during the blasting process, focusing on the geometric morphological changes after crack formation.

[0093] Specifically, the simulation process records the details of crack propagation path and crack surface formation, extracting new crack surface smoothness index data, including information on smoothness, roughness, and local deviations. These data reflect the improvement effect of optimized parameters on crack surface quality.

[0094] For example, in the re-simulation of the aforementioned tunnel construction area, based on the optimized set of blasting parameters, the simulation results showed that the bulging phenomenon near the fracture intersection point was significantly improved, and the overall smoothness was enhanced. The newly obtained fracture smoothness index data recorded the geometric characteristics of each region of the fracture surface in detail, providing a more accurate basis for subsequent evaluation.

[0095] S45. Based on the new crack surface smoothness index data, regenerate the crack surface quality assessment result and determine whether it meets the geological feature requirements. In one embodiment, based on the new crack surface smoothness index data, recalculate the average value and deviation distribution of crack surface smoothness to generate an updated crack surface quality assessment result. Subsequently, compare the assessment result with the geological feature requirements to determine whether the crack surface smoothness meets the expected target. If the requirements are met, proceed to the next step; if not, further parameter adjustments are required.

[0096] For example, in the reassessment of the tunnel construction area, based on the new crack surface smoothness index data, the generated crack surface quality assessment results showed that the overall smoothness of the crack surface was close to the geological characteristic requirements, with only slight deviations in some local areas. By comparing the assessment results with the requirements, it was determined that the crack surface quality basically met expectations, providing confidence for the subsequent parameter determination.

[0097] S46. If the fracture surface quality assessment results meet the geological feature requirements, then the final blasting parameter set is determined. In one possible implementation, if the deviation between the updated fracture surface quality assessment results and the geological feature requirements is within an acceptable range, it indicates that the current optimized blasting parameter set can effectively control the fracture surface quality. In this case, this parameter set is determined as the final blasting parameter set for actual blasting operations. The final parameter set contains key information such as blasting load intensity and application location to ensure that the blasting effect meets expectations.

[0098] For example, in determining the parameters for the aforementioned tunnel construction area, the crack surface quality assessment results showed that the smoothness fully met the geological requirements. Through comprehensive analysis, the optimized blasting parameter set was determined as the final scheme, ensuring that the blasting operation could generate a high-quality crack surface, improving construction efficiency and safety.

[0099] S47. By comparing the final blasting parameter set with the initial blasting parameter set, data on the parameter adjustment process is obtained, and a blasting parameter optimization record is generated. In one embodiment, the final blasting parameter set is compared with the initial blasting parameter set to analyze the changes during the parameter adjustment process. Examples include increasing or decreasing load intensity and moving the application location. The final generated blasting parameter optimization record details each step of the parameter adjustment and its basis, providing reference experience for similar subsequent projects.

[0100] For example, in the parameter records of the tunnel construction area, by comparing the final and initial blasting parameter sets, it was found that the load intensity was reduced by about one-third, and the application location was adjusted from near the fracture intersection to a safer area. The generated blasting parameter optimization records clearly demonstrate the adjustment process and effect, accumulating valuable data for parameter design in future construction.

[0101] S5. If the crack surface smoothness index does not meet the geological characteristics requirements, additional vibration impact data is obtained from the dynamic response and input into the neural network algorithm to further optimize parameter adjustments and generate a corrected blasting parameter set. This step aims to address situations where the crack surface quality still does not meet the requirements by analyzing additional vibration data and using intelligent algorithms to further optimize the blasting parameters, ensuring that the crack surface quality meets the requirements. The specific implementation process includes multiple sub-steps, gradually completing the entire process from data extraction to parameter correction.

[0102] S51. If the crack surface smoothness index does not meet the geological characteristic requirements, supplementary vibration impact data is extracted from the dynamic response data. Key vibration features are separated using the data extraction process to obtain a vibration impact dataset. In one embodiment, if the crack surface smoothness index still does not meet the geological characteristic requirements, it indicates that the current blasting parameters may have caused vibration effects unfavorable to crack formation. In this case, supplementary vibration impact data is extracted from the dynamic response simulation data, focusing on analyzing the propagation characteristics of vibration waves during blasting and their impact on crack formation.

[0103] Specifically, by separating key features such as the frequency, amplitude, and propagation path of vibration waves, a vibration impact dataset is generated to provide data support for subsequent optimization.

[0104] For example, in the extraction of vibration data in the tunnel construction area, it was found that the failure of the crack surface smoothness index to meet the standard may be related to the superposition of vibration waves caused by blasting near the crack. Supplementary vibration impact data was extracted from the dynamic response data, and the high-frequency components and propagation path characteristics of the vibration waves were separated. The generated vibration impact dataset provides a targeted basis for subsequent analysis.

[0105] S52. Based on the vibration impact dataset, a pre-set neural network algorithm is used for training, parameter adjustments are optimized, and a preliminary blasting parameter set is generated. In one possible implementation, the vibration impact dataset is input into a pre-trained neural network model, and the model analyzes the correlation between vibration characteristics and blasting parameters to predict parameter combinations that can reduce vibration impact.

[0106] Specifically, the neural network model learns from historical data to establish a mapping relationship between vibration characteristics and blasting parameters, generating a preliminary set of blasting parameters, which includes information such as optimized load strength and application location.

[0107] For example, in the parameter optimization of the tunnel construction area mentioned above, the vibration impact dataset was input into a neural network model. The model predicted that reducing the blasting load intensity and adjusting the application position could effectively reduce the impact of vibration on crack formation. The generated preliminary blasting parameter set contained a variety of parameter combinations, providing diverse options for subsequent adjustments.

[0108] S53. If the deviation between the initial blasting parameter set and the geological feature requirements exceeds a preset threshold, the neural network weights are adjusted using an iterative optimization algorithm to obtain a corrected blasting parameter set. In one embodiment, the simulation results generated from the initial blasting parameter set are compared with the geological feature requirements. If the deviation still exceeds the preset threshold, it indicates that the current parameter combination still needs improvement. At this time, the weights of the neural network model are adjusted using iterative optimization techniques, the parameter combination is recalculated, and a corrected blasting parameter set is generated. The corrected parameter set is more accurate in terms of vibration control and crack surface quality improvement.

[0109] For example, in the parameter correction process in the tunnel construction area, it was found that the surface smoothness generated by the initial blasting parameter set did not fully meet the geological requirements. By iteratively adjusting the weights of the neural network model, the generated corrected blasting parameter set further reduced the blasting load intensity, ensuring that the vibration impact was minimized. This correction method can gradually approach the optimal parameter combination.

[0110] S54. Based on the revised blasting parameter set, geological data analysis techniques are used to verify the degree of matching between the parameter set and the geological characteristic requirements, and to determine the optimized blasting parameters. In one possible implementation, based on the revised blasting parameter set and combined with rock mass geological data, the parameter combination is analyzed to determine whether it meets the geological characteristic requirements.

[0111] Specifically, the impact of parameters on crack surface smoothness is verified through simulation or experimentation, and the degree of matching between the parameters and geological feature requirements is calculated. If the degree of matching is high, the optimized blasting parameters are determined and used for subsequent operations.

[0112] For example, in the parameter verification of the aforementioned tunnel construction area, based on the modified blasting parameter set and combined with the rock mass fracture distribution characteristics, the improvement effect of the parameters on the smoothness of the fracture surface was verified. The results showed that the parameters matched the geological characteristics well, and were thus determined as the optimized blasting parameters. This verification method ensured the effectiveness of the parameter adjustment.

[0113] S55. If the optimized blasting parameters still do not meet the geological requirements, new vibration impact data is obtained from the dynamic response data, input into the neural network algorithm, and the optimization process is repeated to obtain updated blasting parameters. In one embodiment, if the optimized blasting parameters still do not meet the geological requirements, it indicates that the vibration impact or parameter settings are still insufficient. At this time, new vibration impact data is extracted from the dynamic response data, re-input into the neural network model for optimization, and updated blasting parameters are generated. Through multiple iterations, it is ensured that the parameters gradually approach the optimal state.

[0114] For example, during the further optimization of the tunnel construction area, it was found that the optimized blasting parameters still did not achieve the expected results in local fracture areas. New vibration impact data were extracted from the dynamic response data and re-input into the neural network model. The generated updated blasting parameters were further adjusted in terms of application location and load intensity to ensure a gradual improvement in the quality of the fracture surface.

[0115] S56. Using vibration data processing technology, analyze the impact of updated blasting parameters on the crack surface smoothness index to determine whether it meets the geological characteristics requirements. In one possible implementation, based on the updated blasting parameters and combined with vibration data processing technology, analyze the impact of parameter adjustment on the crack surface smoothness index.

[0116] Specifically, by simulating the blasting process, the changes in the smoothness of the crack surface are calculated to determine whether it meets the geological requirements. If the requirements are met, the process proceeds to the next step; if not, further optimization is required.

[0117] For example, in the impact analysis of the tunnel construction area mentioned above, based on the updated blasting parameters, the influence of vibration on the crack surface smoothness index was analyzed. It was found that the overall smoothness of the crack surface has been significantly improved, basically meeting the requirements of geological characteristics. This analysis method can intuitively reflect the effect of parameter adjustment and provide a basis for the final parameter determination.

[0118] S57. Based on the judgment result, data fusion technology is used to combine the updated blasting parameters with the geological data analysis results to generate the final blasting parameter set. In one embodiment, if the impact of the updated blasting parameters on the crack surface smoothness index meets the geological characteristic requirements, the parameters and geological characteristics are further fused together with the rock mass geological data analysis results to generate the final blasting parameter set.

[0119] Specifically, the fusion process takes into account factors such as fracture distribution, rock mass strength, and vibration effects to ensure that the parameter set is applicable under various conditions.

[0120] For example, in the parameter fusion of the tunnel construction area, the updated blasting parameters are combined with the geological data analysis results, taking into account the distribution of cracks and the influence of vibration. The resulting final blasting parameter set can effectively control the quality of the crack surface and ensure the safety and efficiency of the blasting operation.

[0121] S6. Based on the modified blasting parameter set, simulate a multi-field coupling process and calculate the overall effect evaluation score to obtain a comprehensive blasting effect index. This step aims to comprehensively evaluate the impact of the modified blasting parameters on the rock mass response through multi-field coupling simulation and generate a comprehensive blasting effect index, providing a basis for the final scheme determination. The specific implementation process includes multiple sub-steps, gradually completing the entire process from parameter processing to effect evaluation.

[0122] S61. Obtain the corrected set of blasting parameters, and format it using a preset data processing procedure to obtain a standardized parameter set. In one embodiment, the corrected set of blasting parameters is obtained from the aforementioned steps, and the parameters are converted into a standard format through a data processing procedure. For example, it standardizes the units of load intensity and the coordinate system of the application location. The resulting standardized parameter set is easy for subsequent simulation software to read and use, ensuring the accuracy of the simulation process.

[0123] For example, in parameter processing in tunnel construction areas, the load intensity and application location information in the modified blasting parameter set are converted into a unified format. The generated standardized parameter set contains all the necessary information, which facilitates multi-field coupled simulation.

[0124] S62. By standardizing the parameter set, perform multiple coupled process simulations and use the finite element analysis algorithm to obtain the simulation result dataset. In one possible implementation, based on the standardized parameter set, perform multiple multi-field coupled process simulations in the finite element analysis software, comprehensively considering the interaction of the stress field, temperature field, and seepage field.

[0125] Specifically, each simulation records the changes in rock mass response under blasting loads, including stress distribution, crack propagation, and crack formation. The final simulation dataset contains detailed data from multiple simulations, providing a comprehensive basis for performance evaluation.

[0126] For example, in the coupled simulation of the aforementioned tunnel construction area, multiple simulations were performed based on a standardized parameter set, comprehensively analyzing the impact of stress changes and temperature fields induced by blasting loads on the rock mass material. The generated simulation result dataset records the rock mass response characteristics of each simulation in detail, providing diverse data for subsequent evaluation.

[0127] S63. If the stability of the simulation result dataset meets a preset threshold, then based on the simulation result dataset, calculate the effect evaluation score for each coupling process to obtain a score sequence. In one embodiment, the stability of the simulation result dataset is checked. For example, whether the stress distribution and crack propagation path are consistent across multiple simulations. If the stability meets a preset threshold, the simulation results are considered reliable. At this point, an effectiveness evaluation score for each coupling process is calculated based on the dataset. The score comprehensively considers factors such as crack surface quality, vibration influence, and rock mass stability, ultimately generating a score sequence.

[0128] For example, in the effect evaluation of the tunnel construction area, it was found that the simulation result dataset had high stability, and the crack propagation paths of multiple simulations were basically consistent. By comprehensively analyzing the crack surface smoothness and vibration effects, the effect evaluation scores of each simulation were calculated, and the generated score sequence provided a quantitative basis for subsequent parameter adjustments.

[0129] S64. Based on the score sequence, a clustering analysis algorithm is used to determine the parameter correlation relationships in each coupling process and obtain the parameter influence weights. In one possible implementation, based on the score sequence, the correlation between the blasting parameters and the effect evaluation scores in each simulation is analyzed to determine the influence weights of the parameters on the effect.

[0130] Specifically, cluster analysis is used to group parameter combinations according to their effect scores, extracting the degree of influence of key parameters. The resulting parameter influence weights reflect the contribution of different parameters to the blasting effect.

[0131] For example, in the correlation analysis of the aforementioned tunnel construction area, based on the fractional sequence, it was found that the blasting load intensity had the highest influence weight on the crack surface smoothness, while the application location had a more significant control effect on the vibration. The generated parameter influence weights provide scientific guidance for subsequent parameter adjustments.

[0132] S65. By adjusting the parameter influence weights, the key parameters in the blasting parameter set are adjusted to generate an optimized parameter set. In one embodiment, based on the parameter influence weights, the key parameters that have the greatest impact on the blasting effect are adjusted first. For example, the load intensity could be reduced or the application location optimized. The adjusted parameters were compiled into an optimized parameter set, aiming to further improve the blasting effect and reduce adverse effects.

[0133] For example, in the parameter adjustment of the tunnel construction area, the blasting load intensity was reduced first according to the parameter influence weight, and the application position was fine-tuned. The resulting optimized parameter set can more effectively control the crack surface quality and vibration influence.

[0134] S66. Using the optimized parameter set, re-execute the coupled process simulation to obtain an updated simulation result dataset. In one possible implementation, the optimized parameter set is re-imported into the finite element analysis software, a multi-field coupled process simulation is performed, and the changes in rock mass response under blasting loads are recorded. The final updated simulation result dataset reflects the blasting effect after parameter optimization, providing the latest data for comprehensive evaluation.

[0135] For example, in the re-simulation of the aforementioned tunnel construction area, based on the optimized parameter set, the simulation results show that the surface smoothness is further improved and the vibration impact is significantly reduced. The generated updated simulation result dataset provides a reliable basis for the final effect evaluation.

[0136] S67. Based on the updated simulation result dataset, calculate the comprehensive blasting effect index to obtain the final index value. In one embodiment, based on the updated simulation result dataset, comprehensively calculate various indicators of the blasting effect, including crack surface quality, vibration influence, and rock mass stability, and finally generate a comprehensive blasting effect index value. This index value reflects the overall effect of the blasting parameters in a quantitative form, providing a basis for determining the final scheme.

[0137] For example, in the comprehensive assessment of the tunnel construction area, the combined scores of crack surface smoothness and vibration impact were calculated based on the updated simulation results dataset. The generated final index values ​​showed that the blasting effect had reached a high level, laying the foundation for subsequent scheme selection.

[0138] S7. Through the matching analysis of the comprehensive blasting effect indicators and geological characteristics, determine the final blasting scheme and output a simulation result report. This step aims to determine the optimal blasting scheme by assessing the degree of matching between the comprehensive blasting effect indicators and geological characteristics, and to generate a detailed simulation result report to provide guidance for actual construction. The specific implementation process includes multiple sub-steps, gradually completing the entire process from matching analysis to scheme determination.

[0139] S71. A description of geological conditions is obtained by extracting key features from geological feature data. In one embodiment, key features are extracted from rock mass geological data. Information such as fracture distribution density, rock mass strength, and permeability is used to generate a geological condition description. This description, in a quantitative form, reflects the actual state of the rock mass and provides a basis for subsequent matching analysis.

[0140] For example, in the geological analysis of the tunnel construction area, the distribution density of major fractures and the overall strength information of the rock mass were extracted from the geological data. The generated geological condition description recorded the mechanical properties of the rock mass in detail, providing data support for matching blasting schemes.

[0141] S72. Principal component analysis is used to match the geological condition description with the comprehensive blasting index to determine the matching coefficient. In one possible implementation, based on the geological condition description and the comprehensive blasting effect index, principal component analysis is used to extract the main characteristics of both and calculate the matching coefficient.

[0142] Specifically, the matching coefficient reflects the degree of fit between the blasting effect and the geological conditions. The higher the coefficient, the more suitable the blasting scheme is for the current geological environment.

[0143] For example, in the matching analysis of the tunnel construction area mentioned above, principal component analysis technology was used to find that the comprehensive blasting effect index was highly consistent with the geological conditions in terms of fracture control and stability. The calculated matching coefficient was high, indicating that the current blasting scheme was more applicable.

[0144] S73. If the matching coefficient exceeds a preset threshold, a blasting scheme is selected based on the matching result to obtain a preliminary blasting scheme. In one embodiment, if the matching coefficient exceeds the preset threshold, it indicates that the current blasting parameters are highly compatible with the geological conditions. In this case, the corresponding combination of blasting parameters is selected based on the matching result to generate a preliminary blasting scheme. This scheme includes specific blasting operation guidance information.

[0145] For example, in the selection of blasting schemes for tunnel construction areas, if the matching coefficient exceeds a preset threshold, it indicates that the blasting effect matches the geological conditions well. Based on the matching results, the corresponding parameter combination was selected, and the generated preliminary blasting scheme provided initial guidance for actual construction.

[0146] S74. Optimized blasting parameters are generated by adjusting the parameters in the preliminary blasting plan. In one possible implementation, the parameters are fine-tuned for the preliminary blasting plan, taking into account actual construction conditions and safety requirements. For example, adjusting the intensity or location of the blasting load can ensure the plan is more feasible in practice. The resulting optimized blasting parameters provide more precise guidance for construction.

[0147] For example, in the parameter fine-tuning of the aforementioned tunnel construction area, the blasting load strength was fine-tuned based on the initial blasting plan and the safety distance requirements of the construction site. The resulting optimized blasting parameters ensured the safety and effectiveness of the blasting operation.

[0148] S75. Numerical simulation is performed using the optimized blasting parameters to obtain blasting effect simulation data. In one embodiment, based on the optimized blasting parameters, numerical simulation is performed in finite element analysis software to record information such as stress distribution, crack propagation, and crack formation during the blasting process. The final generated blasting effect simulation data reflects the actual effect of the optimized parameters.

[0149] For example, in the numerical simulation of the tunnel construction area, based on the optimized blasting parameters, the simulation results showed that the crack surface smoothness was high and the vibration impact was controlled within a safe range. The generated blasting effect simulation data provided a basis for the final scheme verification.

[0150] S76. Use a support vector machine (SVM) algorithm to classify the simulated blasting effect data and determine whether the effect meets expectations. In one possible implementation, the simulated blasting effect data is input into the SVM model, and the model classifies the data to determine whether the blasting effect meets the expected target.

[0151] Specifically, the model categorizes simulation results into two types based on preset classification criteria: those that meet expectations and those that do not, providing a reference for determining the final solution.

[0152] For example, in the effect classification of the aforementioned tunnel construction area, analysis of the blasting effect simulation data using a support vector machine model revealed that the simulation results basically met the expected goals, with crack surface quality and vibration control both within acceptable ranges. This classification method can quickly determine the quality of the blasting effect.

[0153] S77. The accuracy of the simulation results is verified by comparing the deviations between the simulated data and the actual geological features. In one embodiment, the simulated data of the blasting effect is compared with the actual geological feature data to analyze the deviations in terms of fracture distribution, fracture surface smoothness, and stability. If the deviations are within an acceptable range, the simulation results are considered accurate and reliable, and can be used as the basis for the final scheme.

[0154] For example, in verifying the accuracy of the simulation in the tunnel construction area, by comparing the simulated blasting effect data with the actual geological features, it was found that the deviations between the two in terms of fracture distribution and fracture surface smoothness were small, indicating that the simulation results have high credibility. This verification method provides confidence for the implementation of the final blasting plan.

Claims

1. A rock mass blasting optimization technique based on a three-dimensional fracture network model and multi-field coupled analysis, characterized in that, include: Rock mass geological data is collected using scanning equipment to generate raw geological datasets; The original geological dataset is cleaned and formatted to obtain standardized geological data; Based on the standardized geological data, an initial three-dimensional fracture network model was constructed, and the fracture network framework was determined. When the node density of the fracture network framework is lower than the set standard, the modeling parameters are adjusted to generate an optimized three-dimensional fracture network model. The optimized three-dimensional fracture network model is meshed to obtain a discretized rock mass mesh structure; By analyzing the fracture distribution characteristics in the discretized rock mass grid structure, fracture distribution parameters are extracted to generate a fracture distribution description. Based on the description of the fracture distribution and the analysis of the rock mass structure, stability parameters are calculated and the characteristics of the rock mass structure are determined. Obtain the geometric data and material parameters of the discretized rock mass mesh structure, construct a three-dimensional finite element mesh model, determine the application location and intensity of the initial blasting load, and obtain the discretized rock mass model; The initial blasting load is applied to the discretized rock mass model to simulate the stress propagation process, calculate the stress components of each unit, and obtain stress distribution data. Based on the stress distribution data, the crack propagation direction is calculated, and the crack propagation path is obtained; The multi-field coupling response data is calculated by combining the crack propagation path and the stress distribution data using a multi-field coupling model. When the stress value in the multi-field coupling response data exceeds the set range, the crack propagation path is adjusted to obtain an updated crack propagation path; For the updated crack propagation path, the rock mass stress distribution is recalculated to determine the rock mass response distribution under multi-field coupling.

2. The rock blasting optimization technology according to claim 1, characterized in that, When the stress value in the multi-field coupled response data exceeds a set range, adjusting the crack propagation path to obtain an updated crack propagation path includes: Key stress distribution features are extracted from the multi-field coupled response data to generate a stress feature set; The stress feature set is analyzed to determine the stress concentration region; Based on the stress concentration region, the crack propagation direction is adjusted to generate an intermediate crack path; The stability of the intermediate crack path is verified, and verification results are generated. Based on the verification results, the intermediate crack path is optimized to obtain the updated crack propagation path. By using the updated crack propagation path, the stress distribution data of each element is recalculated to generate an updated stress distribution dataset.

3. The rock blasting optimization technology according to claim 1, characterized in that, The process involves analyzing the fracture distribution characteristics in the discretized rock mass mesh structure, extracting fracture distribution parameters, and generating a fracture distribution description, including: The discretized rock mass grid structure is layered to generate multiple grid levels; For each of the aforementioned grid levels, the fracture distribution density and orientation features are extracted to generate hierarchical fracture data; The hierarchical fracture data are integrated to generate an overall fracture distribution feature set; Based on the overall fracture distribution feature set, calculate the fracture distribution parameters and generate the fracture distribution description; Based on the description of the fracture distribution, the influence weight of the fracture distribution on the rock mass structure is determined, and influence weight data is generated.

4. The rock blasting optimization technology according to claim 2, characterized in that, The stability verification of the intermediate crack path and the generation of verification results include: The intermediate crack path is mapped to a mesh element to generate a path element set; For the aforementioned path element set, the stress variation data of each element is calculated to generate stability assessment data; The verification results are generated based on the stability assessment data.

5. The rock blasting optimization technology according to claim 3, characterized in that, The process of integrating the hierarchical fracture data to generate an overall fracture distribution feature set includes: The hierarchical fracture data is standardized to generate data in a unified format; The unified format data is fused to generate the overall fracture distribution feature set.

6. The rock blasting optimization technology according to claim 4, characterized in that, The step of calculating stress variation data for each element in the path element set and generating stability assessment data includes: For each element in the path element set, extract stress distribution features to generate element stress data; The stability assessment data is generated by comprehensively analyzing the unit stress data.

7. The rock blasting optimization technology according to claim 5, characterized in that, The step of performing feature fusion on the unified format data to generate the overall fracture distribution feature set includes: The unified format data is subjected to feature classification to generate a classification feature set; The classification feature set is weighted to generate the overall crack distribution feature set.