Data-driven tunnel blasting parameter optimization method and system

Abstract

This invention relates to the field of tunnel blasting engineering management and data processing technology, specifically to a data-driven method and system for optimizing tunnel blasting parameters. The method includes: acquiring the functional region to which the blast hole belongs, hole layout parameters, benchmark explosive consumption, and benchmark rock mass resistance parameters; collecting drilling operation data from the rock drilling equipment to determine the dynamic rock mass strength index of the blast hole, thereby identifying strength anomaly regions in the tunnel face and generating weakened topological boundaries enveloping these regions; combining the spatial position characteristics of the blast hole relative to the weakened topological boundary and the dynamic rock mass strength index to determine the local strength mutation coefficient of each blast hole; using this coefficient to correct the benchmark explosive consumption, and combining it with spatial volume parameters to calculate the dynamic charge amount of the blast hole. This method can respond to local heterogeneous mutations in the rock mass at the tunnel face, improving the accuracy of matching the charge energy with the actual rock-breaking resistance.

Description

Technical Field

[0001] This invention relates to the field of tunnel blasting engineering management and data processing technology, specifically to a data-driven method and system for optimizing tunnel blasting parameters. Background Technology

[0002] In hard rock tunnel and underground space excavation operations, the drill-and-blast method is the core construction method. The design accuracy of blasting parameters directly affects the control of over-excavation and under-excavation of the tunnel, construction costs, and the stability of the surrounding rock.

[0003] Currently, the industry has conducted relevant research on the optimization design of tunnel drilling and blasting parameters. For example, the existing Chinese patent application document with publication number CN117973007A discloses a method for controlling tunnel drilling and blasting parameters. This method obtains the drilling rig's parameters while drilling to invert the surrounding rock grade and rock mechanics parameters, and then matches the corresponding blasting and support design parameters.

[0004] However, deeply buried underground rock masses are generally heterogeneous due to geological tectonic movements. In actual tunnel face construction, even if macro-geological exploration classifies them as the same surrounding rock level, they often contain localized dense joint zones, concealed faults, or soft interlayers. Current technology relies on the assumption of homogeneous rock masses to conduct a comprehensive assessment of mechanical properties, using this as the basis for matching blasting parameters. Under this model, parameter design is often uniformly configured on a large functional area basis, making it difficult to reflect the heterogeneous characteristics of local rock masses and lacking effective means of identifying the spatial boundaries of internal geological anomalies.

[0005] Due to geological heterogeneity, abrupt transitions between hard and soft rock interfaces often form within the tunnel face. During blasting dynamics, the intact hard rock on the periphery exerts a spatial constraint on the blasting and fracturing of the soft rock layers inside. This boundary effect means that the actual rock-breaking resistance of a single borehole is not only related to the physical and mechanical properties of the rock at that location but is also affected by the heterogeneous interface. This can easily lead to a mismatch between the charge amount and the actual rock-breaking resistance. Boreholes adjacent to the hard-soft interface often experience local under-excavation due to the stronger constraint effect, while boreholes located inside the soft interlayer are prone to over-excavation due to relatively excessive charge.

[0006] Therefore, there is an urgent need for a technical solution that can adapt to the heterogeneous characteristics of rock masses, effectively identify local geological anomaly boundaries, and comprehensively consider the rock mass material strength and boundary confinement effect to achieve dynamic adaptive optimization of single-hole blasting parameters. Summary of the Invention

[0007] To address the problem that existing technologies cannot accurately identify local rock mass strength anomalies and their topological boundaries during blasting design, and fail to combine the spatial location characteristics of the borehole relative to the anomaly boundary to make targeted corrections to the charge amount, resulting in a mismatch between local charge energy and actual rock-breaking resistance, this invention proposes a data-driven tunnel blasting parameter optimization method and system.

[0008] On the one hand, the data-driven tunnel blasting parameter optimization method provided by this invention includes: Obtain the functional area and hole layout parameters of each blast hole in the tunnel face, as well as the benchmark explosive consumption per unit area and the benchmark rock mass resistance parameters corresponding to the tunnel face; The operation data of the rock drilling equipment during the drilling of each blast hole is collected to determine the dynamic rock mass strength index of each blast hole. Based on the size distribution characteristics of the dynamic rock mass strength index of each blast hole relative to the rock mass resistance benchmark parameter, the strength anomaly area in the tunnel face is identified, and the weakened topological boundary enveloping the strength anomaly area is generated by combining the change gradient of the dynamic rock mass strength index. By combining the spatial location characteristics of each borehole relative to the weakened topological boundary and the dynamic rock mass strength index of each borehole, the local strength mutation coefficient of each borehole is determined. The reference explosive consumption of each functional area is corrected by using the local intensity mutation coefficient of each borehole, and the dynamic charge of each borehole is obtained by combining the spatial volume parameters obtained by analyzing the borehole layout parameters of each borehole, so as to achieve adaptive optimization of blasting parameters.

[0009] This scheme establishes a dynamic optimization system encompassing drilling data perception, anomaly area spatial topology identification, and adaptive control of blasting parameters. During tunnel excavation, by extracting drilling operation data and analyzing rock mass strength, hidden soft layers or fractured zones within the tunnel face can be identified in situ, generating weakened topological boundaries with clear geometric significance. Based on this, the calculation of borehole charge is integrated with the spatial characteristics of these weakened topological boundaries, considering not only the material variations of the rock mass itself but also identifying the complex confinement environments faced by boreholes at different locations inside and outside the anomaly boundary. The generated local strength mutation coefficient can precisely drive the charge model to reduce energy input in soft areas to curb over-excavation or enhance energy in highly constrained areas to eliminate rock embankments, achieving precise balancing of explosive energy and actual rock fracture energy consumption from both data source and spatial geometry dimensions.

[0010] Furthermore, the dynamic rock mass strength index of each borehole is determined based on the following method: the rock mass resistance benchmark parameters include the benchmark mechanical specific energy and the benchmark value of the Protodyakonov hardness coefficient of the surrounding rock; the actual mechanical specific energy is obtained by analyzing the operating data during the drilling of each borehole; the benchmark value of the Protodyakonov hardness coefficient of the surrounding rock is mapped and transformed using the relative ratio of the actual mechanical specific energy and the benchmark mechanical specific energy to obtain the dynamic Protodyakonov hardness coefficient as a dynamic rock mass strength index.

[0011] Furthermore, the actual mechanical energy is obtained by analyzing the operating data during the drilling of each borehole, including: the operating data during the drilling of each borehole includes the rock drilling equipment's propulsion pressure, rotation speed, rotational torque, and axial drilling speed; the axial linear extrusion power consumption generated by the propulsion pressure and the circumferential rotational shear power consumption generated by the rotation speed and rotational torque are extracted; based on the principle of energy conservation, the axial linear extrusion power consumption and the circumferential rotational shear power consumption are aggregated to obtain the actual mechanical energy.

[0012] This technical solution is based on the principle of conservation of rock-breaking energy. It reconstructs the physical parameters of rock drilling equipment during the drilling process into axial and circumferential rock-breaking work parameters with clear mechanical dimensions. By proportionally calibrating the benchmark parameters through actual mechanical specific energy, it realizes the cross-domain conversion between mechanical cutting energy efficiency and rock dynamic strength. While retaining the benchmark evaluation role of the classic Protodyakonov coefficient, it dynamically restores it into an index with hole-level resolution.

[0013] Furthermore, the process involves identifying areas of strength anomaly in the tunnel face and generating a weakened topological boundary enclosing these areas by combining the gradient of dynamic rock mass strength indices. This includes: extracting the coordinates of boreholes whose dynamic rock mass strength indices are lower than the Protodyakonov robustness coefficient benchmark value in the rock mass resistance benchmark parameters, and performing spatial clustering to generate multiple continuously distributed clusters of weakened nodes, defining the spatial range corresponding to each cluster of weakened nodes as the strength anomaly area; statistically analyzing the gradient of dynamic rock mass strength indices of adjacent boreholes in the tunnel face and constructing a dynamic identification threshold to characterize abrupt changes in geological structure; extracting the coordinates of boreholes whose gradients are greater than the dynamic identification threshold, and selecting the outermost coordinate points enclosing each cluster of weakened nodes from the extracted borehole coordinates, performing geometric connection closure processing to obtain the weakened topological boundary enclosing the strength anomaly area.

[0014] This technical solution introduces spatial variation gradient analysis and node clustering technology, which can automatically separate continuous spatial regions with abrupt changes in geological properties from discrete pore-level intensity data and transform them into closed topological boundaries that the system can recognize, providing an accurate foundation for subsequent spatial geometric constraint calculations under complex geological conditions.

[0015] Furthermore, the local strength mutation coefficient of each borehole is determined as follows: the spatial position of each borehole relative to the weakened topological boundary is determined, boreholes located within the weakened topological boundary are identified as anomalous boreholes, and boreholes located outside the weakened topological boundary are identified as normal boreholes; for each anomalous borehole, the shortest topological geometric distance between each anomalous borehole and the weakened topological boundary is obtained, and the dynamic rock mass strength index of each anomalous borehole is subjected to feature fusion processing with the shortest topological geometric distance to determine the local strength mutation coefficient; for each normal borehole, the local strength mutation coefficient is determined as a preset numerical constant.

[0016] Furthermore, the dynamic rock mass strength index and the shortest topological geometric distance of each abnormal blast hole are subjected to feature fusion processing, including: determining the statistical mean of the dynamic rock mass strength index of each normal blast hole as the benchmark rock breaking resistance of the tunnel face; using the ratio of the dynamic rock mass strength index of each abnormal blast hole to the benchmark rock breaking resistance as the strength reduction coefficient to characterize the degree of material softening at the abnormal blast hole; calculating the clamping compensation factor to characterize the clamping effect of the hard rock boundary on the soft interlayer based on the shortest topological geometric distance between each abnormal blast hole and the weakened topological boundary; and using the clamping compensation factor to perform reverse compensation correction on the strength reduction coefficient to obtain the local strength mutation coefficient.

[0017] This technical solution fully deconstructs the blasting stress transmission mechanism at complex geological interfaces. For blast holes falling into soft or abnormal areas, the actual difficulty of rock fragmentation depends not only on the degree of geological softening at the corresponding location of the blast hole, but also on the clamping effect of the surrounding hard and intact rock mass. By calculating the shortest topological geometric distance and introducing a reverse compensation mechanism, the geometric clamping effect of hard rock on soft rock is accurately assessed, avoiding the problem of insufficient charge caused by excessive reduction based solely on a single soft material, and realizing refined parameter adjustment based on blasting dynamics.

[0018] Furthermore, the dynamic charge amount for each borehole is determined as follows: the minimum resistance line, borehole spacing, and borehole depth are extracted from the preset borehole layout parameters of each borehole; the minimum resistance line, borehole spacing, and borehole depth are multiplied together to obtain the load rock volume as a spatial volume parameter; a lower limit constraint for preventing under-excavation is applied to the local strength mutation coefficient of each borehole; the benchmark explosive consumption of the functional area to which each borehole belongs is corrected and calculated using the constrained local strength mutation coefficient to obtain the target explosive consumption of each borehole; the dynamic charge amount is determined by balancing the load rock volume and the target explosive consumption.

[0019] Based on the volumetric energy consumption method, this technical solution combines the lower limit constraint of preventing under-excavation and directly uses the mutation coefficient to numerically correct the energy consumption of the target explosive. Without changing the main structure of the classical blasting calculation equation, it ensures that even in soft areas, the input energy can maintain the minimum rock breaking threshold required to break the rock structure, thus taking into account both cost control and the bottom line requirements of excavation quality.

[0020] Furthermore, obtaining the dynamic charge amount for each blast hole also includes: acquiring the three-dimensional contour scan data of the tunnel face after the blasting operation is completed, and dividing the three-dimensional contour scan data into local evaluation regions based on the spatial position of each blast hole to extract the over-excavation and under-excavation state characteristics of each blast hole; determining the feedback compensation coefficient of each blast hole based on the over-excavation and under-excavation state characteristics of each blast hole, and iteratively updating the benchmark explosive consumption of the corresponding functional area in the next blasting cycle based on the statistical value of the feedback compensation coefficient of the blast holes in the same functional area.

[0021] This technical solution constructs a complete closed-loop correction system from pre-detonation data acquisition to dynamic optimization of charge quantity and post-blasting effect verification. It introduces three-dimensional point cloud scanning for hole-level over- and under-excavation spatial tracing and provides a feedback compensation strategy for matching different functional boreholes, giving the system the ability to learn and optimize parameters in response to long-term geological condition drift.

[0022] Furthermore, the feedback compensation coefficient for each borehole is determined based on the following: the functional areas include the cutting area, the auxiliary area, and the surrounding area; for boreholes belonging to the cutting area or the auxiliary area, the feedback compensation coefficient is determined based on the ratio of the borehole's load-bearing rock volume to the actual formed volume obtained from the analysis of the three-dimensional contour scan data; for each borehole belonging to the surrounding area, the feedback compensation coefficient is determined based on the ratio of the borehole's minimum resistance line to the actual excavation depth obtained from the analysis of the three-dimensional contour scan data.

[0023] This technical solution takes into account that in tunnel blasting dynamics, the core task of the cut zone and auxiliary zone is large-scale volumetric rock breaking to create internal free surfaces. Therefore, the ratio of the designed load volume to the actual formed volume can accurately assess the volumetric fracturing conversion efficiency of the explosive energy. The core task of the surrounding zone is high-precision contour control to sever the rock mass connection and protect the surrounding rock. Therefore, the ratio of the minimum resistance line representing the radial rock breaking scale boundary to the actual excavation depth can most intuitively capture the radial deviation of the contour control. This differentiated evaluation mechanism ensures that the extracted feedback compensation coefficient can truly map and isolate various types of borehole-specific failure modes, thereby guiding subsequent blasting cycles to perform targeted energy allocation and parameter iteration, and improving the adaptive correction accuracy.

[0024] On the other hand, the present invention provides a data-driven tunnel blasting parameter optimization system, the tunnel blasting parameter optimization system including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of any of the data-driven tunnel blasting parameter optimization methods described above.

[0025] The present invention has the following effects: This invention reconstructs the closed topological boundary of hidden geological anomalies by deeply analyzing single-hole drilling data. It deeply couples the softening reduction of rock mass material in the anomaly area with the geometric confinement effect of the surrounding intact hard rock, adaptively balances the actual rock-breaking resistance and explosive input energy in a single hole, improves the matching accuracy between local explosive energy and actual rock-breaking resistance, avoids the problem of excessive or insufficient local explosive charge at complex heterogeneous interfaces, and reduces explosive consumption and the risk of disturbance to the remaining surrounding rock. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a graph showing the changes in dynamic charge amount and dynamic rock mass strength index of the present invention; Figure 3 This is a comparison chart of the average explosive consumption of the present invention and the traditional method. Detailed Implementation

[0027] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0028] The data-driven tunnel blasting parameter optimization method provided by this invention, such as... Figure 1 As shown, it includes: S1: Obtain the reference parameters of the tunnel face.

[0029] Considering the core issue of poor compatibility between existing blasting design benchmark parameters and actual geological conditions on site, this step involves completing the entire process of dividing the functional areas of blast holes, extracting hole layout parameters, aligning design and actual operation coordinates, and obtaining benchmark explosive consumption and rock mass resistance parameters for functional areas. This ensures the availability of a complete set of standardized and feasible basic data sources for blasting optimization, guaranteeing the compliance and adaptability of blasting parameter optimization from the outset, and providing a unified spatial and parameter benchmark for subsequent full-process calculations.

[0030] The specific operating procedure is as follows: S11: Obtain the borehole functional area and borehole layout parameters.

[0031] The drill-and-blast design grid is a standardized design document completed in advance by blasting designers using CAD and a professional tunnel blasting design system. It contains core design parameters such as the design three-dimensional coordinates of all blast holes at the tunnel face, blast hole spacing, minimum resistance line, blast hole depth, blast hole row spacing, and tunnel design excavation outline, and serves as the basis for construction operations.

[0032] First, based on the spatial geometric relationship between the three-dimensional coordinates of the blast holes and the designed excavation outline of the tunnel, the functional areas to which the blast holes belong are divided. The functional areas are divided into three categories: the cutting area, the auxiliary area, and the surrounding area. The specific division rules are as follows: Cut-out area: The blast holes are located within a preset range corresponding to the geometric center of the tunnel design excavation outline. The preset range is a circular area with a radius of 1.5-2.5m centered on the geometric center of the tunnel face. It can be flexibly adjusted according to the size of the tunnel cross section. The blast holes in this area are cut-out holes. Their core function is to create an initial cavity and a new free face by blasting first when there is only one free face at the tunnel face. This provides free face conditions for the subsequent blasting of auxiliary holes and surrounding holes. It is the core of the success of the entire tunnel face blasting.

[0033] Surrounding area: The blast holes located at the outermost edge of the tunnel design excavation outline. These blast holes are peripheral holes, and their core function is to sever the connection between the retained rock mass and the blasting area rock mass, and to precisely control the tunnel excavation outline. The blasting effect directly determines the tunnel over-excavation and under-excavation rate, the quality of the outline formation, and the stability of the retained surrounding rock.

[0034] Auxiliary area: All blast holes distributed between the surrounding area and the slotting area. The blast holes in this area are auxiliary holes. Their core function is to use the new free surface created by the slotting holes to expand the volume of rock mass fragmentation, complete the blasting of the main rock mass at the working face, and take into account both the fragmentation effect and the control of blasting vibration.

[0035] Subsequently, the hole layout parameters for each borehole were extracted from the drill-blast design grid, including: three-dimensional coordinates of the borehole, borehole spacing, minimum resistance line, borehole depth, and borehole row spacing. All parameters were in SI units, providing a basis for subsequent spatial volume parameter analysis.

[0036] S12: Align the borehole design coordinates with the actual operational coordinates.

[0037] To ensure accurate mapping between the CAD design coordinate system and the actual engineering coordinate system of the drilling rig, and to avoid coordinate deviations causing subsequent calculation failures of topological boundaries and clamping distances, the specific operation strictly follows the general specifications of the tunnel drilling rig construction industry, and the process is as follows: Using a total station, which is standard equipment at the tunnel construction site, two or more stable traverse control points are set up near the tunnel face. The control points are incorporated into the overall engineering control network of the tunnel section. The coordinate system is the independent engineering coordinate system or the 2000 National Geodetic Coordinate System used in the section design, which is completely consistent with the coordinate system of the CAD drilling and blasting design grid.

[0038] The total station was connected to the onboard positioning system of the rock drilling rig for communication and measurement. The spatial positioning and attitude calibration of the rig body were completed. After calibration, the positioning accuracy of the rig's drill arm was not less than ±5mm, which met the accuracy requirements for tunnel blast hole construction.

[0039] The three-dimensional coordinates of the borehole design in the CAD drilling and blasting design grid are imported in batches into the onboard control system of the rock drilling rig. The system automatically completes the mapping and transformation between the design coordinates and the rig's working coordinate system, so that the actual drilling trajectory of the rig's drilling arm corresponds one-to-one with the design borehole coordinates.

[0040] After a single hole is drilled, the trolley-mounted system automatically records the actual three-dimensional coordinates of the hole opening and checks them against the design coordinates. Holes with a deviation of more than 2 cm need to be re-constructed to ensure the consistency between the actual spatial position of all holes and the design coordinates, providing an accurate coordinate basis for subsequent spatial topology analysis.

[0041] S13: Obtain the benchmark explosive consumption and rock mass resistance benchmark parameters for each functional area.

[0042] For the benchmark explosive consumption per unit area in each functional area: Based on the tunnel surrounding rock grade, cross-sectional dimensions, explosive type, and mature construction experience of this section, and in accordance with the requirements of the "Safety Regulations for Blasting" GB6722-2014 and the "Technical Guidelines for Railway Tunnel Engineering Construction" TZ204-2008, the benchmark explosive consumption per unit area for different functional areas is determined. The benchmark explosive consumption per unit area for the cut area is recorded as follows: The standard explosive consumption per unit in the auxiliary area is recorded as follows: The benchmark explosive consumption per unit area in the surrounding area is recorded as follows: All units are The benchmark explosive consumption is the standard explosive consumption benchmark value for intact rock mass within the corresponding functional area. It is fixed in the system as a benchmark value in this blasting cycle and does not need to be repeatedly adjusted unless there are special geological changes.

[0043] For rock mass resistance benchmark parameters, including: Benchmark value of Protodyakonov's roughness coefficient for surrounding rock: Before the commencement of circulating drilling operations, based on the preliminary geological survey report and advanced geological forecast data for the tunnel section (including tunnel seismic wave exploration and ground-penetrating radar detection), combined with the on-site geological logging results at the tunnel face, the overall surrounding rock grade of the tunnel face for blasting operations is determined, and the benchmark Protodyakonov's roughness coefficient for that surrounding rock grade is obtained accordingly. The Protodyakonov rock mass hardness coefficient is an industry-standard indicator that reflects the overall hardness of a rock mass. It was determined by the Protodyakonov rock grading theory proposed by Russian scholar Protodyakonov. The higher the value, the harder the rock mass is and the more difficult it is to break. Conversely, the lower the value, the softer and more easily broken the rock mass is.

[0044] Benchmark Mechanical Energy: A complete rock mass area with uniform geological conditions, no jointed fracture zones, and no soft interlayers is selected in the central area of ​​the tunnel face. This area needs to be confirmed on-site by a geological engineer and consistent with the conclusions of the previous geological survey. The depth of the test hole is consistent with the depth of the boreholes designed for this cycle. One standard test hole is constructed using the same type of rock drilling equipment and the same drilling parameters as this drilling operation. Physical parameters of the entire drilling process are collected during the drilling of the test hole, and the benchmark mechanical energy of the tunnel face is calculated. The benchmark mechanical specific energy reflects the benchmark mechanical energy required to break a unit volume of intact standard rock mass at the tunnel face, and its unit is 1. The larger the value, the higher the energy consumption for breaking the intact rock mass and the greater the inherent strength. It serves as the benchmark for subsequent single-hole rock mass strength correction and is solidified into the system as a fixed benchmark value in this cycle of operation.

[0045] This embodiment takes a Class III surrounding rock section of a double-track railway hard rock tunnel as an example. The excavation width of the tunnel face in this cycle is 12.6m, the excavation height is 9.8m, the total number of blast holes is 128, the cycle advance is 3.5m, No. 2 rock emulsion explosive is used, and a three-arm hydraulic drilling rig is used for drilling. The drill bit diameter is... =45mm.

[0046] Preliminary geological surveys determined that the working face is classified as Class III surrounding rock, corresponding to the Protodyakonov benchmark value for the surrounding rock. A standard test hole was drilled in the central intact rock mass area of ​​the working face to calculate the baseline mechanical specific energy. Based on the previous construction experience and specifications of this section, the benchmark explosive consumption per unit area in the trenching area was determined. Auxiliary area benchmark explosive consumption The benchmark explosive consumption per unit area .

[0047] The drilling rig was calibrated and aligned with the design coordinate system using a total station. The drilling arm positioning accuracy of the rig was ±3mm, and the deviation between the actual construction coordinates and the design coordinates of all blast holes was less than 1cm, meeting the accuracy requirements. Through the extraction of the drill-blast design grid, out of 128 blast holes, 16 were located in the cut area, 88 in the auxiliary area, and 24 in the surrounding area. Typical blast hole layout parameters are as follows: Hole No. 1 located in the cut area: Hole spacing Line of least resistance Hole depth ; Hole No. 50 in the auxiliary area: Hole spacing Line of least resistance Hole depth ; Hole No. 100 located in the surrounding area: Hole spacing Line of least resistance Hole depth .

[0048] This completes the acquisition and preprocessing of all basic parameters required for blasting optimization. It not only obtains the rock mass resistance benchmark parameters that are suitable for the overall geological conditions of the tunnel face, but also clarifies the functional affiliation of each blast hole, the hole layout parameters, and the benchmark explosive consumption of the corresponding functional area. This provides a standardized and feasible basic data source for subsequent full-process calculations, ensuring the compliance and adaptability of blasting parameter optimization from the source.

[0049] S2: Determine the dynamic rock mass strength index and generate the weakened topological boundary.

[0050] After completing the unified alignment of the borehole layout parameters, benchmark rock mass resistance parameters, and actual coordinate system of the entire face, this step further considers the pain points of existing technologies that cannot identify continuous geological anomaly areas from discrete borehole-level strength data and cannot assess the spatial location of the interface between soft and hard rocks.

[0051] Therefore, by analyzing the actual mechanical specific energy of a single hole through the native drilling operation data of the rock drilling rig, the dynamic rock mass strength index at the hole level is obtained through mapping and transformation. At the same time, through spatial gradient analysis and clustering, the strength anomaly areas are automatically identified and closed weakened topological boundaries are generated to achieve accurate hole-level assessment of the rock breaking difficulty of the face rock mass. This provides an accurate spatial geometric basis and strength data support for the subsequent calculation of the local strength mutation coefficient.

[0052] The specific operating procedure is as follows: S21: Extract the actual mechanical specific energy of a single borehole based on drilling operation data.

[0053] First, data collection and standardization were performed during drilling operations. This drilling operation utilized a two-arm / three-arm hydraulic drilling rig for face borehole drilling. The rig's onboard control system incorporated hydraulic sensors, speed sensors, torque sensors, and displacement encoders. Throughout the drilling process of each borehole, the rig's control system synchronously collected real-time operational data at a 100-millisecond sampling period, including drill bit advance pressure. Drill bit rotation speed Drill bit rotation torque Axial drilling speed of the drill bit The physical meanings of each parameter are as follows: Drill bit propulsion pressure : Reflects the axial thrust applied to the drill bit by the hydraulic system of the rock drilling rig, measured in units of The higher the value, the higher the driving force required for axial rock breaking, and the greater the axial compressive strength of the rock mass; drill bit rotation speed : Reflects the number of revolutions of the drill bit per minute, in units of The core setting parameter for rock drilling operations directly affects the rock cutting efficiency; the drill bit rotation torque : Reflects the frictional resistance torque experienced by the drill bit when it rotates and cuts the rock mass, with units of . The higher the value, the harder the rock mass and the greater the cutting resistance; conversely, the lower the value, the softer the rock mass. Axial drilling speed of the drill bit. : Reflects the axial drilling depth of the drill bit per unit time, with units of . The higher the value, the easier the rock mass is to break and the higher the drilling efficiency; conversely, the harder the rock mass is and the more difficult it is to break.

[0054] After drilling of a single borehole is completed, the system automatically standardizes the time-series parameters collected during the entire drilling process, converting them into a single steady-state index value corresponding to the single borehole. The processing method strictly follows the industry-standard specifications for tunnel drilling MWD (Measurement While Drilling), and the specific operation is as follows: Drilling parameter data for the first 0.5m of the hole opening section and the last 0.3m of the hole closing section are discarded. This data is affected by the free face at the hole opening and the step effect at the hole closing, and cannot truly reflect the inherent characteristics of the rock mass inside the hole. This is a common rule for discarding invalid sections in the industry.

[0055] Box plots were used to remove outliers from the remaining valid segments of the time series data. The abnormal fluctuation data within the range eliminates parameter deviations caused by instantaneous anomalies such as fluctuations in the hydraulic system of the rock drilling rig, stuck drill, and drill jumping. Here, Q1 is the lower quartile (25th percentile) of the time series data, Q3 is the upper quartile (75th percentile) of the time series data, and IQR is the interquartile range, the difference between Q3 and Q1.

[0056] For the valid data after removing outliers, the arithmetic mean method is used to calculate the individual numerical index of each dimension parameter corresponding to the borehole, which serves as the basic input parameter for subsequent calculations.

[0057] At this point, after completing the drilling operations and parameter standardization of all blast holes on the entire face, the steady-state operation dataset of all blast holes can be obtained. This data can be collected simultaneously during the normal drilling process of the rock drilling rig without taking up additional construction time.

[0058] Then, based on the standardized operating data, the actual mechanical specific energy of a single hole is obtained through analysis.

[0059] Calculate the cross-sectional area of ​​the borehole:

[0060] In the formula, The cross-sectional area of ​​the borehole is given in units of 1000 m². , The diameter of the drill bit used in this drilling operation is given in units of [diameter value missing]. These are the fixed, known parameters for rock drilling operations. Let π be 3.1415926.

[0061] This formula is a classic formula for calculating the area of ​​a circle in plane geometry, and it is the industry-standard formula for calculating the cross-sectional area of ​​blast holes in tunnel drilling operations.

[0062] Extracting axial linear extrusion power consumption:

[0063] In the formula, This refers to the linear extrusion power per unit volume of rock that the drill bit consumes to overcome the axial compressive strength of the rock during borehole drilling, expressed in units of... This reflects the energy consumption per unit volume of axial rock-breaking propulsion. The steady-state drill bit propulsion pressure after the borehole has been standardized, in units of... .

[0064] This formula is based on the theory of conservation of energy for axial rock breaking in rock mechanics. It converts the axial propulsion force into the axial work done to break a unit volume of rock. The larger the value, the higher the axial compressive strength of the rock mass and the greater the difficulty of axial rock breaking.

[0065] Extract the power of rotating machinery and convert it into power consumption per unit volume of rotating shear: Calculate the power of rotating machinery:

[0066] In the formula, This refers to the rotary mechanical power generated by the drill bit during drilling of the borehole, expressed in units of... , The steady-state drill bit rotation speed of the borehole, in units of , The steady-state drill bit rotation torque of the borehole, in units of ;60 is the time conversion factor, which converts minutes to seconds to ensure the uniformity of power dimensions.

[0067] This formula is a classic formula for calculating the power of rotating machinery. It is a commonly used torque-speed-power conversion model in mechanical engineering, following the power calculation principle for rigid body rotation about a fixed axis. In the formula, The angular velocity of the drill bit's rotation, and the torque Multiplying these values ​​directly yields the mechanical power of the rotary cutting. The larger the value, the higher the energy required for rotary cutting and the greater the resistance to rock cutting.

[0068] Determine the circumferential rotational shear power consumption:

[0069] In the formula, This refers to the circumferential rotational shearing power consumption (rotational shearing power consumption per unit volume) of the drill bit overcoming the cutting friction of the rock mass during borehole drilling. The unit is... This reflects the energy consumption per unit volume of rotary cutting rock breaking. The steady-state axial drilling speed of the drill bit at the borehole is given in units of 1. ;60 is the time conversion factor to ensure dimensional consistency.

[0070] This formula is based on the theory of energy conservation in rock breaking by rotation. It converts the rotational power per unit time into the work done by the rotation of a unit volume of rock. The larger the value, the greater the resistance to the rotational cutting of the rock mass and the higher the difficulty of shearing and breaking it.

[0071] Based on the principle of energy conservation, the energy consumption of axial linear extrusion and circumferential rotational shearing is aggregated to obtain the actual mechanical specific energy. :

[0072] In the formula, The actual mechanical specific energy during the borehole drilling process, in units of , can be converted This value reflects the total mechanical energy required to break a unit volume of rock at the location of the borehole. The larger the value, the harder the rock at the location of the borehole and the greater the breaking resistance. Conversely, the smaller the value, the softer the rock and the presence of jointed fracture zones or soft interlayers.

[0073] The essence of rock breaking by a rock drill bit is the coupling effect of axial crushing and circumferential shearing. Therefore, the total rock breaking energy consumption is a linear superposition of axial linear compression power consumption and circumferential rotational shearing power consumption, which is a mature mechanical energy calculation model commonly used in the field of tunnel rock drilling.

[0074] S22: Determine the dynamic rock mass strength index of a single borehole.

[0075] By utilizing the relative ratio between the actual mechanical specific energy and the benchmark mechanical specific energy, the benchmark value of the Protodyakonov robustness coefficient of the surrounding rock is mapped and transformed to obtain the dynamic Protodyakonov robustness coefficient, which serves as a dynamic rock mass strength index. The specific steps are as follows: Calculate the rock mass strength mapping coefficient:

[0076] In the formula, It is the rock mass strength mapping coefficient of the borehole, which reflects the strength difference of the rock mass at a single borehole location relative to the intact reference rock mass at the working face; The actual mechanical specific energy of the borehole, in units of , The reference mechanical specific energy at the tunnel face, in units of .

[0077] The strength of rock mass is linearly positively correlated with the mechanical energy required for fracturing. Therefore, the ratio of the actual mechanical specific energy of a single hole to the benchmark mechanical specific energy can be directly equivalent to the ratio of the rock mass strength of a single hole to the benchmark rock mass strength. This indicates that the rock mass strength at the location of the blast hole is higher than the average level at the working face, resulting in greater resistance to breaking. This indicates that the rock mass strength at the location of the blast hole is lower than the average level of the working face, indicating the existence of a soft and abnormal area with even less resistance to breaking.

[0078] Obtain the dynamic Protodyakonov strength factor:

[0079] In the formula, The dynamic Protodyakonov strength coefficient of the borehole is dimensionless, representing the dynamic rock mass strength index required in this step. It has hole-level spatial resolution and reflects the true strength of the rock mass at the borehole location. This is the benchmark value for the Protodyakonov strength coefficient of the surrounding rock at the tunnel face.

[0080] This formula is based on the energy equivalence principle of mechanical rock breaking and explosive rock breaking, realizing the cross-domain conversion between mechanical cutting energy efficiency and rock dynamic strength. It not only retains the benchmark evaluation role of the classic Protodyakonov coefficient in blasting engineering calculations and conforms to the industry's common calculation habits, but also realizes the hole-level dynamic evaluation of rock mass strength.

[0081] This embodiment uses a 45mm rock drill bit, and the cross-sectional area of ​​the blast hole is... Taking borehole No. 1 as an example, the standardized steady-state operating data is as follows: propulsion pressure Rotation speed Rotational torque Axial drilling speed .

[0082] Axial linear extrusion power consumption: ; Rotating machinery power: ; Circumferential rotational shear power consumption: ; Actual mechanical specific energy: ; Rock mass strength mapping coefficient: ; Dynamic Protodextrins strength coefficient ; The results indicate that the rock mass strength at the location of this borehole is lower than the average level at the working face, belonging to an abnormally soft region. Similarly, the dynamic Protodyakonov robustness coefficient was calculated for all 128 boreholes at the working face, resulting in a dataset of dynamic rock mass strength indices for all boreholes.

[0083] S23: Identify regions with abnormal intensity and generate weakened topological boundaries.

[0084] After obtaining the dynamic rock mass strength index at the face level of the entire working face, in order to identify continuously distributed geological anomaly areas from discrete single-hole data and accurately locate the spatial position of the interface between soft and hard rocks, this step automatically identifies the strength anomaly areas in the working face through spatial gradient analysis, clustering, and boundary fitting, and generates a closed weakened topological boundary that encloses the area, providing accurate spatial geometric basis for subsequent charge correction considering the boundary clamping effect.

[0085] The specific operating procedure is as follows: First, identify areas of abnormal strength. Extract the 3D coordinates of boreholes with a dynamic Protodyakonov strength coefficient below f0. A coefficient below the benchmark indicates softening of the rock mass at that location, and these boreholes are defined as weakened nodes. Perform spatial clustering on the 3D coordinates of all weakened nodes using the DBSCAN density clustering algorithm, a widely used spatial point clustering algorithm in the industry. This algorithm can automatically identify continuously distributed spatial point clusters without pre-setting the number of clusters. Set the neighborhood radius to the average spacing between adjacent boreholes, and the minimum sample size to 3. This minimum sample size is determined based on the borehole density at the tunnel face, ensuring the identification of continuously distributed abnormal areas and avoiding misjudgments of single-hole abnormal fluctuations. For large-section tunnels with more than 200 boreholes, this can be adjusted to 5 to ensure clustering accuracy.

[0086] Clustering is used to generate multiple continuously distributed weak node clusters. The spatial range corresponding to each weak node cluster is determined as the strength anomaly region, that is, a continuous geological region in which the rock mass strength in the tunnel face is significantly lower than the benchmark level and there are soft interlayers or joint fracture zones.

[0087] Then, the gradient of dynamic rock mass strength indices between adjacent blast holes in the tunnel face is statistically analyzed. The formula for calculating the gradient is:

[0088] In the formula, For two adjacent blast holes The gradient of changes in dynamic rock mass strength indices between them , respectively the gun holes The dynamic Protodextrin robustness coefficient, For gun holes With the gun hole The straight-line distance between them, in units of This gradient reflects the degree of abrupt change in rock mass strength between adjacent boreholes. The larger the gradient value, the more drastic the change in geological properties between the two boreholes, and the more likely there is a soft-hard rock interface.

[0089] This formula is based on the fundamental principle of spatial gradient analysis. It assesses the degree of spatial abrupt change in rock mass strength by using the ratio of the strength difference between adjacent boreholes to the spatial distance. It is a common analytical method in the industry for identifying geological anomaly areas.

[0090] It should be noted that the determination of adjacent blast holes in the tunnel face is not limited by the functional area to which the blast hole belongs. A spatial topological grid is constructed based on the three-dimensional coordinates of all blast holes in the entire tunnel face. Two blast holes in the grid that have a direct line connection are determined to be adjacent blast holes.

[0091] Finally, based on the dynamic rock mass strength index variation gradient data of all boreholes in the entire working face, a dynamic identification threshold for characterizing geological structural abrupt changes is constructed. The dynamic identification threshold adopts the upper quartile (75th quartile) of the variation gradient data of the entire working face. Alternatively, it can be determined by dynamic statistical methods such as the average value of the variation gradient data of the entire working face plus one standard deviation and the 90th quartile, depending on the complexity of the on-site geological conditions. This ensures that the threshold can be adaptively adjusted according to the actual distribution of rock mass strength in the working face, accurately identifying areas of geological structural abrupt changes.

[0092] The coordinates of boreholes whose dynamic rock mass strength index change gradient is greater than the dynamic identification threshold are extracted. The outermost coordinates of each weakened node cluster are then enclosed in the extracted coordinates of boreholes whose change gradient is greater than the dynamic identification threshold. The convex hull algorithm is used to perform geometric connection closure. By tracing the outermost coordinates of each weakened node cluster, the closed weakened topological boundary of the corresponding strength anomaly region is obtained. This boundary can accurately characterize the spatial position of the soft and hard rock interface.

[0093] The weakened topological boundary is a closed polygon with clear geometric meaning, which can accurately represent the spatial position of the interface between soft and hard rock. It is the core foundation for subsequent calculations of the relative positions of boreholes and anomaly areas and the evaluation of boundary clamping effects. The convex hull algorithm is a general and mature algorithm in the field of computational geometry. It can automatically generate a closed topological boundary by simply inputting the outer coordinates of the weakened node cluster.

[0094] In this embodiment, the coordinates of 18 boreholes with dynamic Protodyakonov robustness coefficients lower than the baseline value f0 are extracted. A continuous weakened node cluster is generated by DBSCAN clustering. At the same time, the changing gradient is calculated and the coordinates of boreholes with gradients greater than the dynamic identification threshold are extracted. The convex hull algorithm is used to connect and fit the outermost coordinate points of the extracted coordinates that enclose the weakened node cluster to obtain the weakened topological boundary that encloses the strength anomaly region. This boundary covers 12 boreholes, including the No. 1 slotted borehole.

[0095] S3: Determine the local intensity mutation coefficient based on the weakened topological boundary.

[0096] After obtaining the dynamic rock mass strength index at the full-face hole level, identifying areas of abnormal strength, and generating closed weakened topological boundaries, this step further considers the blasting stress transmission mechanism at the interface between hard and soft rocks. The hard and intact rock mass on the outside will have a spatial clamping effect on the soft interlayer inside, which will also affect the accurate assessment of the rock breaking difficulty of the borehole and cause the problem of inaccurate charge quantity in the later stage.

[0097] Therefore, this step combines the spatial location characteristics of the borehole relative to the weakened topological boundary, as well as the dynamic rock mass strength index, and also considers the clamping compensation effect of the hard rock boundary to calculate the local strength mutation coefficient that can simultaneously reflect the degree of material softening and the boundary clamping effect, providing a precise core basis for subsequent charge correction.

[0098] The consideration of rock mass material strength and the neglect of the clamping effect between soft and hard rock boundaries lead to an underestimation of rock breaking difficulty and excessive reduction of explosive charge, resulting in under-excavation.

[0099] Therefore, by classifying the spatial location of boreholes and coupling the strength reduction coefficient with the clamping compensation factor, the local strength mutation coefficient is solved to accurately assess the actual rock breaking difficulty of a single borehole location, providing a core basis for subsequent explosive consumption correction.

[0100] The specific operating procedures include: S31: Determine the spatial location and classification of boreholes.

[0101] Based on the spatial geometric relationship between the three-dimensional coordinates of each borehole and the weakened topological boundary, the spatial position of the borehole relative to the weakened topological boundary is determined, and the boreholes are divided into two categories.

[0102] The first type is anomalous boreholes: boreholes located within weakened topological boundaries, that is, boreholes within areas of abnormal strength. The rock mass corresponding to these boreholes is softened in strength and is also constrained by the surrounding hard rock boundaries.

[0103] The second type is normal boreholes: boreholes located outside the weakened topological boundary, that is, boreholes in the complete rock mass area outside the strength anomaly area and whose rock mass strength is consistent with the benchmark level. The rock breaking difficulty of these boreholes is determined only by the inherent strength of the rock mass itself and is not affected by the clamping effect of the boundary of the anomaly area.

[0104] S32: For different types of boreholes, a differentiated calculation method is used to determine the local intensity mutation coefficient.

[0105] For normal blast holes, the local strength mutation coefficient is determined as a preset numerical constant. The core principle for determining this preset numerical constant is: it matches the rock-breaking resistance of the intact benchmark rock mass at the working face, with no strength anomalies or boundary clamping effects, and no need to correct the benchmark explosive consumption. In this embodiment, the preset numerical constant is set to 1 by default. This value represents that the rock mass strength of the normal blast hole is completely consistent with the benchmark rock-breaking resistance at the working face, and no additional correction is needed for the benchmark explosive consumption, which conforms to the blasting design specifications for intact rock masses. Only when there is a systematic deviation between the dynamic rock mass strength index of the normal blast hole and the benchmark rock-breaking resistance, it can be adjusted within a narrow range of [0.9, 1.1] based on the statistical ratio of the dynamic rock mass strength index of the normal blast hole to the benchmark rock-breaking resistance. If it exceeds this range, the surrounding rock benchmark parameters and the validity of data acquisition need to be re-verified to avoid systematic deviations in the charge parameters.

[0106] For abnormal boreholes, the local intensity mutation coefficient is determined according to the following procedure: First, obtain the shortest topological geometric distance d between each anomalous borehole and the weakened topological boundary, in meters. The shortest topological geometric distance refers to the minimum perpendicular distance from the three-dimensional coordinate point of the anomalous borehole to each edge of the closed polygon of the weakened topological boundary. This can be automatically solved using CAD software or spatial geometric calculation algorithms. The smaller the value, the closer the anomalous borehole is to the interface between soft and hard rock, and the stronger the clamping effect of the surrounding hard rock. The larger the value, the closer the anomalous borehole is to the center of the intensity anomaly area, and the weaker the clamping effect.

[0107] Subsequently, feature fusion processing was performed on the dynamic rock mass strength indices and shortest topological geometric distances of each abnormal borehole. Specifically, this included: The statistical average of the dynamic rock mass strength indices of each normal borehole is determined as the benchmark rock-breaking resistance at the tunnel face. The average dynamic Protodyakonov hardness coefficient of all normal boreholes is used as the benchmark rock breaking resistance, representing the average true rock breaking resistance of intact rock mass within the working face. It is a benchmark for assessing the degree of rock mass strength reduction in abnormal areas.

[0108] The ratio of the dynamic rock mass strength index of each abnormal borehole to the benchmark rock breaking resistance is used as a strength reduction factor to characterize the degree of material softening at the abnormal borehole. , When the value is lower, it indicates that the rock mass strength at the location of the borehole is lower than the baseline level of intact rock mass. The smaller the value, the higher the degree of rock softening, and theoretically the lower the rock breaking energy required.

[0109] Based on the shortest topological geometric distance between each anomalous borehole and the weakened topological boundary, a clamping compensation factor is calculated to characterize the clamping effect of the hard rock boundary on the soft interlayer. The formula is:

[0110] In the formula, The clamping compensation factor for abnormal boreholes is dimensionless and its value range is [value range missing]. , The shortest topological geometric distance between the anomalous borehole and the weakened topological boundary, in meters; The attenuation characteristic distance for clamping effect, in meters.

[0111] In this embodiment, The statistical average of the minimum resistance lines of all normal blast holes within the functional area to which the abnormal blast hole belongs is taken. This value is determined based on the constraint range of the clamping effect commonly used in tunnel blasting engineering. The influence range of the blasting clamping effect is directly related to the minimum resistance line of the blast hole. After exceeding the minimum resistance line, the constraint effect of the soft and hard rock interface can be ignored. In conventional tunnel blasting, the minimum resistance line is taken as 0.5-0.8m for Class III surrounding rock, 0.6-0.9m for Class IV surrounding rock, and 0.8-1m for Class V surrounding rock. It can be slightly adjusted according to the degree of joint development of the geological conditions on site.

[0112] This formula is based on the linear proportional correction theory commonly used in tunnel blasting engineering. It conforms to the fundamental conclusion that the clamping effect strength and the constraint distance are linearly negatively correlated in blasting engineering. The physical logic is as follows: the closer the abnormal blast hole is to the interface between soft and hard rock, the stronger the reflection constraint effect of the blasting stress wave at the interface, the greater the spatial clamping effect on the rock mass, and the higher the required charge compensation range; the farther the distance, the weaker the clamping effect, and the lower the compensation range; when the distance exceeds the minimum resistance line of the blast hole, the clamping effect disappears completely, and no additional compensation is required.

[0113] Finally, using the clamping compensation factor, the strength reduction coefficient is corrected by reverse compensation to obtain the local strength abrupt change coefficient, as shown in the formula:

[0114] In the formula, This is the local intensity abrupt change coefficient for abnormal boreholes, dimensionless, and a core coefficient for subsequent explosive consumption correction. This is the strength reduction factor. This is the clamping compensation factor for abnormal blast holes.

[0115] The formulation of this formula closely aligns with the stress wave reflection theory and the matching criterion between blasting energy and wave impedance in rock blasting dynamics, and mathematically employs a classic multi-factor linear superposition coupling model. Specifically: In classical rock blasting mechanics, the actual rock-breaking resistance of a single borehole is not a single variable, but is determined by both the intrinsic mechanical properties of the rock and the spatial structural constraints. The formula... Following the theory of rock mass damage evolution, this formula characterizes the absolute reduction in intrinsic fracturing energy consumption when soft interlayers or densely jointed zones exist within the rock mass. As a clamping compensation term, its underlying logic stems from the impedance abrupt change and boundary reflection effect during stress wave propagation. According to the basics of blasting dynamics, when the explosive stress wave propagates from the soft anomalous region to the surrounding intact hard rock and reaches the topological interface, due to the difference in wave impedance between the soft and hard rock masses, the hard rock boundary will hinder the radial displacement of the soft rock and generate stress wave reflection. Macroscopically, this manifests as a strong spatial clamping effect; the closer to the boundary, the stronger the spatial clamping effect. The smaller the value, the greater the boundary constraint stiffness, and the higher the additional compensation energy required to overcome the clamping.

[0116] By linearly superimposing the strength reduction factor and the clamping compensation term, it simultaneously reflects the strength softening of the rock mass itself and the boundary clamping effect of the surrounding hard rock. When the blast hole is close to the interface between soft and hard rock, the clamping compensation term is automatically amplified to avoid under-excavation caused by excessive reduction of charge based solely on material strength. When the blast hole is located in the center of an abnormal area, the clamping compensation term is automatically reset to zero, and the charge is reduced solely based on material strength, thus achieving an accurate assessment of the rock breaking difficulty.

[0117] Meanwhile, to prevent under-digging caused by excessive charge reduction in extremely soft areas, a lower limit constraint for preventing under-digging is applied to the local strength mutation coefficient. In this embodiment, the constraint... That is, when the calculated result When it is less than 0.6, force take This lower limit is determined based on the minimum rock-breaking energy threshold for tunnel blasting. Verified through multiple field tests, this lower limit ensures that the charge energy in extremely soft areas is sufficient to break the rock structure and prevent under-excavation. For different surrounding rock grades, the lower limit can be adjusted proportionally: 0.6 for Grade III, 0.55 for Grade IV, and 0.5 for Grade V, ensuring a minimum rock-breaking threshold suitable for different rock masses.

[0118] In this embodiment, the arithmetic mean of the dynamic Protodyakonov hardening coefficients of all normal boreholes is used. It is consistent with the benchmark Protodyakonov robustness factor.

[0119] Taking borehole No. 1 as an example, this borehole is located within the weakened topological boundary and is considered an anomalous borehole. Its dynamic Protodyakonov robustness coefficient is 5.02, its shortest topological geometric distance from the weakened topological boundary is 0.3m, and its clamping effect attenuation characteristic distance is... Calculate the strength reduction factor Squeezing compensation factor Local intensity abrupt change coefficient ,satisfy The lower bound constraint.

[0120] The results show that although the borehole is located in a soft area and the rock mass itself is reduced to 62.8% of the baseline, due to its proximity to the interface between soft and hard rocks and the strong clamping effect of the surrounding hard rock, the final local strength mutation coefficient is corrected to 0.814, thus avoiding the problem of under-excavation caused by excessive reduction of the charge.

[0121] Similarly, the local strength mutation coefficient of all boreholes on the entire face is calculated to obtain the local strength mutation coefficient corresponding to each borehole.

[0122] S4: Optimize the charge amount based on the local intensity mutation coefficient.

[0123] After calculating the local strength mutation coefficient for all boreholes on the entire face, this step further addresses the mismatch between the single-hole charge energy and the actual rock-breaking resistance in existing technologies. By analyzing the rock mass volume borne by a single hole, correcting the benchmark explosive consumption based on the local strength mutation coefficient, and completing the dynamic charge balance calculation, the hole-level adaptive optimization of blasting parameters is achieved, ensuring a precise match between the explosive input energy and the actual rock-breaking resistance.

[0124] The specific operating procedure is as follows: S41: Analyze the spatial volume parameters of the borehole.

[0125] The minimum resistance line, borehole spacing, and borehole depth are extracted from the preset borehole layout parameters for each borehole to calculate the volume of the burdened rock mass, which serves as a spatial volume parameter. The formula is as follows:

[0126] In the formula, This represents the volume of rock mass that the borehole is responsible for breaking during blasting, expressed in m³. The larger the value, the larger the volume of rock mass that the borehole is responsible for breaking, and the higher the total charge required. The distance between the boreholes. The line of least resistance, This represents the depth of the borehole.

[0127] This formula is a classic and universal formula for calculating the load volume of tunnel blasting boreholes. Based on the principle of column volume calculation, it multiplies the two-dimensional load area of ​​the tunnel face with the axial drilling depth to obtain the three-dimensional rock mass volume that the borehole is responsible for breaking. It is the core basic parameter for calculating the charge amount using the volumetric unit consumption method.

[0128] S42: Calculate the target explosive consumption per unit.

[0129] First, a lower limit constraint to prevent under-drilling is applied to the local strength mutation coefficient of each borehole. In this embodiment, the constraint... Ensure that the explosive energy is not lower than the minimum rock-breaking threshold.

[0130] Then, using the constrained local intensity mutation coefficient, the baseline explosive consumption of each borehole's functional area is corrected and calculated to obtain the target explosive consumption of each borehole, as shown in the formula:

[0131] In the formula, The target explosive consumption per blast hole, in units of It is adapted to the actual rock-breaking difficulty of the blast hole; This is the baseline explosive consumption per unit area for the functional zone to which this borehole belongs, in units of ; The coefficient of change in local intensity of the borehole is dimensionless.

[0132] This formula, without altering the industry-standard volumetric energy consumption method design framework, introduces a local intensity mutation coefficient as a correction factor, achieving an adaptive match between explosive energy consumption and the actual rock-breaking difficulty of a single borehole: when At this time, the explosive consumption per unit automatically increases to ensure that hard rock is fully broken and to avoid under-excavation; when At the same time, the explosive consumption is automatically reduced to avoid over-excavation and surrounding rock disturbance caused by excessive explosive loading in soft areas. At the same time, the lower limit constraint avoids under-excavation caused by excessive reduction.

[0133] S43: Determine the dynamic charge amount.

[0134] Energy balancing calculations are performed using the volume of the burdened rock mass and the unit explosive consumption of the target to determine the dynamic charge amount for the borehole. The formula is as follows:

[0135] In the formula, The dynamic charge amount for the borehole, in units of The calculation results can be directly used for on-site explosive loading operations. The target explosive consumption is calculated using the local intensity mutation coefficient correction, and the unit is... , The volume of the rock mass supporting the borehole, in units of .

[0136] This formula is a classic and universal formula for calculating the charge amount using the volumetric consumption method in tunnel blasting. It is a common standard method for tunnel blasting design in China. The total charge amount is the product of the explosive consumption per unit volume and the volume of the supporting rock mass. The value is linearly positively correlated with the explosive consumption per unit volume and the volume of the supporting rock mass.

[0137] In this embodiment, the functional area to which borehole No. 1 belongs is the slotting area, and its benchmark explosive consumption is substituted into the formula. The baseline explosive consumption per unit area for the cut-out region ,Right now Local intensity abrupt change coefficient The hole layout parameters include: hole spacing. Line of least resistance Hole depth The volume of the burdened rock mass Target explosive consumption Dynamic charge quantity .

[0138] It can be seen that, compared with the traditional fixed unit consumption of explosive charge... This method automatically reduces the amount of explosive charge in the boreholes in the soft area, while avoiding excessive reduction through clamping compensation, thus curbing the risk of over-excavation.

[0139] In this embodiment, the functional area to which borehole No. 50 belongs is an auxiliary area, and its benchmark explosive consumption is substituted into the formula. Take the baseline explosive consumption of the auxiliary area ,Right now Located outside the weakened topological boundary, it is a normal borehole with a local intensity abrupt change coefficient. The hole spacing parameters are: hole spacing Line of least resistance Hole depth The volume of the burdened rock mass Target explosive consumption Dynamic charge quantity It has the same charge quantity as traditional fixed-consumption explosives and is suitable for breaking up intact rock masses.

[0140] In this embodiment, the functional area to which borehole No. 100 belongs is the surrounding area, and its benchmark explosive consumption is substituted into the formula. Take the benchmark explosive consumption per unit area ,Right now Located outside the weakened topological boundary, it is a normal borehole with a local intensity abrupt change coefficient. The hole spacing parameters are: hole spacing Line of least resistance Hole depth Then the volume of the supporting rock mass Target explosive consumption Dynamic charge quantity This meets the core requirement of adapting to the control of the surrounding hole contour.

[0141] Similarly, the dynamic charge calculation for all 128 blast holes on the entire face is completed, generating a blasting charge table that can be directly used for on-site construction, thus achieving adaptive optimization of blasting parameters.

[0142] S5: Source analysis of blasting effects and iterative update of benchmark parameters.

[0143] After completing the calculation of dynamic charge quantity at the face level for the entire blasting face and implementing the on-site blasting operation, this step further considers the long-term drift effect of geological conditions in order to avoid the blasting design parameters failing to adapt and iterate with changes in geological conditions.

[0144] This step involves post-blasting 3D contour scanning, hole-level over- and under-excavation status tracing, and feedback compensation coefficient calculation to complete the iterative update of the benchmark explosive consumption for the next cycle. This constructs a complete technical closed loop from data acquisition and parameter optimization to effect verification, giving the system the ability to learn in response to long-term changes in geological conditions.

[0145] The specific operating procedure is as follows: S51: Obtain the 3D contour scan data after the blast.

[0146] After the blasting operation is completed and the muck is removed, a 3D laser scanner, which can be a tunnel cross-section scanner or a handheld 3D laser scanner, is used to comprehensively scan the outline of the tunnel face after blasting, obtain high-density 3D point cloud outline data, and the scanning accuracy is not less than 5mm, so as to ensure accurate identification of over-excavation and under-excavation areas, and provide an accurate data source for subsequent blasting effect evaluation.

[0147] S52: Divide the local evaluation area and extract the over- or under-mining state features.

[0148] Based on the three-dimensional coordinates of each blast hole at the tunnel face, the three-dimensional contour scanning data is divided into local evaluation regions. In this embodiment, the Dirichlet space division method is adopted: using the three-dimensional coordinates of each blast hole as discrete points, Thiessen polygons are generated with one click using CAD software or a tunnel blasting design system. Each polygon region is the local evaluation region of the corresponding blast hole. This method can achieve non-overlapping and full-coverage division of the tunnel face, accurately match the blasting influence range of each blast hole, and realize hole-level spatial positioning of blasting effect.

[0149] The contour scan point cloud data of each local evaluation area is compared with the tunnel design excavation contour line to extract the over-excavation and under-excavation characteristics of each blast hole: For boreholes whose functional areas are slotting areas or auxiliary areas, extract the actual forming volume within their local evaluation area. The unit is m³, which is the actual volume of rock mass broken after the blast hole is blasted. For blast holes whose functional area is the surrounding area, extract the actual excavation depth within their local evaluation area. The unit is meters (m), which represents the actual radial excavation depth after the blast hole is opened.

[0150] S53: Calculate the feedback compensation coefficient.

[0151] For blast holes in different functional areas, a differentiated approach is used to determine the feedback compensation coefficient λ, ensuring that the feedback coefficient can accurately map the blasting energy conversion efficiency and failure mode of the corresponding blast hole.

[0152] For blast holes whose functional areas are slotting areas or auxiliary areas, the volume of the rock mass bearing the blast hole is determined. Compared with actual molded volume The ratio of these values ​​is used as the feedback compensation coefficient. Based on the principle of evaluating the volumetric fracturing efficiency of blasting, the core task of the cut-out zone and auxiliary zone is large-scale volumetric rock breaking. This ratio can accurately assess the volumetric fracturing conversion efficiency of blast energy: when Over-excavation occurred. The baseline explosive consumption per unit area needs to be reduced in the next cycle; when At that time, under-excavation occurred. The baseline explosive consumption per unit area needs to be increased in the next cycle; when hour, The baseline explosive consumption remains unchanged.

[0153] For blast holes whose functional area is the surrounding area, the minimum resistance line (design excavation depth of the blast hole) Wk and the actual excavation depth are considered. The ratio of these values ​​is used as the feedback compensation coefficient. Based on the principle of tunnel contour control accuracy evaluation, the core task in the surrounding area is high-precision contour control. This ratio can most intuitively capture the radial deviation of contour control: when Over-excavation occurred. It is necessary to reduce the baseline explosive consumption; when At that time, under-excavation occurred. It is necessary to increase the baseline explosive consumption per unit; when hour, The baseline explosive consumption remains unchanged.

[0154] Meanwhile, to prevent data divergence caused by extreme non-explosive geological events such as fault collapses, a preset feedback compensation coefficient constraint range is implemented within the system. In this embodiment, the constraint range is... When the calculated When the value exceeds this range, its upper and lower thresholds are forcibly applied, and an early warning log is generated to prompt manual review. This range is determined based on the conventional over- and under-excavation control range for tunnel blasting construction. Deviations exceeding this range are usually caused by non-blasting factors such as fault collapse and geological abrupt changes, rather than by charge parameters. The range is fine-tuned for different surrounding rock grades: [0.7, 1.3] for Grade III surrounding rock, [0.65, 1.35] for Grade IV surrounding rock, and [0.6, 1.4] for Grade V surrounding rock, to adapt to the forming fluctuation characteristics of different surrounding rocks.

[0155] S54: Iteratively update the baseline explosive consumption.

[0156] Based on the statistical values ​​of the feedback compensation coefficients of all boreholes within the same functional area, the baseline explosive consumption per unit area for that functional area in the next blasting cycle is iteratively updated. In this embodiment, the statistical value is the arithmetic mean of the feedback compensation coefficients of all boreholes within the same functional area. Alternatively, the median, weighted average, or other statistical methods can be used depending on the on-site construction requirements. The iterative formula is as follows:

[0157] In the formula, This serves as the baseline explosive consumption per unit area for the next blasting cycle after iteration of this functional region, in units of... , This is the baseline explosive consumption per unit area for this functional region in this cycle, in units of ; This is the arithmetic mean of the feedback compensation coefficients for all boreholes within this functional area.

[0158] This formula is based on the adaptive iterative logic of closed-loop feedback of blasting effect. By using the feedback compensation coefficient of measured blasting effect, the reference explosive consumption is dynamically corrected, realizing the self-learning and self-optimization of blasting parameters as geological conditions change, which is in line with the iterative optimization method of tunnel blasting construction parameters.

[0159] Through this iterative approach, the system can automatically adapt to long-term drift in geological conditions along the tunnel excavation direction, continuously optimize blasting design benchmark parameters, and constantly improve blasting effects and explosive energy utilization.

[0160] After the completion of this blasting cycle, a 3D laser scanner was used to acquire the tunnel face contour scan data. The Dirichlet space partitioning method was used to divide the area into 128 local evaluation regions, and the over-excavation and under-excavation characteristics of each blast hole were extracted. Hole No. 1 is designed to bear the volume of rock mass. Actual molded volume measured Slight over-excavation occurred, and the feedback compensation coefficient was adjusted. It is within the constraint range of [0.7, 1.3]; Hole No. 50, designed to bear the rock mass volume. Actual molded volume measured Minor under-excavation occurred. It is within the constrained range; blast hole No. 100, minimum resistance line Actual excavation depth measured Minor under-excavation occurred. It is within the constraint range.

[0161] Statistical analysis shows that the average feedback compensation coefficient for all 16 boreholes in the blasting area of ​​this cycle is 0.96, the average for the 88 boreholes in the auxiliary area is 1.01, and the average for the 24 boreholes in the surrounding area is 1.02. Therefore, the baseline explosive consumption for the next blasting cycle is iteratively updated as follows: The baseline explosive consumption per unit area after the slotted region iteration ; The baseline explosive consumption after auxiliary region iteration ; The benchmark explosive consumption per unit area after iteration in the surrounding area ; Through this iterative update, the system automatically corrected the over-excavation trend in the slotted area and compensated for the under-excavation trend in the auxiliary and surrounding areas, thus achieving closed-loop self-optimization of blasting parameters.

[0162] To verify the practical application effect of the data-driven tunnel blasting parameter optimization method of this invention, a Class III surrounding rock section of a double-track railway hard rock tunnel was selected. Under complex conditions with heterogeneous lithological abrupt changes and hidden soft joint zones at the tunnel face, multi-cycle field excavation comparison tests were conducted using both the traditional static blasting design method and the method of this invention. During the tests, the single-hole comprehensive rock-breaking resistance inversion data and dynamic charge issuance instructions from the bottom layer of the system were retrieved, and the overall explosive consumption of each cycle was macroscopically calculated. The data were extracted and plotted as follows: Figure 2 and Figure 3 The comparison images shown are for reference only. like Figure 2As shown, the transient matching curve of the dynamic charge amount in the cut area fluctuating with local geological resistance within the same tunneling cycle is presented. Numerical trend observation reveals that traditional static methods, lacking a front-end sensing closed loop for the rock mass condition at the borehole level, output a uniform and fixed charge amount of 1.2 kg for all cut holes, failing to respond to the significant spatial fluctuations in the local dynamic rock mass strength index reflected in the histogram. In contrast, the dynamic charge amount output by this invention accurately tracks the fluctuations in the dynamic rock mass strength index. For example, in boreholes 6 and 7, the system detects a sudden drop in rock mass resistance and automatically issues a charge reduction command as low as approximately 0.75 kg, while in the hard boreholes 3 and 4, it automatically compensates to approximately 1.4 kg. This high-frequency follow-up of the underlying parameters ensures that the chemical explosion input energy of a single hole achieves a strict physical impedance match with the actual fracturing energy consumption threshold of the local rock mass, avoiding excessive explosion energy in soft rock areas and insufficient fracturing energy in hard rock areas from the energy source.

[0163] like Figure 3 As shown, a macroscopic comparison of the average explosive consumption per unit volume under multi-cycle operations is presented between the two methods. This is based on the dynamic optimization of hole-level parameters and the closed-loop feedback of the spatial grid after blasting. With the dual intervention of coefficient compensation, the average explosive consumption per unit volume of the traditional method is 1.32 kg / m³; while the method of this invention steadily reduces the average explosive consumption per unit volume to 1.15 kg / m³, achieving a reduction of approximately 12.9%. This decrease in macroscopic data reflects, from an engineering physics perspective, the effective elimination of ineffective redundant energy during the explosion process. The optimization strategy of this invention isolates the nonlinear interference of heterogeneous geological conditions on the single-hole rock breaking effect. Under the premise of ensuring sufficient rock fragmentation at the tunnel face and strict control of the excavation profile, it improves energy utilization and achieves effective reduction and refined control of engineering material costs.

[0164] This embodiment also provides a data-driven tunnel blasting parameter optimization system. The system includes a memory and a processor. The memory stores a computer program, and the processor executes the program to implement the steps of a data-driven tunnel blasting parameter optimization method. This system can be directly integrated into the onboard control system of a tunnel drilling rig, seamlessly connecting with the rig's native PLC system to achieve automatic acquisition of drilling parameters, real-time data calculation, and automatic output of blasting parameters. Alternatively, it can be used as a standalone portable laptop terminal for on-site blasting technicians, supporting the import of drilling and blasting design CAD files, manual or batch import of drilling parameters acquired by the rig, automatic completion of the entire calculation process, and output of blasting charge tables, adapting to the conventional operating modes of tunnel construction sites.

Claims

1. A data-driven method for optimizing tunnel blasting parameters, characterized in that, include: Obtain the functional area and hole layout parameters of each blast hole in the tunnel face, as well as the benchmark explosive consumption per unit area and the benchmark rock resistance parameters corresponding to the tunnel face; The operation data of the rock drilling equipment during the drilling of each blast hole is collected to determine the dynamic rock mass strength index of each blast hole. Based on the size distribution characteristics of the dynamic rock mass strength index of each blast hole relative to the rock mass resistance benchmark parameter, the strength anomaly area in the tunnel face is identified, and the weakened topological boundary enveloping the strength anomaly area is generated by combining the change gradient of the dynamic rock mass strength index. By combining the spatial location characteristics of each borehole relative to the weakened topological boundary and the dynamic rock mass strength index of each borehole, the local strength mutation coefficient of each borehole is determined. The reference explosive consumption of each functional area is corrected by using the local intensity mutation coefficient of each borehole, and the dynamic charge of each borehole is obtained by combining the spatial volume parameters obtained by analyzing the borehole layout parameters of each borehole, so as to achieve adaptive optimization of blasting parameters.

2. The tunnel blasting parameter optimization method according to claim 1, characterized in that, The dynamic rock mass strength index for each borehole is determined based on the following method: The benchmark parameters for rock mass resistance include the benchmark mechanical specific energy and the benchmark value of the Protodyakonov robustness coefficient of the surrounding rock; The actual mechanical specific energy is obtained by analyzing the operational data during the drilling process of each borehole; By using the relative ratio of actual mechanical specific energy to benchmark mechanical specific energy, the benchmark value of the Protodyakonov robustness coefficient of the surrounding rock is mapped and transformed to obtain the dynamic Protodyakonov robustness coefficient as a dynamic rock mass strength index.

3. The tunnel blasting parameter optimization method according to claim 2, characterized in that, The actual mechanical specific energy is obtained based on the analysis of operational data during the drilling process of each borehole, including: The operational data during the drilling of each borehole includes the rock drilling equipment's propulsion pressure, rotational speed, rotational torque, and axial drilling speed; Extract the axial linear extrusion power consumption generated by the propulsion pressure, and the circumferential rotational shear power consumption generated by the rotational speed and rotational torque; Based on the principle of energy conservation, the axial linear extrusion power consumption and the circumferential rotational shear power consumption are combined to obtain the actual mechanical specific energy.

4. The tunnel blasting parameter optimization method according to claim 2, characterized in that, Identify areas of strength anomalies in the tunnel face and generate weakened topological boundaries that enclose these areas by combining the gradient of dynamic rock mass strength indices, including: The coordinates of boreholes whose dynamic rock mass strength index is lower than the Protodyakonov robustness coefficient benchmark value in the rock mass resistance benchmark parameters are extracted, and spatial clustering is performed to generate multiple continuously distributed weakened node clusters. The spatial range corresponding to each weakened node cluster is determined as the strength anomaly region. The change gradient of dynamic rock mass strength index of adjacent boreholes in the tunnel face is statistically analyzed, and a dynamic identification threshold for characterizing abrupt changes in geological structure is constructed. The coordinates of boreholes whose change gradient is greater than the dynamic identification threshold are extracted, and the outermost coordinate points enveloping each weakened node cluster are selected from the extracted borehole coordinates. Geometric connection closure processing is performed to obtain the weakened topological boundary enveloping the strength anomaly region.

5. The tunnel blasting parameter optimization method according to claim 1, characterized in that, The local intensity abrupt change coefficient for each borehole is determined based on the following method: Determine the spatial position of each shot hole relative to the weakened topological boundary, identify shot holes located within the weakened topological boundary as anomalous shot holes, and identify shot holes located outside the weakened topological boundary as normal shot holes. For each anomalous borehole, the shortest topological geometric distance between each anomalous borehole and the weakened topological boundary is obtained, and the dynamic rock mass strength index of each anomalous borehole and the shortest topological geometric distance are subjected to feature fusion processing to determine the local strength mutation coefficient. For each normal borehole, the local intensity mutation coefficient is determined as a preset numerical constant.

6. The tunnel blasting parameter optimization method according to claim 5, characterized in that, Feature fusion processing is performed on the dynamic rock mass strength indices and the shortest topological geometric distance for each abnormal borehole, including: The statistical mean of the dynamic rock mass strength index of each normal blast hole is determined as the benchmark rock breaking resistance of the tunnel face. The ratio of the dynamic rock mass strength index of each abnormal borehole to the benchmark rock breaking resistance is used as the strength reduction factor to characterize the degree of material softening at the abnormal borehole. Based on the shortest topological geometric distance between each anomalous borehole and the weakened topological boundary, the clamping compensation factor used to characterize the clamping effect of the hard rock boundary on the soft interlayer is calculated. By using the clamping compensation factor, the strength reduction coefficient is corrected by reverse compensation to obtain the local strength mutation coefficient.

7. The tunnel blasting parameter optimization method according to claim 1, characterized in that, The dynamic charge amount for each borehole is determined as follows: Extract the minimum resistance line, borehole spacing, and borehole depth from the preset borehole layout parameters of each borehole; multiply the minimum resistance line, borehole spacing, and borehole depth together to obtain the burden rock volume as a spatial volume parameter. Apply a lower limit constraint to prevent under-drilling to the local strength mutation coefficient of each borehole, and use the constrained local strength mutation coefficient to correct the benchmark explosive consumption of the functional area to which each borehole belongs, so as to obtain the target explosive consumption of each borehole. The dynamic charge amount is determined by balancing calculations using the volume of the burdened rock mass and the unit consumption of the target explosive.

8. The tunnel blasting parameter optimization method according to claim 7, characterized in that, Obtaining the dynamic charge amount for each borehole also includes: The three-dimensional contour scan data of the tunnel face after blasting is obtained, and the three-dimensional contour scan data is divided into local evaluation areas based on the spatial position of each blast hole in order to extract the over-excavation and under-excavation characteristics of each blast hole. Based on the over-drilling and under-drilling characteristics of each blast hole, the feedback compensation coefficient of each blast hole is determined. Based on the statistical value of the feedback compensation coefficient of the blast holes in the same functional area, the benchmark explosive consumption of the corresponding functional area in the next blasting cycle is iteratively updated.

9. The tunnel blasting parameter optimization method according to claim 8, characterized in that, The feedback compensation coefficient for each borehole is determined based on the following method: The functional areas include: the slotting area, the auxiliary area, and the surrounding area; For blast holes whose functional areas are slotting areas or auxiliary areas, the feedback compensation coefficient is determined based on the ratio of the volume of the rock mass bearing the blast hole to the actual forming volume obtained by parsing the three-dimensional contour scanning data. For each blast hole whose functional area is the surrounding area, the feedback compensation coefficient is determined based on the ratio of the minimum resistance line of the blast hole to the actual excavation depth obtained by analyzing the three-dimensional contour scan data.

10. A data-driven tunnel blasting parameter optimization system, characterized in that, The tunnel blasting parameter optimization system includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the data-driven tunnel blasting parameter optimization method as described in any one of claims 1-9.

Images