Method for determining blasting scheme considering geological structure and blasting effect

Abstract

The application relates to a blasting scheme determination method considering a geological structure and a blasting effect, comprising the following steps: acquiring hole network parameters, a charging structure and a charging amount of a blast hole for blasting a region to be blasted; determining three-dimensional coordinates of the blast hole according to the hole network parameters; determining a target blast hole with a geological structure and a first geological structure of the target blast hole based on drilling parameters; collecting an image of the target blast hole, identifying a second geological structure of the image, and establishing a three-dimensional geological structure model based on the first geological structure, the second geological structure and geological information of the region to be blasted; optimizing the charging structure and the charging amount based on the three-dimensional geological structure model; acquiring blasting characteristics after blasting the region to be blasted; training a prediction model based on an optimized scheme, the hole network parameters, the blasting characteristics and the three-dimensional geological structure model to obtain an optimal blasting scheme prediction model. The method can accurately identify the geological structure, accurately evaluate the blasting effect and output the optimal blasting scheme, and the safety of open pit slope deep hole blasting is improved.

Description

Technical Field

[0001] This application relates to the field of blasting technology, and in particular to a method for determining blasting schemes that takes into account geological structure and blasting effects. Background Technology

[0002] Deep-hole blasting of open-pit slopes is a commonly used and efficient fracturing method in geotechnical engineering. Its blasting effect directly affects slope stability, loading efficiency, and construction safety. In actual engineering projects, geological structures such as joints, fissures, and faults are widely distributed in rock masses, significantly influencing block size distribution, blasting vibration, blast pile morphology, and blast smoke characteristics. Failure to accurately identify and fully consider these factors can lead to poor blasting control, increased flyrock, and a higher proportion of large blocks. However, existing blasting design methods generally rely on empirical parameters, and traditional blasting effect evaluation methods are characterized by single indicators and strong subjectivity, making it difficult to meet the current demand for refined and intelligent blasting management in construction processes.

[0003] In recent years, deep learning technology has been widely applied in the field of intelligent engineering, demonstrating significant advantages, particularly in modeling complex nonlinear relationships, automatic feature extraction, and adaptive decision-making. Introducing deep learning into the blasting design and optimization process can fully explore the deep-seated correlation between geological structural parameters and blasting effects, avoiding the problems of subjective parameter selection and poor adaptability in traditional models. However, existing blasting design methods lack self-learning capabilities and struggle to perform intelligent optimization based on historical data, limiting the intelligence, multidimensionality, and efficiency of blasting design. Therefore, there is an urgent need to establish a method for determining blasting schemes that integrates multi-source sensing technology and deep learning, considering both geological structure and blasting effects. Summary of the Invention

[0004] Therefore, it is necessary to provide a method for determining blasting schemes that considers geological structure and blasting effect, which can accurately identify geological structure, accurately evaluate blasting effect, output optimal blasting scheme, and improve the safety of deep hole blasting on open slopes, in order to address the above-mentioned technical problems.

[0005] A method for determining a blasting scheme that considers geological structure and blasting effect, the method comprising:

[0006] S1. Obtain the hole network parameters, the charge structure and charge amount of each blast hole for the area to be blasted, and determine the three-dimensional coordinates of the blast holes based on the hole network parameters.

[0007] S2. Obtain the drilling parameters for drilling the blast hole according to the three-dimensional coordinates, and based on the drilling parameters, determine the target blast hole with a geological structure and the first geological structure of the target blast hole;

[0008] S3. Acquire images of the target blast holes, identify the second geological structure in the images, and establish a three-dimensional geological structure model based on the first geological structure, the second geological structure, and the geological information of the area to be blasted.

[0009] S4. Based on the three-dimensional geological structure model, optimize the charge structure and the charge quantity to obtain an optimized solution;

[0010] S5. Obtain multiple blasting features after blasting the area to be blasted using the optimization scheme; the blasting features are used to determine the blasting effect;

[0011] S6. Based on the optimization scheme, the hole mesh parameters, the blasting characteristics, and the three-dimensional geological structure model, the prediction model used to predict the blasting characteristics is trained to obtain the optimal blasting scheme prediction model.

[0012] The aforementioned method for determining blasting schemes, considering both geological structure and blasting effects, acquires the borehole network parameters, charge structure, and charge quantity for each borehole in the area to be blasted. Based on the borehole network parameters, it determines the three-dimensional coordinates of the boreholes, obtains drilling parameters for drilling the boreholes according to these three-dimensional coordinates, and, based on these parameters, identifies the target boreholes with existing geological structures and their primary geological structures. Images of the target boreholes are acquired, and secondary geological structures within the images are identified. This allows for precise identification of the geological structure of the area to be blasted, enabling the establishment of a relatively accurate three-dimensional geological structure model based on the primary and secondary geological structures and the geological information of the area to be blasted. By optimizing the charge structure and charge quantity based on this three-dimensional geological structure model, an optimized scheme is obtained. Multiple blasting characteristics after blasting the area using the optimized scheme are acquired, allowing for multi-dimensional evaluation of the blasting effect and improving its accuracy. By training a prediction model for blasting characteristics based on an optimized scheme, borehole parameters, blasting characteristics, and a three-dimensional geological structure model, an optimal blasting scheme prediction model is obtained. This allows the prediction model to continuously learn the intrinsic relationship between the geological structure, optimized scheme, borehole parameters, and blasting characteristics, outputting accurate predictions of blasting characteristics. Based on these predictions, the optimal blasting scheme can be determined, thereby improving the safety of deep-hole blasting on open-pit slopes.

[0013] In one embodiment, step S1 includes:

[0014] Based on the geological survey report of the area to be blasted, the geological information of the area to be blasted is determined; the geological information includes stratigraphic lithology and macroscopic geological structure information, and the macroscopic geological structure information includes the distribution location, scale and orientation of each geological structure in the area to be blasted;

[0015] Obtain the rock mechanics parameters and topographic parameters of the area to be blasted;

[0016] Based on the geological information, the rock mechanics parameters, and the topographic parameters, the hole network parameters, the charge structure of each borehole, and the charge amount are determined for the area to be blasted.

[0017] In this embodiment, the geological information of the area to be blasted is determined based on the geological survey report of the area to be blasted, and the rock mechanics parameters and topographic parameters of the area to be blasted are obtained. Based on the geological information, rock mechanics parameters, and topographic parameters, the hole pattern parameters, the charge structure and charge amount of each blast hole are determined. This makes the hole pattern parameters, the charge structure and charge amount of each blast hole more accurate, which helps to achieve a reasonable distribution of blasting energy, improve the uniformity of rock fragmentation, reduce the proportion of large pieces and the base, and improve blasting efficiency and subsequent loading and unloading efficiency.

[0018] In one embodiment, step S2 includes:

[0019] When drilling boreholes according to the three-dimensional coordinates, the drilling parameters of each borehole are obtained; the drilling parameters include at least the thrust variation curve, rotation speed variation curve, and torque variation curve of the borehole drilling equipment.

[0020] The curve change characteristics of the top thrust change curve, the rotational speed change curve, and the torque change curve are determined, and a preset geological structure mapping library is obtained; the preset geological structure mapping library includes multiple preset curve change characteristics and the geological structure associated with each preset curve change characteristic;

[0021] When a first target feature exists in the preset curve change features that matches the curve change features of the borehole, the borehole is determined to be a target borehole;

[0022] The geological structure associated with the first target feature is identified as the first geological structure of the target borehole.

[0023] In this embodiment, by utilizing the curve variation characteristics of drilling parameters and combining them with a preset geological mapping database, the real-time identification and positioning of geological structures traversed by the borehole drilling equipment can be achieved, thereby identifying target boreholes with geological structures and improving the accuracy of the geological structure of the area to be blasted.

[0024] In one embodiment, step S3 includes:

[0025] Images of the target boreholes are acquired and input into a recognition model for identifying geological structures to obtain the second geological structure of the target boreholes;

[0026] When the first geological structure and the second geological structure of the target blast hole do not match, the second geological structure is determined as the geological structure of the target blast hole, and the three-dimensional spatial coordinates of the geological structure of the target blast hole are obtained;

[0027] Taking any of the target blast holes as the target object, the horizontal distance and elevation distance between the target object and the associated blast hole are determined based on the three-dimensional spatial coordinates of the geological structure of the target object and the three-dimensional spatial coordinates of the geological structure of the associated blast hole; the associated blast hole is a blast hole among the adjacent blast holes of the target object that has the same geological structure type as the target object.

[0028] When the horizontal distance is less than a first threshold, the elevation distance is less than a second threshold, and the geological structure of the target object and the associated borehole are aligned, a geological structure association result is obtained in which the geological structure of the associated borehole and the target object is the same. Each target borehole is traversed to obtain the geological structure association result of each target borehole.

[0029] Based on the geological information of the area to be blasted and the correlation results of the geological structure, a three-dimensional geological structure model of the area to be blasted is established.

[0030] In this embodiment, when the first and second geological structures of the target blast hole do not match, the second geological structure is determined as the geological structure of the target blast hole. This reduces geological identification errors and improves the accuracy of geological structure judgment. Using any target blast hole as the target object, the horizontal and vertical distances between the target object and the associated blast holes are determined based on their three-dimensional spatial coordinates and the three-dimensional spatial coordinates of their geological structures. When the horizontal distance is less than a first threshold, the vertical distance is less than a second threshold, and the geological structures of the target object and the associated blast holes are aligned, a geological structure association result is obtained, indicating that the geological structures of the associated blast holes and the target object are of the same structure. This process is repeated for each target blast hole to obtain the geological structure association results for each target blast hole. This allows for a scientific determination of spatial continuity, extension, and geological structure, achieving a leap from "point-like identification" to "line / area structure deduction." By establishing a three-dimensional geological structure model of the area to be blasted based on the geological information and geological structure association results, the accuracy of the three-dimensional geological structure model is significantly higher than that of traditional geological exploration inference models, facilitating refined guidance for blasting design.

[0031] In one embodiment, step S5 includes:

[0032] Image acquisition is performed on the blasted area after the blasting of the area to be blasted, and blasting images are obtained;

[0033] Based on the blasting images, blasting block size analysis is performed to obtain the blasting block size distribution characteristics of the blasting area; the blasting block size distribution characteristics include median block size, blasting index, and large block proportion;

[0034] The median block size, the blasting index, and the proportion of large blocks are determined as blasting characteristics.

[0035] In this embodiment, images of the blasted area after blasting are acquired to obtain blasting images. Based on the blasting images, blasting block size analysis is performed to obtain the blasting block size distribution characteristics of the blasting area. The blasting block size distribution characteristics include median block size, blasting index, and large block ratio. The median block size, blasting index, and large block ratio are determined as blasting characteristics, so that the blasting effect can be accurately evaluated based on the median block size, blasting index, and large block ratio.

[0036] In one embodiment, step S5 is characterized by comprising:

[0037] Pre-set blasting vibration monitoring points, and collect vibration velocity data from the blasting vibration monitoring points during the blasting process;

[0038] Regression analysis was performed on the vibration velocity data to obtain the peak vibration velocity, dominant vibration frequency, and vibration duration at each of the blasting vibration monitoring points.

[0039] The peak vibration velocity, the dominant vibration frequency, and the duration of vibration are defined as the blasting characteristics.

[0040] In this embodiment, by pre-setting blasting vibration monitoring points, vibration velocity data is collected from these points during the blasting process. Regression analysis is then performed on the vibration velocity data to obtain the peak vibration velocity, dominant vibration frequency, and vibration duration at each monitoring point. This reduces the need for manual operation and improves work efficiency. By defining the peak vibration velocity, dominant vibration frequency, and vibration duration as blasting characteristics, accurate assessment of the blasting effect can be achieved based on these parameters.

[0041] In one embodiment, step S5 includes:

[0042] Obtain the point cloud data of the blast pile after the blasting of the area to be blasted;

[0043] Based on the point cloud data of the blast pile and the three-dimensional geological structure model, the throwing angle, accumulation height, and loosening coefficient are determined.

[0044] The throwing angle, the accumulation height, and the loosening coefficient are determined as blasting characteristics.

[0045] In this embodiment, by acquiring point cloud data of the blast pile and combining it with a three-dimensional geological structure model, key parameters such as throwing angle, accumulation height, and loosening coefficient can be accurately calculated. This enables a systematic and quantitative description of the rock mass movement and accumulation state after blasting, overcoming the subjectivity and inaccuracy of traditional experience-based judgment.

[0046] In one embodiment, step S5 includes:

[0047] During the blasting process in the area to be blasted, the control and acquisition device collects the concentration changes and diffusion trajectory of harmful gases generated during the blasting process, as well as the aerodynamic diameter of particulate matter, according to the preset flight route and flight altitude.

[0048] The concentration changes, the diffusion trajectory, and the aerodynamic diameter of the particles are determined as the blasting characteristics.

[0049] In this embodiment, during the blasting process in the area to be blasted, the data acquisition device is controlled to collect the concentration changes and diffusion trajectories of harmful gases generated during the blasting process, as well as the aerodynamic diameter of particulate matter, according to a preset flight path and altitude. This allows for real-time capture of the changes in pollutants in both spatial and temporal dimensions, overcoming the limitations of traditional fixed-point, static monitoring. By defining the concentration changes, diffusion trajectories, and aerodynamic diameter of particulate matter as blasting characteristics, accurate assessment of the blasting effect can be achieved based on these factors.

[0050] In one embodiment, step S6 includes:

[0051] Data augmentation processing is performed on the optimization scheme, the borehole parameters, the blasting characteristics, and the three-dimensional geological structure model to obtain multiple training samples;

[0052] The training samples are used to train the prediction model for predicting blasting features, and the optimal blasting scheme prediction model is obtained after training.

[0053] Step S6 is followed by:

[0054] The current geological structure information of the area to be predicted and each blasting scheme for the area to be predicted are input into the prediction model to obtain the predicted blasting characteristics of each blasting scheme.

[0055] Select a second target feature from the predicted blasting features that matches the preset blasting features, and determine the blasting scheme corresponding to the second target feature as the optimal blasting scheme.

[0056] In this embodiment, data augmentation technology is used to expand the training samples. Combined with multi-source heterogeneous data such as 3D geological structure models, borehole parameters, optimization schemes, and blasting characteristics, the prediction model is systematically trained, realizing a shift from "experience-driven" to "data and model-driven" approaches, thus improving the intelligence and scientific level of blasting prediction. By matching and filtering predicted blasting characteristics with preset blasting characteristics, the system automatically identifies the "second target characteristic" that best meets engineering requirements and its corresponding blasting scheme, achieving intelligent recommendation and optimal decision-making for blasting schemes, improving selection efficiency and objectivity. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating a method for determining a blasting scheme that considers geological structure and blasting effects in one embodiment.

[0058] Figure 2 This is a schematic diagram of the exposed surface captured in one embodiment;

[0059] Figure 3 This is a schematic diagram illustrating the identification of block size in one embodiment;

[0060] Figure 4 This is a block size distribution diagram after blasting in one embodiment;

[0061] Figure 5 This is a schematic diagram of vibration velocity data in the X direction in one embodiment;

[0062] Figure 6 This is a schematic diagram of vibration velocity data in the Y direction in one embodiment;

[0063] Figure 7 This is a schematic diagram of vibration velocity data in the Z direction in one embodiment;

[0064] Figure 8 This is a schematic diagram of scanning the blast area in one embodiment;

[0065] Figure 9 This is a schematic diagram of the overall process for determining a blasting scheme that considers geological structure and blasting effect in one embodiment. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0067] In one embodiment, such as Figure 1 As shown, a method for determining a blasting scheme that considers geological structure and blasting effect is provided, including the following steps:

[0068] S1. Obtain the hole network parameters, the charge structure and charge amount of each blast hole for the area to be blasted, and determine the three-dimensional coordinates of the blast holes based on the hole network parameters.

[0069] The hole grid parameters include hole spacing, row spacing, and resistance line. Hole spacing is the distance between adjacent holes, row spacing is the vertical distance between two adjacent rows of holes, and resistance line is the shortest distance from the explosive to the nearest free surface (free plane).

[0070] The charge structure refers to the arrangement of the explosives within the borehole and the interaction between the explosives and the surrounding medium. The charge amount refers to the total weight of the explosives within the borehole.

[0071] The method for determining the three-dimensional coordinates of blast holes based on the perforation parameters is as follows: First, obtain the terrain parameters of the area to be blasted. Then, based on the perforation parameters and terrain parameters, generate multiple design three-dimensional coordinates for each blast hole, where each design three-dimensional coordinate includes a horizontal coordinate and an elevation coordinate determined based on the terrain parameters. Next, import the design three-dimensional coordinates of multiple blast holes into the rover terminal of an RTK (Real-Time Kinematic) positioning system. Finally, use the base station and rover of the RTK positioning system for real-time dynamic positioning. Based on the deviation between the real-time position displayed on the rover terminal and the design three-dimensional coordinates, guide the rover to move to the design three-dimensional coordinates of each blast hole and mark that position to obtain the three-dimensional coordinates of each blast hole.

[0072] S2. Obtain the drilling parameters of the borehole according to the three-dimensional coordinates, and based on the drilling parameters, determine the target borehole with geological structure and the first geological structure of the target borehole.

[0073] Drilling parameters are those obtained during the drilling process, based on the three-dimensional coordinates of the borehole. These parameters include the variation curves of the thrust, rotational speed, and torque of the borehole drilling equipment during the drilling process. Drilling parameters can be acquired in real time using drilling-while-feeding (DWP) technology. DWP technology is crucial for precise trajectory control and intelligent drilling technologies; it allows the borehole drilling equipment to continuously monitor, transmit, and store drilling parameters while drilling is in progress.

[0074] The presence of geological structures in target boreholes and their primary geological structure are primarily determined based on the changing trends of drilling parameters. For example, when the drill bit of the borehole drilling equipment passes through a structural surface, the drilling parameters exhibit brief and sharp abrupt changes. These parameters typically show a characteristic of "a slight decrease in thrust, a slight increase in rotational speed, and a decrease in torque." If the drilling parameters of borehole A exhibit this characteristic, then borehole A can be identified as a target borehole with a geological structure, and the primary geological structure of borehole A is the structural surface.

[0075] In some embodiments, the drilling sensing technology can also identify the lithology and rock mass strength of the area to be blasted. For example, if the drill pressure is higher than a first preset value, it is hard rock; if the torque is higher than a second preset value, it is a high-hardness or abrasive rock formation.

[0076] In some embodiments, the boreholes may also include boreholes without geological structures. For boreholes without geological structures, no image acquisition is performed.

[0077] S3. Acquire images of the target blast holes, identify the second geological structure in the images, and establish a three-dimensional geological structure model based on the first geological structure, the second geological structure, and the geological information of the area to be blasted.

[0078] The image of the target blast hole can be acquired using a borehole television system containing a high-resolution camera. Specifically, the borehole television system with a high-resolution camera is placed inside the target blast hole, and the camera is controlled to rotate 360° to perform a panoramic scan, acquiring a two-dimensional color image unfolded along the hole wall, thus obtaining the image of the target blast hole.

[0079] The second geological structure serves as a further verification of the first geological structure to avoid errors in the first geological structure. For example, if the first geological structure of borehole A is a structural plane and the second geological structure is a fault, then it can be determined that the first geological structure is incorrect, and the actual geological structure of borehole A is a fault.

[0080] The second geological structure in the image can be identified using a convolutional neural network (CNN) model. Specifically, the image of the target borehole is input into the CNN, and the second geological structure in the image is output. The training process of the CNN includes:

[0081] (1) Data acquisition using borehole television equipment. First, a borehole television equipment containing a high-resolution camera is placed inside the borehole. The camera rotates 360° to perform a panoramic scan, acquiring a two-dimensional color image unfolded along the borehole wall. This yields key geological structure information such as structural planes, fault fracture zones, and karst caves, and establishes a database of various geological structure images.

[0082] (2) Data preprocessing. All data are image sharpened to enhance edge clarity, and professional personnel manually label areas such as structural planes, faults, and caves on the images. Different labels are assigned to different geological structures, with the background set to 0, and structural planes, faults, and caves set to 1, 2, and 3 respectively, to form a high-quality multi-category label map.

[0083] (3) Model Architecture. A convolutional neural network structure is adopted, namely a fully convolutional network with an encoder-decoder structure. Encoder: Extracts high-level features through convolution and pooling. Decoder: Restores spatial resolution through upsampling and preserves local details using skip connections. The network input consists of two-channel images, the original drill image and the sharpened image. Sobel edge detection results are introduced to enhance the network's sensitivity to edge regions.

[0084] (4) Network training. Input sample number, use cross-entropy loss function, use Adam optimizer, learning rate 1e-5 (0.00001), activation function is ReLU, total training times are 200.

[0085] The geological information of the area to be blasted includes the stratigraphy and lithology of the area and the macro-geological structure information, including the distribution, size and orientation of the geological structures.

[0086] The first and second geological structures are further refinements of the macroscopic geological structure information. For example, macroscopic geological structure information indicates that the upper layer of the area to be blasted has structural planes, the lower layer has karst caves, and there may be a near-horizontal fault fracture zone in between. This is a macroscopic, regional description that tells the staff what geological structures are roughly present, but does not know the specific locations of these geological structures. By acquiring images of the target blast holes and performing image recognition, the location of the geological structures in the area to be blasted can be refined.

[0087] A three-dimensional geological structure model can be constructed using 3D modeling software. Specifically, after obtaining the geological information of the first geological structure, the second geological structure, and the area to be blasted, the correlation between the geological structures of different boreholes is analyzed using the Kriging interpolation method. This allows for the inference of whether the geological structures of adjacent boreholes are the same, the inference of whether large geological structures exist, and the establishment of a three-dimensional geological structure model using 3D modeling software.

[0088] S4. Based on the three-dimensional geological structure model, the charge structure and charge amount are optimized to obtain the optimized scheme;

[0089] The optimization scheme includes the optimized charge structure and charge amount for each borehole.

[0090] Optimizing the charge structure and charge quantity based on a three-dimensional geological structure model essentially involves adjusting the charge structure and charge quantity of each borehole according to the geological structure and lithology of the strata shown in the three-dimensional geological structure model. For example, the charge structure and charge quantity of borehole A obtained in step S1 are determined based on the charge quantity and charge structure of the karst cave geological structure. However, the three-dimensional geological structure model shows that the actual geological structure of borehole A is a structural plane, and the lithology of the strata is hard rock. Therefore, the charge structure of borehole A should be adjusted to a slotted charge technique, and the charge quantity should be increased.

[0091] Furthermore, the geological structure includes at least structural planes, karst caves, and fault fracture zones. For areas with structural planes, the charge structure should prioritize slotting and fracture control techniques. For areas with fault fracture zones, the charge density should be reduced, and flexible sealing should be used. For areas with karst caves, holes should be avoided or weak charges should be used, supplemented by casing support.

[0092] S5. Obtain multiple blasting features after blasting the area to be blasted using the optimized scheme; the blasting features are used to determine the blasting effect.

[0093] Among them, blasting characteristics refer to the features of the blasted area formed after blasting according to the optimized plan. Blasting characteristics include blast fragment size distribution characteristics, blasting vibration characteristics, blast pile morphology characteristics, and blast smoke characteristics. Blasting fragment size distribution characteristics refer to the quantity or mass distribution of rock fragments formed after blasting within different particle size ranges, reflecting the degree and uniformity of rock fragmentation. Blasting vibration characteristics refer to the propagation characteristics of vibration waves on the ground or structures caused by the release of explosive energy during blasting, mainly including vibration velocity, frequency, and duration. Blasting pile morphology characteristics refer to the spatial geometry and distribution of the pile formed by the collapse and accumulation of rock after blasting. Blasting smoke characteristics refer to the physical and chemical properties of particulate matter, toxic and harmful gases, and their diffusion behavior generated after blasting. Furthermore, the blasting effect is evaluated based on the blast fragment size distribution characteristics, blasting vibration characteristics, blast pile morphology characteristics, and blast smoke characteristics.

[0094] S6. Based on the optimization scheme, borehole parameters, blasting characteristics and three-dimensional geological structure model, the prediction model used to predict blasting characteristics is trained to obtain the optimal blasting scheme prediction model.

[0095] Among them, the optimization scheme, borehole parameters, blasting characteristics, and three-dimensional geological structure model are used as training samples for the prediction model, mainly for training the prediction model.

[0096] The optimal blasting scheme prediction model is used to predict the blasting characteristics of the area to be predicted. When using the optimal blasting scheme prediction model to determine the optimal blasting scheme for the area to be predicted, each blasting scheme and a three-dimensional geological structure model of the area to be predicted are input into the optimal blasting scheme prediction model. The model outputs the predicted blasting characteristics corresponding to each blasting scheme. From these predicted blasting characteristics, a second target feature that matches the preset blasting characteristics is selected, and the blasting scheme corresponding to the second target feature is determined as the optimal blasting scheme.

[0097] The aforementioned method for determining blasting schemes considering geological structure and blasting effects involves acquiring the borehole network parameters, charge structure, and charge quantity for each borehole in the area to be blasted. Based on the borehole network parameters, the three-dimensional coordinates of the boreholes are determined, and drilling parameters for drilling the boreholes according to these three-dimensional coordinates are obtained. Based on these drilling parameters, target boreholes with existing geological structures and their primary geological structures are identified. Images of the target boreholes are acquired, and secondary geological structures within the images are identified. This allows for precise identification of the geological structure of the area to be blasted, enabling the establishment of a relatively accurate three-dimensional geological structure model based on the primary and secondary geological structures and the geological information of the area to be blasted. By optimizing the charge structure and charge quantity based on this three-dimensional geological structure model, an optimized scheme is obtained. Multiple blasting characteristics after blasting the area using the optimized scheme are acquired. This allows for evaluation of the blasting effect from multiple dimensions, improving the accuracy of the blasting results. By training a prediction model for blasting characteristics based on an optimized scheme, borehole parameters, blasting characteristics, and a three-dimensional geological structure model, an optimal blasting scheme prediction model is obtained. This allows the prediction model to continuously learn the intrinsic relationship between the geological structure, optimized scheme, borehole parameters, and blasting characteristics, outputting accurate predictions of blasting characteristics. Based on these predictions, the optimal blasting scheme can be determined, thereby improving the safety of deep-hole blasting on open-pit slopes.

[0098] In one embodiment, step S1 includes:

[0099] Based on the geological survey report of the area to be blasted, the geological information of the area to be blasted is determined. The geological information includes stratigraphic lithology and macroscopic geological structure information. The macroscopic geological structure information includes the distribution, scale and orientation of each geological structure in the area to be blasted.

[0100] Obtain the rock mechanics parameters and topographic parameters of the area to be blasted;

[0101] Based on geological information, rock mechanics parameters, and topographic parameters, the hole network parameters, the charge structure of each borehole, and the charge amount are determined for the area to be blasted.

[0102] Among them, the geological exploration report refers to a comprehensive technical document formed after systematically investigating and analyzing the geological conditions of the area to be blasted through geological surveying, geophysical exploration and other means.

[0103] Stratigraphic lithology is a geological description of the properties and characteristics of rocks in different strata. It includes information such as the type, composition, structure, texture, and relationships between rocks.

[0104] The methods for obtaining terrain parameters include using drones to conduct multi-angle photogrammetry of the area to be blasted, and obtaining terrain parameters such as elevation, slope, and surface undulation of the area to be blasted.

[0105] Rock mass mechanics parameters include at least the uniaxial compressive strength, tensile strength, and elastic modulus of the rock. These parameters can be obtained through in-situ field tests or laboratory tests. In-situ tests can be point load tests, while laboratory tests can be uniaxial tests, triaxial compression tests, or Brazilian splitting tests.

[0106] The borehole network parameters include borehole spacing, row spacing, and resistance line. Based on geological information, rock mechanics parameters, and topographic parameters, the principles for determining the borehole network parameters, charge structure, and charge quantity for the area to be blasted are as follows: For areas with structural planes, slotting and fracture control techniques should be prioritized; for areas with fault fracture zones, the charge density should be reduced, and flexible plugging should be used; for areas with karst caves, holes should be avoided or weak charges should be used, supplemented by casing support; the charge structure and charge quantity should be guided by rock mechanics parameters. For example, for intact and dense rock masses, deep-hole continuous charging is used, and the charge density is controlled within the range of 1.2 kg / m to 1.8 kg / m. Intact and dense rock masses refer to rock masses with a UCS (Uniaxial Compressive Strength) greater than 80 MPa; for weak and fractured rock masses, dispersed charging or weak charging is used, and the charge density is controlled below 0.5 kg / m to 0.8 kg / m. The UCS of weak and fractured rock masses is less than 30 MPa. For sloping terrain, segmented charging or inclined drilling is used to balance the distribution of blasting energy. For uneven terrain, the charge can be increased in raised areas and decreased in recessed areas to achieve uniform fragmentation. If the rock mass has significant structural planes, the number of delayed detonation stages should be increased according to the rock mass's density and orientation to control fragmentation and vibration. The borehole grid parameters depend on the formation homogeneity and free surface conditions. Hole spacing, row spacing, and hole depth should be designed based on the "minimum resistance line," typically controlled within 8 to 12 times the blasting hole diameter. For areas with uneven rock formations, the row spacing should be shortened and the detonation unit density increased. Furthermore, in flat areas, hole and row spacing should be arranged in a uniform grid; in sloping areas, the hole spacing needs to be denser to compensate for energy loss due to gravity. In areas with complex structural planes, multiple delayed detonators are used to achieve segmented controlled detonation. For example, in intact and dense limestone areas, a deep-hole blasting and multi-stage delayed detonation strategy is employed. The borehole diameter is 100 to 115 mm, the borehole spacing is 8 to 12 times the borehole diameter, and the row spacing is determined based on the target rock mass area. The borehole depth is slightly greater than the target rock mass height, and the minimum resistance line is designed to be 2 to 3 meters. A columnar continuous charge structure is used for single-hole charging, with a multi-stage delayed detonator placed at the upper and lower thirds of the charge to achieve segmented delayed detonation within the borehole.

[0107] In this embodiment, the geological information of the area to be blasted is determined based on the geological survey report of the area to be blasted, and the rock mechanics parameters and topographic parameters of the area to be blasted are obtained. Based on the geological information, rock mechanics parameters, and topographic parameters, the hole pattern parameters, the charge structure and charge amount of each blast hole are determined. This makes the hole pattern parameters, the charge structure and charge amount of each blast hole more accurate, which helps to achieve a reasonable distribution of blasting energy, improve the uniformity of rock fragmentation, reduce the proportion of large pieces and the base, and improve blasting efficiency and subsequent loading and unloading efficiency.

[0108] In one embodiment, step S2 includes:

[0109] When drilling boreholes according to three-dimensional coordinates, the drilling parameters of each borehole are obtained; the drilling parameters include at least the thrust variation curve, rotation speed variation curve, and torque variation curve of the borehole drilling equipment.

[0110] The curve variation characteristics of the top thrust variation curve, rotational speed variation curve, and torque variation curve are determined, and a preset geological structure mapping library is obtained; the preset geological structure mapping library includes multiple preset curve variation characteristics and the geological structures associated with each preset curve variation characteristic;

[0111] When a first target feature that matches the curve change feature of a borehole exists in the preset curve change features, the borehole is determined to be the target borehole.

[0112] The geological structure associated with the first target feature is identified as the first geological structure of the target borehole.

[0113] Drilling parameters can be obtained in real time from the borehole drilling equipment. The thrust variation curve is constructed from the thrust obtained during drilling. The rotational speed variation curve is constructed from the rotational speed obtained during drilling. The torque variation curve is constructed from the torque obtained during drilling.

[0114] The curve variation characteristics refer to the changes in the top thrust variation curve, rotation speed variation curve, and torque variation curve over time or drilling depth.

[0115] The preset geological structure mapping library includes multiple preset curve variation features and the geological structures associated with each preset curve variation feature. For example, the curve variation feature of "slight decrease in thrust variation curve, slight increase in rotational speed variation curve, and decrease in torque variation curve" is associated with the geological structure of a structural surface.

[0116] A target borehole refers to a borehole whose curve variation characteristics are consistent with or approximately consistent with the first target characteristics. For example, continuing the example above, the thrust variation curve, rotational speed variation curve, and torque variation curve are curves that change with drilling depth. If, at the same drilling depth, the thrust variation curve of borehole A slightly decreases, the rotational speed variation curve slightly increases, and the torque variation curve decreases, it can be determined that the curve variation characteristics of borehole A match the preset curve variation characteristics of "slight decrease in thrust variation curve, slight increase in rotational speed variation curve, and decrease in torque variation curve". In this case, borehole A is the target borehole, and the first geological structure of borehole A is the structural plane.

[0117] In a specific application, when the borehole drilling equipment passes through the structural surface, the drilling parameters will exhibit brief and sharp abrupt changes. The drilling parameters typically show the characteristics of "a slight decrease in the thrust curve, a slight increase in the rotational speed curve, and a decrease in the torque curve." Fault fracture zones are usually composed of fractured rock or weak mudstone, with strength much lower than that of intact surrounding rock. When the borehole drilling equipment passes through the fault fracture zone, the drilling parameters show the characteristics of "a significant decrease in the thrust curve, an accelerated increase in the rotational speed curve, and a decrease in the torque curve." The changes and recovery of drilling parameters are relatively slow, and the duration is affected by the fault width. When the borehole drilling equipment enters the cavity of a karst cave, it instantly loses all support and resistance. The drilling parameters typically show the characteristics of "a sharp decrease in the thrust and torque curves, and a sharp increase in the rotational speed curve." When the borehole drilling equipment touches the bottom of the karst cave, the thrust and torque curves will rise sharply again.

[0118] In this embodiment, by utilizing the curve variation characteristics of drilling parameters and combining them with a preset geological mapping database, the real-time identification and positioning of geological structures traversed by the borehole drilling equipment can be achieved, thereby identifying target boreholes with geological structures and improving the accuracy of the geological structure of the area to be blasted.

[0119] In one embodiment, step S3 includes:

[0120] Images of the target blast holes are acquired and input into a recognition model used to identify geological structures, thereby obtaining the second geological structure of the target blast holes;

[0121] When the first and second geological structures of the target borehole do not match, the second geological structure is determined as the geological structure of the target borehole, and the three-dimensional spatial coordinates of the geological structure of the target borehole are obtained.

[0122] Taking any target borehole as the target object, the horizontal and elevation distances between the target object and the associated boreholes are determined based on the three-dimensional spatial coordinates of the geological structure of the target object and the three-dimensional spatial coordinates of the geological structure of the associated boreholes; the associated boreholes are the adjacent boreholes of the target object that have the same geological structure type as the target object.

[0123] When the horizontal distance is less than the first threshold, the elevation distance is less than the second threshold, and the geological structure of the target object and the associated borehole are aligned, a geological structure association result is obtained in which the geological structure of the associated borehole and the target object is the same. The geological structure association result of each target borehole is obtained by traversing each target borehole.

[0124] Based on the geological information and geological structure correlation results of the area to be blasted, a three-dimensional geological structure model of the area to be blasted is established.

[0125] The recognition model can be a convolutional neural network model.

[0126] If the first geological structure and the second geological structure do not match, it means that the first geological structure determined by the drilling parameters is incorrect. In this case, the second geological structure needs to be determined as the actual geological structure of the target borehole. That is, the second geological structure is used as the geological structure of the target borehole.

[0127] The three-dimensional spatial coordinates of the geological structure of the target blast hole are determined based on the planar coordinates and elevation coordinates of the target blast hole. Specifically, the planar coordinates of the target blast hole are used as the planar coordinates of the geological structure, and the elevation of the geological structure is used as the elevation coordinates. The elevation of the geological structure is the elevation coordinates of the target blast hole minus the distance from the geological structure to the ground surface.

[0128] In some embodiments, after obtaining the three-dimensional spatial coordinates of the geological structure of the target borehole, the geological structure features can be quantified and the corresponding attributes can be labeled. For example, for structural planes, information such as their location, density, and attitude can be recorded; for faults, their depth, thickness, and degree of fracturing can be recorded; and for karst caves, their cave height, size, and filling conditions can be recorded.

[0129] Associated boreholes are boreholes adjacent to the target borehole that share the same geological structure type as the target borehole. For example, if target borehole A is the target borehole, and its geological structure is a structural plane, and its adjacent borehole B also has a structural plane geological structure, then adjacent borehole B is an associated borehole of target borehole A.

[0130] When traversing each target borehole, each target borehole must be taken as the target object until the geological structure correlation results of each target borehole are obtained.

[0131] The horizontal distance between the target object and the associated borehole refers to the horizontal distance between the geological structures in the target object and the geological structures in the associated borehole. The horizontal distance can be calculated using the planar coordinates of the geological structures in the three-dimensional space of the target object and the associated borehole. Specifically, the horizontal distance is calculated using the Euclidean distance formula based on the planar coordinates of the geological structures in the three-dimensional space of the target object and the associated borehole.

[0132] The elevation distance between the target object and the associated borehole refers to the elevation distance between the geological structures in the target object and the geological structures in the associated borehole. The elevation distance can be calculated based on the elevation coordinates of the geological structures in the three-dimensional space of the target object and the associated borehole. Specifically, the elevation distance is determined by the difference between the elevation coordinates of the geological structures in the three-dimensional space of the target object and the associated borehole.

[0133] The method for determining whether the geological structure of the target object and the associated blast hole are consistent is to measure the azimuth of the geological structure of the target object and the associated blast hole and compare them. If the azimuth deviation is less than the deviation threshold, then the directions are consistent.

[0134] The geological information and geological structure correlation results of the area to be blasted are input into 3D modeling software for modeling, thereby establishing a 3D geological structure model of the area to be blasted. Furthermore, target boreholes with the same geological structure can be connected in 3D to form an extension trajectory, and the extension trajectory can be input into the 3D modeling software for modeling.

[0135] In some embodiments, if the geological structure in N target boreholes is the same and the length of the geological structure is greater than a set threshold, then the structure is a large geological structure, where N is an integer greater than 3 and less than the total number of target boreholes, so that the large geological structure can be input into the 3D modeling software for modeling.

[0136] Furthermore, after the blasting was completed, the site was cleared of debris, and the blasted area was photographed using drones to locate the exposed surface, verify the accuracy of the 3D geological structure model, and adjust the model based on the on-site photographic results. A schematic diagram of the exposed surface is shown below. Figure 2 As shown.

[0137] In this embodiment, when the first and second geological structures of the target blast hole do not match, the second geological structure is determined as the geological structure of the target blast hole. This reduces geological identification errors and improves the accuracy of geological structure judgment. Using any target blast hole as the target object, the horizontal and vertical distances between the target object and the associated blast holes are determined based on their three-dimensional spatial coordinates and the three-dimensional spatial coordinates of their geological structures. When the horizontal distance is less than a first threshold, the vertical distance is less than a second threshold, and the geological structures of the target object and the associated blast holes are aligned, a geological structure association result is obtained, indicating that the geological structures of the associated blast holes and the target object are of the same structure. This process is repeated for each target blast hole to obtain the geological structure association results for each target blast hole. This allows for a scientific determination of spatial continuity, extension, and geological structure, achieving a leap from "point-like identification" to "line / area structure deduction." By establishing a three-dimensional geological structure model of the area to be blasted based on the geological information and geological structure association results, the accuracy of the three-dimensional geological structure model is significantly higher than that of traditional geological exploration inference models, facilitating refined guidance for blasting design.

[0138] In one embodiment, step S5 includes:

[0139] Images of the blasted area after blasting are acquired to obtain blasting images;

[0140] Based on the blasting images, perform blasting fragmentation analysis to obtain the blasting fragmentation distribution characteristics of the blasting area; the blasting fragmentation distribution characteristics include median fragmentation, blasting index, and large block ratio;

[0141] Determine the median fragmentation, blasting index, and large block ratio as blasting characteristics.

[0142] Among them, the blasting images can be obtained by the camera carried on the unmanned aerial vehicle taking multi-angle photos of the blasting area.

[0143] The blasting fragmentation analysis is carried out by the image processing screening method. The image processing screening method refers to a method of performing image preprocessing, feature extraction, and structure separation on the blasting images, and then realizing the automatic recognition, screening, and classification of different blasting fragmentations. The image processing screening method depends on the significant features such as the gray level, edges, and textures of the images for differential analysis, and is applicable to the non-contact recognition and distribution statistics of complex fragmentation sizes in blasting engineering. The schematic diagram of the image processing screening method for identifying fragmentation sizes is as Figure 3 shown. The fragmentation distribution diagram after blasting is as Figure 4 shown.

[0144] The median fragmentation refers to the value that 50% of the rock blocks are less than or equal to. For example, the median fragmentation = 30 cm, which means that after blasting, 50% of the rock block sizes are less than or equal to 30 cm, and 50% of the rock blocks are larger than 30 cm. The smaller the median fragmentation, the more uniform the fragmentation, and the better the blasting effect. The ideal value of the median fragmentation is usually 10 - 15 times the charge diameter. The fragmentation can be understood as the size of the rock.

[0145] The calculation formula for the blasting index is x0 is the characteristic fragmentation, n is the blasting index, P(x) represents the cumulative passing rate of fragmentation x, and the characteristic fragmentation x0 can be directly defined as the fragmentation size when P(x) = 0.632. The larger the blasting index n, the steeper the curve, the more concentrated the fragmentation distribution, and the more uniform the fragmentation, and the better the blasting effect. When n > 1.2, the fragmentation is relatively uniform; when 0.8 < n ≤ 1.2, the fragmentation distribution is relatively wide; when n < 0.8, it means there are more large blocks / fine powders. The large block ratio refers to the proportion of blocks with sizes larger than the maximum allowable block size. When the proportion exceeds 10%, it indicates that the blasting effect is not good and secondary crushing is required, increasing the cost. When calculating the blasting index n, in addition to using the defined characteristic fragmentation x0, a data point with a known P(x) and corresponding fragmentation x is also required, so as to calculate the blasting index n according to the known cumulative passing rate P(x), fragmentation x, and characteristic fragmentation x0. After screening, it is found that the proportion of blasted stones with fragmentation x ≤ 80 mm is 40%, then x = 80 mm, and P(x) = 0.40.

[0146] In this embodiment, images of the blasted area after blasting are acquired to obtain blasting images. Based on the blasting images, blasting block size analysis is performed to obtain the blasting block size distribution characteristics of the blasting area. The blasting block size distribution characteristics include median block size, blasting index, and large block ratio. The median block size, blasting index, and large block ratio are determined as blasting characteristics, so that the blasting effect can be accurately evaluated based on the median block size, blasting index, and large block ratio.

[0147] In one embodiment, step S5 includes:

[0148] Pre-set blasting vibration monitoring points and collect vibration velocity data from these points during the blasting process;

[0149] Regression analysis was performed on the vibration velocity data to obtain the peak vibration velocity, dominant vibration frequency, and vibration duration at each blasting vibration monitoring point.

[0150] The peak vibration velocity, dominant vibration frequency, and vibration duration are defined as blasting characteristics.

[0151] Among them, blasting vibration monitoring points are set up before blasting. When setting up blasting vibration monitoring points, blasting vibration meters are also deployed at these points to collect vibration velocity data. A schematic diagram of the vibration velocity data is shown below. Figure 5 , Figure 6 , Figure 7 As shown.

[0152] When performing regression analysis on vibration velocity data, the Sadovsky formula is used to calculate the peak vibration velocity, dominant vibration frequency, and duration of vibration at each blasting vibration monitoring point.

[0153] Peak vibration velocity refers to the maximum instantaneous velocity of a particle at the blasting vibration monitoring point in the vibration direction. The dominant vibration frequency refers to the frequency with the largest vibration amplitude. Vibration duration refers to the time elapsed from the start of significant fluctuations at the blasting vibration monitoring point to its end and return to a stable state.

[0154] Furthermore, if the peak vibration velocity is within a safe range, the dominant vibration frequency is higher than the frequency threshold, and the vibration duration is within the preset duration, it indicates that the blasting effect is good.

[0155] In this embodiment, by pre-setting blasting vibration monitoring points, vibration velocity data is collected from these points during the blasting process. Regression analysis is then performed on the vibration velocity data to obtain the peak vibration velocity, dominant vibration frequency, and vibration duration at each monitoring point. This reduces the need for manual operation and improves work efficiency. By defining the peak vibration velocity, dominant vibration frequency, and vibration duration as blasting characteristics, accurate assessment of the blasting effect can be achieved based on these parameters.

[0156] In one embodiment, step S5 includes:

[0157] Acquire point cloud data of the blast pile after blasting in the area to be blasted;

[0158] Based on the point cloud data of the blast pile and the three-dimensional geological structure model, the throwing angle, accumulation height and loosening coefficient were determined.

[0159] The throwing angle, accumulation height, and loosening coefficient were determined as blasting characteristics.

[0160] The acquisition method for blast pile point cloud data involves using a drone equipped with a 3D scanning device to scan the blast pile area from multiple angles after the blasting operation, obtaining raw point cloud data containing spatial 3D coordinate information. The raw point cloud data is then preprocessed to obtain high-precision, seamlessly stitched blast pile point cloud data. The blast pile area refers to the concentrated area of ​​loose material formed on the surface or near the excavation face after blasting, due to the breaking and throwing of rock or soil by the energy of the explosive. A schematic diagram of scanning the blast pile area is shown below. Figure 8 As shown.

[0161] The process of determining the throwing angle is as follows: based on the three-dimensional geological structure model and the three-dimensional terrain model of the area to be blasted before blasting, the original center position of the rock mass to be blasted is determined; the centroid position of the point cloud data of the blast pile is calculated; a spatial displacement vector is constructed from the original center position to the centroid position of the blast pile; and the angle between the displacement vector and the horizontal plane is determined as the throwing angle.

[0162] The process of determining the accumulation height involves extracting the maximum and minimum elevation values ​​within the blasting area; the difference between the maximum and minimum elevation values ​​is taken as the accumulation height, or the difference between the maximum elevation value and the elevation of the corresponding area of ​​the pre-blast terrain is taken as the accumulation height. The elevation of the corresponding area of ​​the pre-blast terrain refers to the elevation of the blasting area before blasting.

[0163] The process of determining the loosening coefficient is as follows: obtain the original rock mass volume of the area to be blasted; calculate the actual accumulation volume of the blasting area; and determine the loosening coefficient as the ratio of the actual accumulation volume to the original rock mass volume.

[0164] Furthermore, the blasting effect is determined to be good if the preset throwing angle, preset pile height, and preset loosening coefficient are all within their respective ranges. Specifically, if the preset throwing angle θ∈[30°, 45°], the preset pile height H≤ the excavator boom length × 0.8, and the preset loosening coefficient K∈[1.35, 1.50], the blasted pile can be considered to meet loading, transportation, and safety requirements. For example, if the throwing angle is within the preset throwing angle θ∈[30°, 45°], the pile height is less than or equal to the excavator boom length × 0.8, and the loosening coefficient is within the preset loosening coefficient K∈[1.35, 1.50], the blasting effect can be determined to be good. Further, if any one or more of the following conditions exist—the throwing angle not being within the preset throwing angle range, the pile height not being within the preset pile height range, or the loosening coefficient not being within the preset loosening coefficient range—the blasting effect can be determined to be average or poor.

[0165] In this embodiment, by acquiring point cloud data of the blast pile and combining it with a three-dimensional geological structure model, key parameters such as throwing angle, accumulation height, and loosening coefficient can be accurately calculated. This enables a systematic and quantitative description of the rock mass movement and accumulation state after blasting, overcoming the subjectivity and inaccuracy of traditional experience-based judgment.

[0166] In one embodiment, step S5 includes:

[0167] During the blasting process in the area to be blasted, the control and data acquisition device collects the concentration changes and diffusion trajectories of harmful gases generated during the blasting process, as well as the aerodynamic diameter of particulate matter, according to the preset flight path and flight altitude.

[0168] The changes in concentration, the diffusion trajectory, and the aerodynamic diameter of the particles were determined as the characteristics of the explosion.

[0169] The data collection device is a device capable of flying along a preset flight path and at a preset altitude. For example, a drone. The data collection device is equipped with a laser particle sensor and a gas sensor. The laser particle sensor can monitor particulate matter in the fumes, while the gas sensor can detect changes in the concentration and diffusion trajectory of harmful gases, including but not limited to CO, NO2, and SO2.

[0170] Furthermore, before the blast, the flight path and altitude are preset, and the laser particle sensor and gas sensor mounted on the data acquisition device are calibrated.

[0171] Furthermore, if the concentration of harmful gases rises rapidly, reaches a high peak, and decays quickly, while the diffusion trajectory extends orderly along the prevailing wind direction, it indicates that the explosive energy is fully released and the blasting effect is good; if the proportion of particles with an average aerodynamic diameter of less than 10 μm is relatively high, it indicates that the rock is finely broken, the explosive utilization rate is high, and the blasting effect is good.

[0172] In this embodiment, during the blasting process in the area to be blasted, the data acquisition device is controlled to collect the concentration changes and diffusion trajectories of harmful gases generated during the blasting process, as well as the aerodynamic diameter of particulate matter, according to a preset flight path and altitude. This allows for real-time capture of the changes in pollutants in both spatial and temporal dimensions, overcoming the limitations of traditional fixed-point, static monitoring. By defining the concentration changes, diffusion trajectories, and aerodynamic diameter of particulate matter as blasting characteristics, accurate assessment of the blasting effect can be achieved based on these factors.

[0173] In one embodiment, step S6 includes:

[0174] Data augmentation processing was performed on the optimization scheme, borehole parameters, blasting characteristics, and three-dimensional geological structure model to obtain multiple training samples;

[0175] The prediction model for predicting blasting features is trained using training samples to obtain the optimal blasting scheme prediction model after training.

[0176] Step S6 is followed by

[0177] The current geological structure information of the area to be predicted and the blasting schemes for the area to be predicted are input into the prediction model to obtain the predicted blasting characteristics of each blasting scheme.

[0178] Select the second target feature that matches the preset blasting feature from the predicted blasting features, and determine the blasting scheme corresponding to the second target feature as the optimal blasting scheme.

[0179] The purpose of data augmentation is to increase the sample size.

[0180] Furthermore, blasting characteristics include blast fragment size distribution, blast vibration characteristics, blast pile morphology characteristics, and blast smoke characteristics. Blasting fragment size distribution characteristics include median fragment size, blast index, and the proportion of large fragments. Blasting vibration characteristics include peak vibration velocity, dominant vibration frequency, and vibration duration at each blast vibration monitoring point. Blasting pile morphology characteristics include throwing angle, pile height, and loosening coefficient. Blasting smoke characteristics include concentration variations, diffusion trajectory, and aerodynamic diameter of particles.

[0181] When training a prediction model for predicting blasting features using training samples, the training samples serve as the input parameters of the prediction model, and the predicted blasting features serve as the output parameters. Further, the model is divided into training, validation, and test sets. Mean squared error or weighted mean squared error is used as the loss function for the prediction model, and AdamW is used as the optimizer. The learning rate can be set to 1e-4, and the batch size is 32 to 64. In each training round, the root mean square error and coefficient of determination R of the validation set are recorded. 2 Values. Root mean square error, mean absolute error, and coefficient of determination R0. 2 Regression metrics are used to evaluate the training performance of the prediction model.

[0182] Each blasting scheme for the area to be predicted includes borehole network parameters, the charge structure of each borehole, and the charge quantity. Current geological structure information can be determined from the geological survey report of the area to be predicted.

[0183] Furthermore, after blasting the area to be predicted according to the optimal blasting scheme, the current geological structure information of the area to be predicted, the optimal blasting scheme, and the blasting characteristics after blasting can be used as training samples for the optimal blasting scheme prediction model, triggering incremental training, so that the performance of the optimal blasting scheme prediction model is gradually optimized, thereby realizing the optimization cycle of the adaptive blasting design scheme.

[0184] Furthermore, after blasting the area to be predicted according to the optimal blasting scheme, the actual blasting characteristics of the area to be predicted are obtained, and the actual blasting characteristics are compared and analyzed with the preset blasting characteristics. The error is recorded and input into the optimal blasting scheme prediction model to further train the optimal blasting scheme prediction model.

[0185] In this embodiment, data augmentation technology is used to expand the training samples. Combined with multi-source heterogeneous data such as 3D geological structure models, borehole parameters, optimization schemes, and blasting characteristics, the prediction model is systematically trained. This achieves a shift from "experience-driven" to "data and model-driven" approaches, significantly improving the intelligence and scientific rigor of blasting prediction. By matching and filtering predicted blasting characteristics with preset blasting characteristics, the system automatically identifies the "second target characteristic" that best meets engineering requirements and its corresponding blasting scheme. This enables intelligent recommendation and optimal decision-making for blasting schemes, improving selection efficiency and objectivity.

[0186] This application also provides an application scenario in which the above-described method for determining a blasting scheme that considers geological structure and blasting effects is applied. Specifically, the application of this method for determining a blasting scheme that considers geological structure and blasting effects in this scenario is as follows:

[0187] The overall flowchart of the method for determining blasting schemes considering geological structure and blasting effects is as follows: Figure 9As shown, specifically, based on the geological survey report of the area to be blasted, the geological information of the area is determined. Rock mechanics and topographic parameters of the area to be blasted are acquired. Based on the geological information, rock mechanics parameters, and topographic parameters, a preliminary blasting scheme is determined, including borehole network parameters, the charge structure of each borehole, and the charge quantity. Drilling parameters are acquired in real time using drilling sensing technology. Based on these parameters, target boreholes with existing geological structures and the first geological structure of the target boreholes are identified, while the lithology and rock mass strength of the strata in the area to be blasted are also identified. Images of the target boreholes are acquired using borehole television equipment, and a convolutional neural network is used to identify the second geological structure in the images. The correlation of geological structures between different boreholes is analyzed using Kriging interpolation to infer whether the geological structures between adjacent boreholes are the same structure, to infer the existence of large geological structures, and to establish a three-dimensional geological structure model. Based on the three-dimensional geological structure model, the charge structure and charge quantity are optimized to obtain an optimized scheme. Images of the blasted area after blasting are acquired, and blast images are analyzed based on these images to obtain the blast block size distribution characteristics of the blasted area. These characteristics include median block size, blasting index, and the proportion of large blocks, which are then defined as blasting features. Pre-set blasting vibration monitoring points are established, and vibration velocity data is collected from these points during blasting. Regression analysis of the vibration velocity data yields the peak vibration velocity, dominant frequency, and duration of vibration at each monitoring point. These peak vibration velocity, dominant frequency, and duration are then defined as blasting features. Point cloud data of the blast pile after blasting is obtained. Based on the point cloud data and a 3D geological structure model, the throwing angle, accumulation height, and loosening coefficient are determined. These factors are then defined as blasting features. During the blasting process in the area to be blasted, the control and acquisition device collects the concentration changes and diffusion trajectories of harmful gases generated during the blasting process, as well as the aerodynamic diameter of particles, according to the preset flight path and altitude. The concentration changes, diffusion trajectories, and aerodynamic diameter of particles are determined as blasting characteristics. Data augmentation processing is performed on the optimized scheme, borehole parameters, blasting characteristics, and three-dimensional geological structure model to obtain multiple training samples. The prediction model used to predict blasting characteristics is trained using the training samples to obtain the optimal blasting scheme prediction model. The current geological structure information of the area to be predicted and each blasting scheme for the area to be predicted are input into the optimal blasting scheme prediction model to obtain the predicted blasting characteristics of each blasting scheme. From the predicted blasting characteristics, a second target feature that matches the preset blasting characteristics is selected, and the blasting scheme corresponding to the second target feature is determined as the optimal blasting scheme.

[0188] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0189] Based on the same inventive concept, this application also provides an apparatus for determining a blasting scheme that considers geological structure and blasting effect, used to implement the above-described method for determining a blasting scheme that considers geological structure and blasting effect. The solution provided by this apparatus is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more embodiments of the apparatus for determining a blasting scheme that considers geological structure and blasting effect provided below can be found in the limitations of the method for determining a blasting scheme that considers geological structure and blasting effect described above, and will not be repeated here.

[0190] In one embodiment, a blasting scheme determination device considering geological structure and blasting effect is provided, comprising:

[0191] The parameter acquisition module is used to acquire the hole network parameters, the charge structure and charge amount of each blast hole for blasting the area to be blasted, and to determine the three-dimensional coordinates of the blast holes based on the hole network parameters.

[0192] The first structure determination module is used to obtain the drilling parameters of the borehole drilled according to the three-dimensional coordinates, and based on the drilling parameters, determine the target borehole with a geological structure and the first geological structure of the target borehole.

[0193] The second structure determination module is used to acquire images of the target boreholes, identify the second geological information in the images, and establish a three-dimensional geological structure model.

[0194] The scheme optimization module is used to optimize the charge structure and the charge amount based on the three-dimensional geological structure model to obtain an optimized scheme;

[0195] The feature acquisition module is used to acquire multiple blasting features after the area to be blasted is blasted using the optimization scheme; the blasting features are used to determine the blasting effect.

[0196] The scheme determination module is used to train a prediction model for predicting blasting characteristics based on the optimization scheme, the hole mesh parameters, the blasting characteristics, and the three-dimensional geological structure model, so as to obtain the optimal blasting scheme prediction model.

[0197] The modules in the aforementioned blasting scheme determination device, which considers geological structure and blasting effects, can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0198] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0199] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0200] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0201] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0202] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0203] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0204] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for determining a blasting scheme that considers geological structure and blasting effect, characterized in that, The method includes: S1. Obtain the hole network parameters, the charge structure and charge amount of each blast hole for the area to be blasted, and determine the three-dimensional coordinates of the blast holes based on the hole network parameters. S2. Obtain the drilling parameters for drilling the blast hole according to the three-dimensional coordinates, and based on the drilling parameters, determine the target blast hole with a geological structure and the first geological structure of the target blast hole; S3. Acquire images of the target blast holes, identify the second geological structure in the images, and establish a three-dimensional geological structure model based on the first geological structure, the second geological structure, and the geological information of the area to be blasted. S4. Based on the three-dimensional geological structure model, optimize the charge structure and the charge quantity to obtain an optimized solution; S5. Obtain multiple blasting features after blasting the area to be blasted using the optimization scheme; the blasting features are used to determine the blasting effect; S6. Based on the optimization scheme, the hole mesh parameters, the blasting characteristics, and the three-dimensional geological structure model, the prediction model used to predict the blasting characteristics is trained to obtain the optimal blasting scheme prediction model.

2. The method according to claim 1, characterized in that, Step S1 includes: Based on the geological survey report of the area to be blasted, the geological information of the area to be blasted is determined; the geological information includes stratigraphic lithology and macroscopic geological structure information, and the macroscopic geological structure information includes the distribution location, scale and orientation of each geological structure in the area to be blasted; Obtain the rock mechanics parameters and topographic parameters of the area to be blasted; Based on the geological information, the rock mechanics parameters, and the topographic parameters, the hole network parameters, the charge structure of each borehole, and the charge amount are determined for the area to be blasted.

3. The method according to claim 1, characterized in that, Step S2 includes: When drilling boreholes according to the three-dimensional coordinates, the drilling parameters of each borehole are obtained; the drilling parameters include at least the thrust variation curve, rotation speed variation curve, and torque variation curve of the borehole drilling equipment; The curve change characteristics of the top thrust change curve, the rotational speed change curve, and the torque change curve are determined, and a preset geological structure mapping library is obtained; the preset geological structure mapping library includes multiple preset curve change characteristics and the geological structure associated with each preset curve change characteristic; When a first target feature exists in the preset curve change features that matches the curve change features of the borehole, the borehole is determined to be a target borehole; The geological structure associated with the first target feature is identified as the first geological structure of the target borehole.

4. The method according to claim 1, characterized in that, Step S3 includes: Images of the target boreholes are acquired and input into a recognition model for identifying geological structures to obtain the second geological structure of the target boreholes; When the first geological structure and the second geological structure of the target blast hole do not match, the second geological structure is determined as the geological structure of the target blast hole, and the three-dimensional spatial coordinates of the geological structure of the target blast hole are obtained; Taking any of the target blast holes as the target object, the horizontal distance and elevation distance between the target object and the associated blast hole are determined based on the three-dimensional spatial coordinates of the geological structure of the target object and the three-dimensional spatial coordinates of the geological structure of the associated blast hole; the associated blast hole is a blast hole among the adjacent blast holes of the target object that has the same geological structure type as the target object. When the horizontal distance is less than a first threshold, the elevation distance is less than a second threshold, and the geological structure of the target object and the associated borehole are aligned, a geological structure association result is obtained in which the geological structure of the associated borehole and the target object is the same. Each target borehole is traversed to obtain the geological structure association result of each target borehole. Based on the geological information of the area to be blasted and the correlation results of the geological structure, a three-dimensional geological structure model of the area to be blasted is established.

5. The method according to claim 1, characterized in that, Step S5 includes: Image acquisition is performed on the blasted area after the blasting of the area to be blasted, and blasting images are obtained; Based on the blasting images, blasting block size analysis is performed to obtain the blasting block size distribution characteristics of the blasting area; the blasting block size distribution characteristics include median block size, blasting index, and large block proportion; The median block size, the blasting index, and the proportion of large blocks are determined as blasting characteristics.

6. The method according to claim 1, characterized in that... As stated, step S5 includes: Pre-set blasting vibration monitoring points, and collect vibration velocity data from the blasting vibration monitoring points during the blasting process; Regression analysis was performed on the vibration velocity data to obtain the peak vibration velocity, dominant vibration frequency, and vibration duration at each of the blasting vibration monitoring points. The peak vibration velocity, the dominant vibration frequency, and the duration of vibration are defined as the blasting characteristics.

7. The method according to claim 1, characterized in that, Step S5 includes: Obtain the point cloud data of the blast pile after the blasting of the area to be blasted; Based on the point cloud data of the blast pile and the three-dimensional geological structure model, the throwing angle, accumulation height, and loosening coefficient are determined. The throwing angle, the accumulation height, and the loosening coefficient are determined as blasting characteristics.

8. The method according to claim 1, characterized in that, Step S5 includes: During the blasting process in the area to be blasted, the control and acquisition device collects the concentration changes and diffusion trajectory of harmful gases generated during the blasting process, as well as the aerodynamic diameter of particulate matter, according to the preset flight route and flight altitude. The concentration changes, the diffusion trajectory, and the aerodynamic diameter of the particles are determined as the blasting characteristics.

9. The method according to claim 1, characterized in that, Step S6 includes: Data augmentation processing is performed on the optimization scheme, the borehole parameters, the blasting characteristics, and the three-dimensional geological structure model to obtain multiple training samples; The training samples are used to train the prediction model for predicting blasting features, and the optimal blasting scheme prediction model is obtained after training. Step S6 is followed by: The current geological structure information of the area to be predicted and each blasting scheme for the area to be predicted are input into the prediction model to obtain the predicted blasting characteristics of each blasting scheme. Select a second target feature from the predicted blasting features that matches the preset blasting features, and determine the blasting scheme corresponding to the second target feature as the optimal blasting scheme.

Images