A Method for Controlling Rock Blasting Effects and Semi-porosity Based on Artificial Intelligence Prediction
By collecting drilling rig mechanical response parameters in real time and using intelligent prediction models, the inflation pressure of PE pipes and the detonation delay time are adjusted in a coordinated manner. This solves the problem of unstable energy transfer caused by the heterogeneity of rock mass in traditional blasting design, and achieves precise control of the half-hole ratio and improved blasting effect.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA RAILWAY 19 TH BUREAU GROUP MINING IND INVESTMENT CO LTD
- Filing Date
- 2026-04-28
- Publication Date
- 2026-07-31
AI Technical Summary
Traditional blasting engineering design cannot adapt to the heterogeneity of rock masses, resulting in unstable blasting energy transfer efficiency, difficulty in accurately predicting half-porosity, lack of real-time data feedback and automatic adjustment mechanisms, and reliance on experience-based judgment.
By collecting drilling rig mechanical response parameters in real time, a borehole wall roughness model is established to simulate the contact state between the PE pipe and the borehole wall. Using a half-porosity intelligent prediction model and gradient optimization algorithm, the inflation pressure and detonation delay time are adjusted in a coordinated manner to generate precise blasting construction instructions.
It enables precise prediction and control of rock blasting effects, increases the semi-pore ratio, reduces vibration damage in the rock fracture zone, and improves blasting quality.
Smart Images
Figure CN122490893A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of blasting engineering technology, and in particular to a method for controlling the blasting effect and half-porosity of rock strata based on artificial intelligence prediction. Background Technology
[0002] In open-pit deep-hole bench blasting operations and geotechnical engineering excavation, the quality of smooth blasting at the perimeter holes directly affects the integrity of the preserved rock mass and the long-term stability of the slope. The half-hole ratio is a core quantitative indicator for measuring the effectiveness of smooth blasting and evaluating the level of damage control in the surrounding rock. Existing blasting engineering designs largely rely on macroscopic rock mass data obtained during geological exploration, combined with Protodyakonov coefficients or Sadowsky empirical formulas to determine borehole parameters, charge structure, and detonation sequence. This traditional design model is typically based on the idealized assumption that the rock mass is a homogeneous and isotropic medium. However, in actual engineering sites, the rock mass is widely characterized by invisible joints, fissures, karst structures, and weak interlayers. This high degree of heterogeneity leads to significant differences in the physical and mechanical properties of the rock at different borehole locations, making it difficult to adapt to complex and variable geological conditions using only static parameters from the design phase.
[0003] Furthermore, traditional decoupled charge structures typically ignore the borehole enlargement or reduction phenomena caused by lithological differences during drilling, making it difficult to quantitatively analyze the contact state between the charge medium and the borehole wall's micro-geometry. When uncertain air gaps or media filling exist between the charge and the rough, uneven borehole wall, the stress waves generated by the explosion will experience uncontrollable attenuation or reflection as they pass through the interface, leading to unstable energy transfer efficiency and making it highly prone to over-excavation in fractured rock areas or under-excavation in hard rock areas. Current construction control methods lack a dynamic feedback mechanism based on real-time data, making it impossible to accurately predict possible half-hole ratios before blasting, and also lacking a scientific decision-making method that can automatically and collaboratively adjust charge pressure and detonation time according to changes in geological conditions. As a result, the final blasting quality often relies on the experience and judgment of on-site personnel. Summary of the Invention
[0004] The purpose of this invention is to provide a method for controlling the rock blasting effect and semi-porosity based on artificial intelligence prediction, so as to solve the problems pointed out in the background art.
[0005] This invention provides a method for controlling the rock blasting effect and semi-porosity based on artificial intelligence prediction, the method comprising the following steps: Obtain the rock mass geological survey data and blasting design objectives of the area to be blasted, and determine the basic hole mesh parameters and PE pipe charging structure parameters; During the drilling operation, the drilling rig's mechanical response parameters are collected in real time, and these parameters are mapped to a sequence of single-hole rock mass mechanical specific work that varies with depth. The borehole inner wall was digitally reconstructed in three dimensions using the single-hole rock mass mechanical specific work sequence to establish a borehole wall roughness model. Based on the hole wall roughness model, the contact state between the PE pipe wall and the hole wall under a preset inflation pressure is simulated in a virtual environment, and the medium-rock acoustic impedance coupling index is calculated and generated. The medium-rock acoustic impedance coupling index, the perforation parameters, and the PE pipe charging structure parameters are input into a pre-trained intelligent prediction model for half-porosity, and the predicted half-porosity is output. Determine whether the predicted half-hole ratio meets the preset engineering quality standard. If not, adjust the inflation pressure of the PE pipe wall and the detonation delay time of the electronic detonator based on the gradient optimization algorithm until the predicted half-hole ratio meets the engineering quality standard, and generate the final blasting construction command.
[0006] Optionally, the mechanical response parameters while drilling include drilling pressure, slewing torque, drill rod speed and drilling speed during the drilling process; The step of mapping the drilling mechanical response parameters to a single-hole rock mass mechanical specific work sequence that varies with depth specifically includes: Based on the energy conservation principle of mechanical rock breaking, the mechanical work required to break a unit volume of rock is calculated using the drilling machine response parameters to obtain the original specific work data; The original specific work data is denoised using a moving window filtering algorithm to remove abnormal noise generated during the drill pipe replacement process, resulting in a single-hole rock mass mechanical specific work sequence that reflects the degree of rock mass fracture development.
[0007] Optionally, the basic mesh parameters are calculated and determined through the following steps: Set the borehole diameter to be The height of the step is Super deep The borehole inclination angle is ; According to row spacing Calculate the minimum resistance line ,satisfy ; The density coefficient is determined based on the degree of rock fracture development. And combined with the preset explosive consumption per unit and single-hole charge amount Joint calculation of hole spacing and row spacing To make it satisfy the energy balance equation: .
[0008] Optionally, the step of calculating the medium-rock acoustic impedance coupling index specifically includes: The degree of fit between the outer wall of the PE pipe and the roughness model of the borehole wall is analyzed, and the ratio of the effective contact area to the side surface area of the borehole is calculated and defined as the contact fit degree. Obtain the acoustic impedance values of the PE pipe material and the rock mass; Using the contact fit degree as a weighting factor, the stress wave transmission coefficient between the PE pipe material and the rock mass is corrected to obtain the medium-rock acoustic impedance coupling index, which characterizes the explosion energy transfer efficiency.
[0009] Optionally, the on-site construction requirements for PE pipes in the final blasting construction order include: Cut the PE pipe to a length 0.5m longer than the designed hole depth, and perform heat fusion sealing on the bottom of the PE pipe; When inserting the PE pipe into the borehole, a connecting pipe is used to assist in feeding the PE pipe into the bottom of the borehole along the borehole curve. An air compressor is used to inflate the PE pipe, and a pressure monitoring device is used to ensure that the air pressure inside the pipe reaches the adjusted inflation pressure value, so that the PE pipe forms a cylindrical loading cavity that is tightly coupled to the bore wall.
[0010] Optionally, the construction process of the intelligent prediction model for semi-porosity includes: Collect drilling data, PE pipe inflation parameters, and measured half-porosity images from historical blasting operations; The actual half-porosity value is extracted from the measured half-porosity image using image semantic segmentation technology; Construct a hybrid deep learning model that includes convolutional neural network layers and long short-term memory network layers; A physical constraint term is added to the loss function of the hybrid deep learning model, which restricts the prediction results to conform to the physical laws of explosion energy decay. The collected data is used to train the hybrid deep learning model until the model converges.
[0011] Optionally, in the step of collaboratively adjusting the gas pressure value of the PE pipe wall and the detonation delay time of the electronic detonator based on the gradient optimization algorithm, a multi-objective optimization strategy is adopted: Construct an objective function, wherein the first objective is to maximize the semi-porosity and the second objective is to minimize the explosive consumption. Using the medium-rock acoustic impedance coupling index as a constraint, when the coupling index is lower than a preset threshold, the inflation pressure of the PE pipe wall is increased first to improve the wall adhesion.
[0012] Optionally, the requirements regarding the detonation network in the final blasting command include: Digital electronic detonators are used for hole-by-hole detonation; The inter-hole delay time of the electronic detonator is set according to the adjusted detonation delay time. The value of the inter-hole delay time is controlled between 40ms and 75ms. The interference and vibration reduction effect generated by the micro-delay blasting is used to protect the remaining half-hole after blasting.
[0013] Optionally, the method further includes an adaptive modification step for the groundwater environment: When the single-hole rock mass mechanical specific work sequence identifies the presence of an aquifer within the borehole, a fluid-structure interaction correction coefficient is introduced when calculating the medium-rock acoustic impedance coupling index. The final blasting command instructs that, in the charge structure of the aquifer section, porous granular ammonium nitrate explosives or powdered finished explosives be selected instead of emulsion explosives, taking advantage of the water-proof properties of PE pipes, and the linear density of the charge be adjusted accordingly.
[0014] Optionally, the method further includes a closed-loop feedback step after the blast: After the blasting operation is completed, point cloud data of the actual half-pore ratio is collected using a 3D laser scanner; Calculate the deviation between the actual half-porosity and the predicted half-porosity; The deviation value and the actual geological parameters of this blasting are stored in the historical database, and the online incremental learning program of the intelligent half-porosity prediction model is triggered to update the model weights to adapt to the changes in mine lithology with mining depth.
[0015] The present invention has achieved the following beneficial effects: This invention establishes a mapping relationship between drilling mechanical response parameters and rock mechanical indices, enabling in-situ acquisition and digital reconstruction of rock mass geological information across the entire borehole length. This overcomes the problem of poor parameter adaptability caused by insufficient accuracy of geological models in traditional designs. Utilizing a PE pipe inflation pressure adjustment mechanism and a borehole wall roughness model, this invention constructs a quantifiable coupling index between the medium and rock acoustic impedance, making active control of explosion energy transfer efficiency possible and solving the problem of high randomness in energy dissipation in traditional uncoupled charges. Simultaneously, this invention combines physical information neural networks and gradient optimization algorithms to establish an advanced prediction and closed-loop optimization mechanism for blasting effects. By coordinating the adjustment of inflation pressure and detonation delay time, it achieves precise matching between construction parameters and rock mass characteristics. While ensuring that the rock fragment size meets the loading requirements, it effectively improves the half-hole ratio of smooth blasting and reduces vibration damage to the retained rock mass.
[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.
[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the rock blasting effect and semi-porosity control method based on artificial intelligence prediction in an embodiment of the present invention; Figure 2 This is a schematic diagram of the hardware and software structure of the digital blasting construction system in an embodiment of the present invention. Detailed Implementation
[0019] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.
[0020] This invention provides a method for controlling the rock blasting effect and semi-porosity based on artificial intelligence prediction, including: Step 1: Obtain the rock mass geological survey data and blasting design target of the area to be blasted, and determine the basic hole mesh parameters and PE pipe charging structure parameters; For acquiring rock geological survey data, data obtained by using a UAV equipped with an airborne LiDAR to conduct low-altitude close-range photogrammetry of the bench to be blasted was processed to obtain the strike, dip, and slope angle of the bench slope, as well as the dominant occurrence of exposed joints and fissures in the rock mass (including dip, dip angle, spacing, and trace length). Combined with rock rebound tests and point load tests conducted on site, as well as indoor mechanical test data from historical borehole cores, a rock physical and mechanical parameter database was established. The database covers the uniaxial compressive strength (UCS), uniaxial tensile strength (UTS), rock density, Protodyakonov hardness coefficient (f), and rock mass integrity coefficient (Kv). The blasting design objectives include: the design half-hole ratio must meet the preset engineering quality standards; the particle vibration velocity (PPV) of the retained rock mass must be strictly controlled within the safety threshold (e.g., less than 5 cm / s) to prevent blasting from inducing slope instability; and the block size distribution after rock fragmentation must meet the efficiency requirements of loading and transporting equipment to reduce the proportion of large blocks. For the basic parameters: Set the hole diameter as d (unit: mm), which is usually limited by the drill tool specifications of the on-site drill rig. For example, standard drill bits of 115 mm or 140 mm are selected. Set the bench height as H (unit: m), which is the vertical distance between the upper and lower bench platforms. Set the overburden depth as h (unit: m). To overcome the clamping effect of the rock at the bottom of the bench and offset the obstruction of gravity to the discharge of rock debris and prevent the formation of a bottom after blasting, it is usually taken as 5 to 10 times the hole diameter. Set the hole inclination angle as α (unit: degree). In smooth blasting, to obtain a smooth slope, the hole axis is usually arranged along the designed slope surface, and α is generally taken as 75 degrees to 90 degrees. Secondly, calculate the minimum resistance line W (unit: m). The minimum resistance line is the vertical distance from the center of the explosive charge to the free surface and is the most sensitive geometric parameter in blasting fragmentation. If W is too large, the blasting energy cannot overcome the rock resistance, resulting in large blocks or a bottom; if W is too small, flyrock and strong vibrations will occur, wasting energy. It is calculated and determined according to the geometric relationship formula W = b×sinα, where b is the row spacing. Thirdly, determine the density coefficient m. The density coefficient m is defined as the ratio of the hole spacing a to the row spacing b (m = a / b); configure the mapping rule of the rock mass integrity coefficient Kv: when the rock mass is intact (Kv > 0.75), take m = 0.8 - 1.0 to utilize the brittle fracture characteristics of the rock; when the rock mass is relatively intact (0.55 < Kv ≤ 0.75), take m = 0.7 - 0.8; when the rock mass is fractured (Kv ≤ 0.55), take m = 0.5 - 0.7. Finally, combined with the preset explosive consumption per unit volume q (kg / m³) and the charge per hole Q (kg), jointly calculate the hole spacing a and the row spacing b to satisfy the energy balance equation: Q / q = a×b×H. The physical essence of this energy balance equation is that the chemical energy released by the explosive loaded in a single hole must be sufficient to break the controlled rock volume borne by this hole. Among them, the explosive consumption per unit volume q is an empirical function that is positively correlated with the rock hardness coefficient f. By solving the system of equations regarding W, m, and energy balance simultaneously, the system can analytically calculate the uniquely determined basic hole spacing a and row spacing b.
[0021] For the charging structure parameters of the PE pipe, they include: material, outer diameter, wall thickness of the PE pipe, and the initially preset inflation pressure, etc.; the material can be selected as high-density polyethylene (HDPE) pipe.
[0022] Step 2: During the implementation of the drilling operation, real-time collect the mechanical response parameters of the drill rig during drilling and map the mechanical response parameters during drilling into a sequence of specific mechanical work of the single-hole rock mass varying with depth; Through the sensor group installed on the power head, propulsion beam, and hydraulic system of the drill rig, real-time collect the mechanical response parameters of the drill rig during drilling. The mechanical response parameters during drilling include: drilling pressure, rotary torque, drill pipe rotation speed, and drilling speed; the specific mapping steps are as follows: Based on the energy conservation principle of mechanical rock breaking, the mechanical work required to break a unit volume of rock is calculated using the drilling machine response parameters, yielding the original specific work data. The calculation formula is as follows:
[0023] in, Let MSE be the cross-sectional area of the borehole. The first term in the formula represents the indentation work density done by the axial force, and the second term represents the cutting work density done by the rotational torque. The calculated MSE value (in MPa) shows a significant positive correlation with the uniaxial compressive strength of the rock. A higher MSE value indicates that the rock is harder and denser, requiring more energy to break; a lower MSE value indicates that the rock is softer or more fractured.
[0024] The above-mentioned mechanical downhole response parameters have been standardized in terms of dimensions, and the specific calculation formulas have been revised as follows:
[0025] In the formula, The specific mechanical work of a single-hole rock mass is calculated in megapascals (MPa). The drilling pressure applied to the drilling rig is input in kilonewtons (kN). Multiplying by 1000 in the formula converts it to newtons (N). The cross-sectional area of the borehole is expressed in square millimeters. ), determined by the drill bit diameter Calculation yields ( ); Pi, with a value of 3.14159; The torque of the drill pipe is expressed in Newton-meters (N·m). This is the drill pipe rotation speed; the unit is revolutions per minute (r / min). For the drilling speed (ROP), enter the unit as meters per minute (m / min).
[0026] Step 3: Use the single-hole rock mass mechanical specific work sequence to perform digital three-dimensional reconstruction of the borehole inner wall and establish a borehole wall roughness model; Based on extensive field borehole inspection data (using borehole cameras to obtain the actual borehole diameter), a correlation between mechanical specific work (MSE) and borehole diameter enlargement coefficient was established through machine learning regression analysis. The nonlinear mapping relationship between them is expressed as:
[0027] In the formula, For depth The aperture enlargement factor at that location is a dimensionless ratio, and ; The nominal aperture reference constant represents the nominal aperture under ideal conditions; The maximum hole enlargement disturbance factor is dimensionless and ranges from 0.2 to 0.4. This parameter characterizes the ultimate destructive capability of the drilling rig's mechanical vibration on the hole wall. It is a natural exponential function; The lithological sensitivity index is expressed in inverted megapascals (MPA). The value ranges from 0.05 to 0.1, and is used to control the hole expansion coefficient as a function of rock strength ( The rate at which the value increases and decreases, and ensures that the exponent term is dimensionless; For depth The mechanical specific work value calculated at the drill bit is expressed in megapascals (MPa). A lower MSE value (indicating rock fracturing) indicates greater drill bit disturbance and more severe scouring by the slag discharge airflow. The higher the value (i.e., the more severe the hole enlargement); the higher the MSE value (indicating rock integrity). It approaches 1.0. Therefore, the actual aperture sequence varying with depth can be obtained. .
[0028] Secondly, parametric modeling is performed in virtual 3D space. Using voxelization or parametric surface lofting techniques, a series of annular slices with different diameters are generated, with the borehole center trajectory (usually assumed to be a straight line, or corrected to a curve based on inclination data) as the axis. The diameter of each slice is controlled by a D(Z) sequence.
[0029] Step 4: Based on the borehole wall roughness model, simulate the contact state between the PE pipe wall and the borehole wall under a preset inflation pressure in a virtual environment, and calculate the medium-rock acoustic impedance coupling index. When establishing the flexible finite element model of the PE pipe, the configured properties include: Young's modulus (approximately 0.8~1.2 GPa), Poisson's ratio (approximately 0.4), yield strength, and material density. A preset inflation pressure is then applied inside the PE pipe. (e.g., 0.2MPa~0.6MPa). The simulation uses a nonlinear contact algorithm (such as the penalty function method or the Lagrange multiplier method).
[0030] The total area of the nodes where the outer surface of the PE pipe physically contacts the rock borehole wall is calculated, and the ratio of this area to the total area of the borehole side surface is defined as the contact fit degree. This parameter It is a scalar between 0 and 1 that varies with depth Z, reflecting the geometric coupling quality of the charge structure.
[0031] Obtain the acoustic impedance value of PE pipe from the material library. For the rock mass, the depth-varying acoustic impedance sequence of the rock mass is inverted using the MSE sequence from step two through empirical formulas. Typically, the acoustic impedance of rock is much greater than that of PE pipe, and even greater than that of air.
[0032] According to the stress wave propagation theory, when a wave is incident perpendicularly from medium 1 to medium 2, the transmission coefficient... ,in For the acoustic impedance of medium 1, The acoustic impedance of medium 2.
[0033] In the contact model, there are two typical energy transfer paths: Path A (contact zone): explosive energy → PE pipe → rock. The transmission efficiency of this path is determined by... and The degree of matching is determined. Path B (non-contact zone / air gap): explosive energy → PE pipe → air → rock. Due to the extremely low acoustic impedance of air, the energy in this path is almost totally reflected, and the energy transmitted into the rock mass is negligible. Therefore, a weighted correction model was constructed to calculate the medium-rock acoustic impedance coupling index, which characterizes the energy transfer efficiency of the explosion. The calculation formula is as follows: in, The contact fit varies with depth; the value in square brackets is the ideal transmittance coefficient at the interface between the PE pipe material and the rock. A dimensionless correction factor is used to account for the attenuation effect of PE pipe wall thickness. The correction factor... The specific calculations are based on the acoustic transmission theory of thin-walled cylindrical shells, using the following exponential attenuation formula:
[0034] In the formula, Energy transfer correction factor, range of values ; The material's acoustic damping constant is dimensionless. The selected high-density polyethylene material, under inflated stress conditions... The value range is set to 2.5 to 3.0; The wall thickness of the PE pipe is expressed in millimeters (mm). The outer radius of the PE pipe is expressed in millimeters (mm). ; reflects the change in relative pipe wall thickness ( With the increase of ), the energy dissipation of high-frequency explosive stress waves when penetrating the dielectric layer increases exponentially, and the correction factor... The pressure decreases accordingly. By adjusting the inflation pressure, it can be forcibly increased. This linearly increases the ICI value; Step 5: Input the medium-rock acoustic impedance coupling index, pore network parameters, and PE pipe charging structure parameters into the pre-trained intelligent prediction model for half-porosity, and output the predicted half-porosity. To ensure the model's generalization ability with small samples, a total loss function that includes a physical regularization term is used. Its definition is as follows:
[0035]
[0036] In the formula, This represents the total loss value during neural network training. The data-driven loss is calculated using the mean square error (MSE) between the predicted half-pore ratio and the measured label value; This is the weighting coefficient for the physical constraint term, which is dimensionless and set to 0.15 in this embodiment. This represents the loss value for the physical constraint term; This represents the total number of sample data in the training batch. The index number of the sample data; The linear rectified activation function is defined as follows: ; To calculate the first using automatic differentiation techniques The partial derivative of the predicted semi-porosity of a sample with respect to the medium-rock acoustic impedance coupling index.
[0037] According to explosive physics, the coupling index The improvement should promote the semi-porosity The improvement (i.e., the derivative should) When the model output violates this physical law (i.e., the derivative is negative), the combination of the negative sign and the ReLU function will produce a positive penalty loss, forcing the neural network's weight update direction to revert to the parameter space that conforms to physical logic; Step 6: Determine whether the predicted half-hole ratio meets the preset engineering quality standard. If not, adjust the inflation pressure of the PE pipe wall and the detonation delay time of the electronic detonator based on the gradient optimization algorithm until the predicted half-hole ratio meets the engineering quality standard, and generate the final blasting construction command. The predicted semi-pore ratio is compared with the quality standards specified in the engineering contract; if it meets the requirements, the current basic parameter scheme is adopted; otherwise, model optimization is performed. A multi-objective optimization strategy is employed, the core of which is the synthesis objective function. By searching The optimal parameters are determined by finding the minimum value of the parameter set. The comprehensive objective function is expressed as:
[0038] In the formula, The vector of decision parameters to be optimized includes the PE pipe inflation pressure and the electronic detonator detonation delay time. The predicted semi-porosity (a decimal between 0 and 1) is the output of the model. This represents the explosive consumption per unit volume under the current parameters, expressed in kilograms per cubic meter. ); This represents the maximum allowable explosive consumption per unit volume for the project, expressed in kilograms per cubic meter. ); This is a vibration penalty term, which is set to 0 when the predicted vibration velocity is less than the safety threshold, and otherwise takes the maximum penalty value (e.g., ...). ); The weighting coefficients for each objective, for example: set to , , .
[0039] Calculate the gradient vector of the comprehensive objective function with respect to the input variables, and use the Adam optimizer for iterative optimization.
[0040] After multiple rounds of iterative optimization, until the predicted half-pore ratio converges to the optimal value that meets the engineering quality standards, the final blasting construction command is generated. Step 7: Implement refined on-site operations based on the final blasting command; For the pretreatment and insertion of PE pipes: Cut high-density polyethylene (HDPE) pipes to the appropriate length. The pipe length should be 0.5 meters longer than the designed hole depth, with the excess used for hole fixing and inflation. A hot-melt sealing process is used: A portable hot-melt machine is used to heat the bottom of the PE pipe to a molten state (approximately 220°C), then pressure is applied to close it, forming a sealed bottom integrated with the pipe body, effectively preventing gas leakage under high-pressure inflation. When inserting the PE pipe into the borehole, due to its flexibility, it is prone to buckling or getting stuck in cracks during deep hole operations (over 15 meters). Therefore, a dedicated rigid guide pipe can be used to assist in lowering the pipe. The PE pipe is fitted onto a guide pipe with an internal steel wire skeleton and smoothly fed into the bottom of the borehole along the borehole curve. Once in place, the guide pipe is withdrawn.
[0041] Use a mobile air compressor or high-pressure gas cylinder group to fill the PE pipe with air; finally, load the medicine.
[0042] Step 8: Adaptive modification steps for the groundwater environment; When an aquifer is identified, the algorithm automatically incorporates a fluid-structure interaction correction coefficient when calculating the medium-rock acoustic impedance coupling index. The final blasting command instructs specific parameter adjustments for the aquifer: increasing the inlet pressure to slightly exceed the external hydrostatic pressure, maintaining the pipe shape, and / or, appropriately reducing the charge linear density (reducing the charge quantity) for the corresponding PE pipe location within the aquifer. This segmented, differentiated charging approach adapts to complex hydrogeological conditions. Step Nine: Closed-loop feedback and model self-evolution after the explosion; After the blasting is carried out, the data from the panoramic scan of the newly exposed slope surface after the blasting is processed using a high-precision 3D laser scanner. The original MWD drilling data, the final set pressure and delay parameters, and the measured half-hole ratio results are then packaged into a new training sample for the model's self-evolution training.
[0043] To implement the method disclosed in this invention, the following methods can be employed: Figure 2 The system shown mainly includes, at the hardware level: an intelligent hydraulic down-the-hole drilling rig equipped with a high-frequency drilling monitoring sensor group, an explosion-proof industrial computer for on-site edge computing, a high-precision 3D laser scanner, a PE pipe inflation system with a pressure monitoring feedback device, and a digital electronic detonator detonation control terminal.
[0044] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
Claims
1. A method for controlling the blasting effect and semi-porosity of rock strata based on artificial intelligence prediction, characterized in that, The method includes the following steps: Obtain the rock mass geological survey data and blasting design objectives of the area to be blasted, and determine the basic hole mesh parameters and PE pipe charging structure parameters; During the drilling operation, the drilling rig's mechanical response parameters are collected in real time, and these parameters are mapped to a sequence of single-hole rock mass mechanical specific work that varies with depth. The borehole inner wall was digitally reconstructed in three dimensions using the single-hole rock mass mechanical specific work sequence to establish a borehole wall roughness model. Based on the hole wall roughness model, the contact state between the PE pipe wall and the hole wall under a preset inflation pressure is simulated in a virtual environment, and the medium-rock acoustic impedance coupling index is calculated and generated. The medium-rock acoustic impedance coupling index, the perforation parameters, and the PE pipe charging structure parameters are input into a pre-trained intelligent prediction model for half-porosity, and the predicted half-porosity is output. Determine whether the predicted half-hole ratio meets the preset engineering quality standard. If not, adjust the inflation pressure of the PE pipe wall and the detonation delay time of the electronic detonator based on the gradient optimization algorithm until the predicted half-hole ratio meets the engineering quality standard, and generate the final blasting construction command.
2. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, The mechanical response parameters while drilling include drilling pressure, slewing torque, drill pipe rotation speed, and drilling speed during the drilling process; The step of mapping the drilling mechanical response parameters to a single-hole rock mass mechanical specific work sequence that varies with depth specifically includes: Based on the energy conservation principle of mechanical rock breaking, the mechanical work required to break a unit volume of rock is calculated using the drilling machine response parameters to obtain the original specific work data; The original specific work data is denoised using a moving window filtering algorithm to remove abnormal noise generated during the drill pipe replacement process, resulting in a single-hole rock mass mechanical specific work sequence that reflects the degree of rock mass fracture development.
3. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, The basic mesh parameters are determined by the following steps: Set the borehole diameter to be The height of the step is Super deep The borehole inclination angle is ; According to row spacing Calculate the minimum resistance line ,satisfy ; The density coefficient is determined based on the degree of rock fracture development. And combined with the preset explosive consumption per unit and single-hole charge amount Joint calculation of hole spacing and row spacing To make it satisfy the energy balance equation: .
4. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, The specific steps for calculating the medium-rock acoustic impedance coupling index include: The degree of fit between the outer wall of the PE pipe and the roughness model of the borehole wall is analyzed, and the ratio of the effective contact area to the side surface area of the borehole is calculated and defined as the contact fit degree. Obtain the acoustic impedance values of the PE pipe material and the rock mass; Using the contact fit degree as a weighting factor, the stress wave transmission coefficient between the PE pipe material and the rock mass is corrected to obtain the medium-rock acoustic impedance coupling index, which characterizes the explosion energy transfer efficiency.
5. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, The on-site construction requirements for PE pipes in the final blasting operation order include: Cut the PE pipe to a length 0.5m longer than the designed hole depth, and perform heat fusion sealing on the bottom of the PE pipe; When inserting the PE pipe into the borehole, a connecting pipe is used to assist in feeding the PE pipe into the bottom of the borehole along the borehole curve. An air compressor is used to inflate the PE pipe, and a pressure monitoring device is used to ensure that the air pressure inside the pipe reaches the adjusted inflation pressure value, so that the PE pipe forms a cylindrical loading cavity that is tightly coupled to the bore wall.
6. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, The construction process of the intelligent prediction model for semi-porosity includes: Collect drilling data, PE pipe inflation parameters, and measured half-porosity images from historical blasting operations; The actual half-porosity value is extracted from the measured half-porosity image using image semantic segmentation technology; Construct a hybrid deep learning model that includes convolutional neural network layers and long short-term memory network layers; A physical constraint term is added to the loss function of the hybrid deep learning model, which restricts the prediction results to conform to the physical laws of explosion energy decay. The collected data is used to train the hybrid deep learning model until the model converges.
7. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, In the step of collaboratively adjusting the gas pressure value of the PE pipe wall and the detonation delay time of the electronic detonator based on the gradient optimization algorithm, a multi-objective optimization strategy is adopted: Construct an objective function, wherein the first objective is to maximize the semi-porosity and the second objective is to minimize the explosive consumption. Using the medium-rock acoustic impedance coupling index as a constraint, when the coupling index is lower than a preset threshold, the inflation pressure of the PE pipe wall is increased first to improve the wall adhesion.
8. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, The requirements regarding the detonation network in the final blasting command include: Digital electronic detonators are used for hole-by-hole detonation; The inter-hole delay time of the electronic detonator is set according to the adjusted detonation delay time. The value of the inter-hole delay time is controlled between 40ms and 75ms. The interference and vibration reduction effect generated by the micro-delay blasting is used to protect the remaining half-hole after blasting.
9. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 4, characterized in that, The method also includes adaptive modification steps for the groundwater environment: When the single-hole rock mass mechanical specific work sequence identifies the presence of an aquifer within the borehole, a fluid-structure interaction correction coefficient is introduced when calculating the medium-rock acoustic impedance coupling index. The final blasting command instructs that, in the charge structure of the aquifer section, porous granular ammonium nitrate explosives or powdered finished explosives be selected instead of emulsion explosives, taking advantage of the water-proof properties of PE pipes, and the linear density of the charge be adjusted accordingly.
10. The method for controlling rock blasting effect and semi-porosity based on artificial intelligence prediction according to claim 1, characterized in that, The method also includes a closed-loop feedback step after the blast: After the blasting operation is completed, point cloud data of the actual half-pore ratio is collected using a 3D laser scanner; Calculate the deviation between the actual half-porosity and the predicted half-porosity; The deviation value and the actual geological parameters of this blasting are stored in the historical database, and the online incremental learning program of the intelligent half-porosity prediction model is triggered to update the model weights to adapt to the changes in mine lithology with mining depth.