A blasting hole network parameter optimization method based on image recognition

Abstract

The application is suitable for the technical field of blasting engineering, and provides a blasting hole network parameter optimization method based on image recognition, which comprises the following steps: acquiring images and characteristic information of a plurality of excavation sections; acquiring blasting hole network parameters according to the characteristic information of each excavation section; using a numerical simulation method to optimize and adjust the blasting hole network parameters multiple times, and calculating blasting effect indexes of the blasting hole network parameters before and after the adjustment; determining optimal blasting hole network parameters based on a multiple regression analysis model; fusing the optimal blasting hole network parameters and the optimized and adjusted blasting hole network parameters; generating a training data set based on the images of the excavation sections, the parameter fusion results and the corresponding blasting effect indexes; training a generative adversarial network by using the training data set; and optimizing the blasting hole network parameters by using the trained hole network parameter optimization model. The application can improve the accuracy and scientificity of the blasting hole network parameter optimization.

Description

Technical Field

[0001] This application belongs to the field of blasting engineering technology, and in particular relates to a method for optimizing blasting hole network parameters based on image recognition. Background Technology

[0002] In deep resource extraction and tunnel excavation, blasting is currently the primary method for rock breaking. Whether in mine roadways or tunnel construction, the design of the blasting hole pattern directly affects construction efficiency, quality, and safety. A well-designed blasting hole pattern maximizes blasting effectiveness, reduces unnecessary repetitive work, and thus improves overall construction efficiency. Simultaneously, a precise hole pattern design effectively reduces over-excavation and under-excavation, minimizes disturbance to the rock strata, and reduces the workload of subsequent support work.

[0003] Current methods employ the minimum resistance line combined with theoretical formula derivation for optimization, which can improve the effectiveness of inter-hole micro-delay blasting. However, this approach is ill-suited to complex geological conditions, such as those with well-developed joints and fractures or diverse mineral compositions. The minimum resistance line-based method struggles to provide a scientific and accurate design, and because it relies heavily on static design, it's difficult to dynamically adjust based on real-time field data. Existing experience-based methods, relying primarily on the experience and technical expertise of blasting experts to determine borehole network parameters, are highly subjective, resulting in poor accuracy and scientific rigor in parameter optimization and hindering the guarantee of optimal optimization results. Summary of the Invention

[0004] This application provides an image recognition-based method for optimizing blasting hole mesh parameters, which can solve the problems of poor accuracy and scientific rigor in blasting hole mesh parameter optimization.

[0005] This application provides a method for optimizing the parameters of a blasting hole mesh based on image recognition, including:

[0006] Acquire images and feature information of multiple excavation sections;

[0007] For each excavation section, blasting hole network parameters are obtained based on the characteristic information of the excavation section. Numerical simulation methods are used to calculate the blasting effect index of blasting the excavation section based on the blasting hole network parameters. The blasting hole network parameters are then optimized and adjusted multiple times, and the blasting effect index of blasting the excavation section based on the optimized blasting hole network parameters is calculated for each adjustment. Based on a pre-constructed multiple regression analysis model, all blasting effect indices obtained using numerical simulation methods, and the blasting hole network parameters corresponding to each blasting effect index, the optimal blasting hole network parameters are determined from the existing blasting hole network parameters and the optimized blasting hole network parameters. Finally, the optimal blasting hole network parameters and the optimized blasting hole network parameters are fused together.

[0008] A training dataset is generated based on images of multiple excavation sections, all fusion results corresponding to multiple excavation sections, and blasting effect indicators corresponding to the blasting hole mesh parameters after each optimization and adjustment.

[0009] The generative adversarial network was trained using the training dataset to obtain an optimized model for the mesh parameters.

[0010] The blasting hole mesh parameters of the excavation section were optimized using a hole mesh parameter optimization model.

[0011] Optional, the feature information includes rock type, joint and fracture characteristics, and cross-sectional geometric parameters.

[0012] Optional parameters for the blasting hole network include hole diameter, hole depth, and number of holes.

[0013] Optionally, the blasting effect index is the damage rate, and the formula for calculating the damage rate is:

[0014]

[0015] Where D represents the damage rate, N d This represents the number of mesh elements in the rock model that reach the damage threshold when the numerical simulation method is used, and N represents the total number of mesh elements in the rock model when the numerical simulation method is used.

[0016] Optionally, the expression for the multiple regression analysis model is:

[0017] y = β0 + β1x1 + β2x2 + β3x3 + ε

[0018] Where y represents the blasting effect index, β0 represents the constant term, β1, β2, and β3 all represent regression coefficients, x1, x2, and x3 represent the borehole diameter, borehole depth, and number of boreholes, respectively, and ε is the error term.

[0019] Optionally, based on a pre-constructed multiple regression analysis model, all blasting effect indices obtained using numerical simulation methods, and the blasting hole mesh parameters corresponding to each blasting effect index, the optimal blasting hole mesh parameters are determined from the blasting hole mesh parameters and the blasting hole mesh parameters after each optimization adjustment, including:

[0020] All blasting effect indices obtained by numerical simulation methods, and the blasting hole mesh parameters corresponding to each blasting effect index, are substituted into the multiple regression analysis model to obtain multiple equations for solving.

[0021] Solve multiple equations to obtain the values ​​of the constant term, regression coefficients and error term of the multiple regression analysis model, and use the obtained values ​​to update the multiple regression analysis model;

[0022] Using the updated multiple regression analysis model, the blasting effect index of the blasting hole mesh parameters and the blasting hole mesh parameters after each optimization adjustment was calculated;

[0023] The blasting hole mesh parameters corresponding to the blasting effect indexes that are greater than the preset threshold among the calculated blasting effect indexes are taken as the optimal blasting hole mesh parameters.

[0024] Optionally, the optimal blasting hole network parameters and the blasting hole network parameters after each optimization adjustment are fused, including:

[0025] For each optimized and adjusted blasting hole network parameter, the optimal blasting hole network parameters and the optimized and adjusted blasting hole network parameters are fused using a fusion formula; the fusion formula is:

[0026] θ fused =θ0·ω0+θ1·ω1

[0027] Where, θ fused Let θ0 represent the fused blasting hole mesh parameters, ω0 represent the optimal blasting hole mesh parameters, and θ1 represent the optimized blasting hole mesh parameters, with ω1 representing the weight of θ1.

[0028] Optionally, the fusion result is the fused blast hole mesh parameters, and the training dataset includes multiple training samples;

[0029] A training dataset is generated based on images of multiple excavation sections, all fusion results corresponding to the multiple excavation sections, and blasting effect indices corresponding to the blasting hole mesh parameters after each optimization adjustment. This dataset includes:

[0030] For each of the multiple excavation sections, perform the following steps:

[0031] For each fusion result corresponding to the excavation section, the blasting effect index corresponding to the fusion result is determined from all blasting effect indices obtained by numerical simulation method, and the image of the excavation section, the fusion result, and the determined blasting effect index are used as a training sample.

[0032] The above-mentioned solution in this application has the following beneficial effects:

[0033] In the embodiments of this application, a borehole parameter optimization model is obtained by training a generative adversarial network (GAN) using the acquired training dataset. This model is then used to optimize the blasting borehole parameters of the excavation section, reducing reliance on expert experience. Machine learning technology enables the optimization of blasting borehole parameters, thus improving the scientific rigor of the optimization. Furthermore, during the acquisition of the training dataset, the optimal blasting borehole parameters are determined through multivariate regression analysis. These optimal parameters are then fused with those optimized through numerical simulation, and a training dataset is generated based on the fused optimal parameters. This allows the GAN to learn the optimal blasting borehole parameters during training, thereby improving the accuracy of the optimized borehole parameter optimization model. This provides strong data support for subsequent engineering construction and ensures construction safety.

[0034] Other beneficial effects of this application will be described in detail in the following detailed description section. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A flowchart illustrating an image recognition-based method for optimizing blasting hole mesh parameters, as provided in an embodiment of this application;

[0037] Figure 2 A numerical diagram of borehole depth provided for an embodiment of this application. Detailed Implementation

[0038] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0039] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0040] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0041] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0042] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0043] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0044] To address the current shortcomings in the accuracy and scientific rigor of blasting hole mesh parameter optimization, this application provides an image recognition-based method for optimizing blasting hole mesh parameters. This method trains a generative adversarial network (GAN) using an acquired training dataset to obtain a hole mesh parameter optimization model. This model is then used to optimize the blasting hole mesh parameters of the excavation section, reducing reliance on expert experience. Optimization of blasting hole mesh parameters can be achieved based on machine learning technology, thus improving the scientific rigor of the optimization. Furthermore, during the acquisition of the training dataset, the optimal blasting hole mesh parameters are determined through multivariate regression analysis. These optimal parameters are then fused with those optimized through numerical simulation, and a training dataset is generated based on the fused optimal parameters. This allows the GAN to learn the information about the optimal blasting hole mesh parameters during training, thereby improving the accuracy of the optimized hole mesh parameter optimization model. This provides strong data support for subsequent engineering construction, ensuring the safety and economy of the construction process.

[0045] The method for optimizing blasting hole mesh parameters based on image recognition provided in this application will be illustrated below with reference to specific embodiments.

[0046] like Figure 1 As shown in the embodiments of this application, the method for optimizing blasting hole mesh parameters based on image recognition includes the following steps:

[0047] Step 11: Obtain images and feature information of multiple excavation sections.

[0048] The excavation section described above is the section that requires drilling and blasting construction; the images described above can be obtained by taking pictures with a drone; the feature information described above includes rock type, joint and fracture characteristics, and cross-sectional geometric parameters. Joint and fracture characteristics refer to the features of joints and fractures, and cross-sectional geometric parameters refer to the geometric dimensions of the cross-section.

[0049] In some embodiments of this application, images of different types of rocks stored in the database can be compared with images of the excavation section in step 11 to determine the rock type of the excavation section; joint and fracture characteristics and cross-sectional geometric parameters can be obtained by scanning the excavation section using a three-dimensional laser scanner.

[0050] Step 12: For each excavation section, obtain the blasting hole network parameters based on the characteristic information of the excavation section; use numerical simulation to calculate the blasting effect index of blasting the excavation section based on the blasting hole network parameters, and optimize and adjust the blasting hole network parameters multiple times, calculating the blasting effect index of blasting the excavation section based on the blasting hole network parameters after each optimization and adjustment; based on the pre-constructed multiple regression analysis model, all blasting effect indices obtained by numerical simulation, and the blasting hole network parameters corresponding to each blasting effect index, determine the optimal blasting hole network parameters from the blasting hole network parameters and the blasting hole network parameters after each optimization and adjustment; and fuse the optimal blasting hole network parameters and the blasting hole network parameters after each optimization and adjustment.

[0051] That is, for each excavation section, the process in step 12 above needs to be executed to obtain multiple fusion results corresponding to each excavation section. These fusion results refer to the parameters obtained by fusing the optimal blasting hole network parameters and the blasting hole network parameters optimized and adjusted by numerical simulation. It should be noted that the optimal blasting hole network parameters are the blasting hole network parameters that enable the blasting effect to achieve the expected results.

[0052] In some embodiments of this application, the aforementioned blasting hole network parameters include the hole diameter, hole depth, and number of holes. The hole diameter is primarily determined by the rock drilling equipment. Currently, most horizontal tunnel excavation uses handheld and pneumatic rock drills, with two main types of hole diameters: standard (40-42mm) and small (34-35mm). The hole depth can be initially calculated based on the requirements of the blasting operation or the time required to complete one tunneling cycle. Given the conditions of commonly used rock drills, if a small-diameter hole is selected, a shallower hole depth is preferable; if a standard-diameter hole is selected, the hole depth is determined by… Figure 2 Confirmed. It should be noted that... Figure 2 The tunneling cross-section refers to the area of ​​the excavated cross-section.

[0053] The number of blast holes and the total number of blast holes on the working face of the roadway are related to the cross-sectional size, as well as factors such as rock properties, blast hole diameter, and unit explosive consumption. They can be calculated based on the characteristic information of the excavation cross-section. Specifically, the process of obtaining the blasting hole network parameters based on the characteristic information of the excavation cross-section is as follows: using the formula... The number of boreholes N is calculated. In the formula, f represents the rock firmness coefficient, and S... 2 S represents the area of ​​the tunnel excavation cross-section (i.e., the excavation cross-section), where f can be calculated based on rock type, joint and fracture characteristics, and basic physical and mechanical parameters. 2 It can be calculated based on the cross-sectional geometric parameters. It should be noted that the rock firmness coefficient and the area of ​​the tunnel excavation cross-section can be calculated using existing empirical formulas, and the specific calculation process will not be elaborated on here.

[0054] In some embodiments of this application, the specific implementation method of optimizing and adjusting the blasting hole network parameters using numerical simulation to calculate the blasting effect index of the excavation section based on the blasting hole network parameters before and after adjustment can be as follows: A finite element method fluid-structure interaction model is established using existing basic rock mechanics parameters and hole network parameters (i.e., the characteristic information of the excavation section and the blasting hole network parameters obtained before numerical simulation). The blasting process is solved using LS-DYNA series software. The number of damaged elements is derived from the solution results, and the effective damage rate is calculated to represent the effect of this blasting (i.e., the blasting effect index). Parameters are initially optimized based on the different effects. Specifically, the parameters of the rock damage constitutive model are corrected from the field blasting results to obtain model parameters that represent the field blasting situation; the models and state equations of the explosive and air are selected from commonly used data from the engineering field.

[0055] In some embodiments of this application, the aforementioned blasting effect index can be the damage rate (i.e., blasting damage rate), and the formula for calculating the damage rate is:

[0056]

[0057] Where D represents the damage rate, N d This represents the number of mesh elements in the rock model that reach the damage threshold during numerical simulation, while N represents the total number of mesh elements in the rock model during numerical simulation. The rock model mentioned above refers to the model constructed using basic rock mechanical parameters and pore mesh parameters during numerical simulation. d Both N and N are data output from the numerical simulation process.

[0058] In some embodiments of this application, the expression for the above-mentioned multiple regression analysis model is as follows:

[0059] y = β0 + β1x1 + β2x2 + β3x3 + ε

[0060] Where y represents the blasting effect index, β0 represents the constant term, β1, β2, and β3 all represent regression coefficients, x1, x2, and x3 represent the borehole diameter, borehole depth, and number of boreholes, respectively, and ε is the error term.

[0061] After constructing the aforementioned multiple regression analysis model, based on the pre-constructed multiple regression analysis model, all blasting effect indices obtained using numerical simulation methods, and the blasting hole mesh parameters corresponding to each blasting effect index, the specific implementation steps for determining the optimal blasting hole mesh parameters from the blasting hole mesh parameters and the blasting hole mesh parameters after each optimization adjustment are as follows:

[0062] Step 12.1: Substitute all the blasting effect indices obtained by numerical simulation methods, as well as the blasting hole mesh parameters corresponding to each blasting effect index, into the multiple regression analysis model to obtain multiple equations for solving.

[0063] Step 12.2: Solve the multiple equations to obtain the values ​​of the constant term, regression coefficients and error term of the multiple regression analysis model, and update the multiple regression analysis model using the obtained values.

[0064] In some optional embodiments, statistical analysis software such as SPSS or Excel can be used to solve multiple equations to obtain the values ​​of the constant term, regression coefficients, and error term of the multiple regression analysis model. After obtaining these values, the constant term, regression coefficients, and error term in the constructed multiple regression analysis model are replaced with the corresponding values ​​to obtain the updated multiple regression analysis model.

[0065] Step 12.3: Using the updated multiple regression analysis model, calculate the blasting effect index of the blasting hole mesh parameters and the blasting hole mesh parameters after each optimization and adjustment.

[0066] For each optimized and adjusted blasting hole network parameter obtained based on feature information, the value of the blasting hole network parameter must be substituted into the updated multiple regression analysis model to calculate the blasting effect index (i.e., y in the above multiple regression analysis model) corresponding to the blasting hole network parameter.

[0067] Step 12.4: The blasting hole mesh parameters corresponding to the blasting effect indices that are greater than the preset threshold among the calculated blasting effect indices are taken as the optimal blasting hole mesh parameters.

[0068] After determining the updated multiple regression analysis model, the goodness of fit of the model can be judged by calculating indicators such as the coefficient of determination and mean square error of the regression model, ensuring that the optimal blasting hole mesh parameters that achieve the expected blasting damage rate can be obtained through multiple regression analysis. That is, after calculating the blasting effect index (i.e., damage rate) corresponding to each blasting hole mesh parameter using the updated multiple regression analysis model, the blasting hole mesh parameters corresponding to the blasting effect index exceeding a preset threshold can be taken as the optimal blasting hole mesh parameters. The optimal blasting hole mesh parameters are those that achieve the expected blasting damage rate.

[0069] In some embodiments of this application, the specific implementation of fusing the optimal blasting hole mesh parameters and the blasting hole mesh parameters after each optimization adjustment is as follows: for each optimized blasting hole mesh parameter, the optimal blasting hole mesh parameters and the optimized blasting hole mesh parameters are fused using a fusion formula.

[0070] The above fusion formula is:

[0071] θ fused =θ0·ω0+θ1·ω1

[0072] Where, θ fused The parameters of the fused blasting hole mesh are denoted as follows: θ0 represents the optimal blasting hole mesh parameters, ω0 represents the weight of θ0, and θ1 represents the optimized blasting hole mesh parameters, with ω1 representing the weight of θ1. It should be noted that ω0 and ω1 are weights determined based on different data and can be adjusted according to the implementation. It should also be noted that θ... fused θ0 and θ1 can both be in matrix form, containing information such as borehole diameter, borehole depth and number of boreholes.

[0073] Step 13: Generate a training dataset based on images of multiple excavation sections, all fusion results corresponding to multiple excavation sections, and blasting effect indices corresponding to the blasting hole mesh parameters after each optimization and adjustment.

[0074] In some embodiments of this application, the above fusion result is the fused blast hole mesh parameters, and the training dataset includes multiple training samples.

[0075] In some optional embodiments, for each of the multiple excavation sections, the following steps are performed: for each fusion result corresponding to the excavation section, the blasting effect index corresponding to the fusion result is determined from all blasting effect indices obtained by numerical simulation methods, and the image of the excavation section, the fusion result, and the determined blasting effect index are used as a training sample.

[0076] It is worth mentioning that the training dataset mentioned above takes into account both the results of numerical simulation and regression analysis, and has strong reliability. It can better reflect the actual blasting situation, thereby improving the training effect and generalization ability of the generative adversarial network (GAN) model in subsequent steps.

[0077] Step 14: Train the generative adversarial network using the training dataset to obtain the hole mesh parameter optimization model.

[0078] As an optional example, the generative adversarial network described above could be a pix2pix model.

[0079] In actual training, the training dataset can be divided into two parts in a 7:3 ratio, with 70% of the data used for training and the remaining data used for testing. It should be noted that common training methods for generative adversarial networks (GANs) can be used for training and testing to achieve rapid prediction of blasting hole mesh parameters under different cross-sectional conditions.

[0080] Step 15: Optimize the blasting hole mesh parameters of the excavation section using the hole mesh parameter optimization model.

[0081] In some embodiments of this application, after training the hole mesh parameter optimization model, the image of the excavation section requiring hole mesh parameter optimization is input into the hole mesh parameter optimization model for processing, and the hole mesh parameter optimization model can then output the blasting hole mesh parameters of the excavation section. Alternatively, the image of the excavation section, some blasting hole mesh parameters (which can be obtained based on the feature information of the excavation section), and the desired blasting effect index can be input into the hole mesh parameter optimization model for processing to obtain the blasting hole mesh parameters of the excavation section.

[0082] It is worth mentioning that, during the acquisition of the training dataset, the optimal blasting hole mesh parameters were determined through multivariate regression analysis. These optimal parameters were then fused with those optimized through numerical simulation. The training dataset was generated based on these fused optimal parameters. This allows the generative adversarial network (GAN) to learn the information about the optimal blasting hole mesh parameters when training on the training dataset. Consequently, the trained hole mesh parameter optimization model improves the accuracy of blasting hole mesh parameter optimization, providing strong data support for subsequent engineering construction and ensuring construction safety.

[0083] Furthermore, since this application obtains a hole mesh parameter optimization model by training a generative adversarial network using the acquired training dataset, and then uses this hole mesh parameter optimization model to optimize the blasting hole mesh parameters of the excavation section, this method of optimizing blasting hole mesh parameters based on machine learning technology reduces the reliance on expert experience and avoids the influence of expert subjective factors, thereby improving the scientific nature of blasting hole mesh parameter optimization.

[0084] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. An image recognition-based blast hole pattern parameter optimization method, characterized in that, The method comprises the following steps: obtaining images and feature information of a plurality of excavation sections; for each excavation section, obtaining blast hole pattern parameters based on the feature information of the excavation section; using a numerical simulation method, calculating a blasting effect index of blasting the excavation section based on the blast hole pattern parameters, and performing multiple optimization adjustments on the blast hole pattern parameters, calculating a blasting effect index of blasting the excavation section based on the blast hole pattern parameters after each optimization adjustment; based on a pre-constructed multiple regression analysis model, all blasting effect indexes obtained by using a numerical simulation method, and the blast hole pattern parameters corresponding to each blasting effect index, determining the optimal blast hole pattern parameters from the blast hole pattern parameters and the blast hole pattern parameters after each optimization adjustment; fusing the optimal blast hole pattern parameters and the blast hole pattern parameters after each optimization adjustment; generating a training data set based on the images of the plurality of excavation sections, all fusion results corresponding to the plurality of excavation sections, and the blasting effect indexes corresponding to the blast hole pattern parameters after each optimization adjustment; training a generative adversarial network using the training data set to obtain a blast hole pattern optimization model; optimizing the blast hole pattern parameters of the excavation section using the blast hole pattern optimization model; The blast hole pattern parameters include hole diameter, hole depth and hole number; the fusion result is the fused blast hole pattern parameters, and the training data set includes a plurality of training samples; The training data set is generated based on the images of the plurality of excavation sections, all fusion results corresponding to the plurality of excavation sections, and the blasting effect indexes corresponding to the blast hole pattern parameters after each optimization adjustment, comprising: for each excavation section in the plurality of excavation sections, the following steps are performed: for each fusion result corresponding to the excavation section, determine the blasting effect index corresponding to the fusion result from all blasting effect indexes obtained by using a numerical simulation method, and take the image of the excavation section, the fusion result and the determined blasting effect index as a training sample.

2. The blasthole pattern optimization method of claim 1, wherein, The feature information includes rock type, joint and fracture characteristics, and section geometric parameters.

3. The blasthole pattern optimization method of claim 1, wherein, The blasting effect index is damage rate, and the calculation formula of the damage rate is: ； wherein, D represents the damage rate, D represents the number of grid elements in the rock model that reach the damage threshold when simulated by the numerical simulation method, D represents the total number of grid elements in the rock model when simulated by the numerical simulation method.

4. The blasthole pattern optimization method of claim 3, wherein, The expression of the multiple regression analysis model is: ； wherein, represents the blasting effect index, represents the constant term, each represents the regression coefficient, respectively represents the blast hole diameter, the blast hole depth and the blast hole number, is the error term.

5. The blasthole pattern optimization method of claim 4, wherein, The optimal blast hole pattern parameters are determined from the blast hole pattern parameters and the blast hole pattern parameters after each optimization adjustment based on the pre-constructed multiple regression analysis model, all blasting effect indexes obtained by using a numerical simulation method, and the blast hole pattern parameters corresponding to each blasting effect index, comprising: substitute all blasting effect indexes obtained by using a numerical simulation method and the blast hole pattern parameters corresponding to each blasting effect index into the multiple regression analysis model to obtain a plurality of equations for solving; solve a plurality of equations to obtain the values of the constant term, the regression coefficient and the error term of the multiple regression analysis model, and update the multiple regression analysis model using the solved values; calculate the blasting effect indexes of the blast hole pattern parameters and the blast hole pattern parameters after each optimization adjustment using the updated multiple regression analysis model; ​ The blasting hole network parameter corresponding to the blasting effect index greater than the preset threshold in the calculated blasting effect index is taken as an optimal blasting hole network parameter.

6. The blasthole pattern optimization method of claim 5, wherein, The fusing of the optimal blasting hole network parameter and the blasting hole network parameter after each optimization adjustment comprises: The optimal blasting hole network parameter and the blasting hole network parameter after each optimization adjustment are fused through a fusion formula for each blasting hole network parameter after optimization adjustment; and the fusion formula is: ； wherein, represents the parameter of the blast hole network after fusion, represents the optimal blast hole network parameter, represents the weight of, represents the blast hole network parameter after optimization adjustment, represents the weight of.

Images