Directional shaped charge blasting optimization method based on big data analysis

Abstract

The invention relates to a directional shaped charge blasting optimization method based on big data analysis. The method comprises the steps that drilling parameters of a drilling machine in the tunnel blasting process are obtained; utilizing the drilling parameters to invert rock characteristics; determining blasting parameters based on the rock characteristics; and directional shaped charge blasting is carried out based on the blasting parameters, and the blasting effect is evaluated. By analyzing a large amount of data generated by the drilling machine in the drilling process, the blasting effect can be predicted more accurately, and therefore the blasting parameters are optimized. According to the method, the blasting efficiency can be improved, resource waste is reduced, and the safety of the blasting process is improved.

Description

Technical Field

[0001] The present invention relates to the field of blasting technology, and in particular to a directional focused energy blasting optimization method based on big data analysis. Background Art

[0002] Traditional directional focused blasting technology has some defects and challenges in engineering applications, which limit its application potential in a wider range of scenarios. For example: Traditional focused blasting technology has extremely high requirements for drilling accuracy and quality. Any deviation may affect the blasting effect, resulting in uneven rock crushing or uneven blasting surface. The selection of blasting parameters usually relies on experience and lacks scientific basis, resulting in unstable blasting effects, serious waste of resources, and safety hazards. In traditional blasting design, the strength, hardness and integrity of rocks are often ignored, resulting in the blasting design not being able to accurately match the actual rock conditions, affecting blasting efficiency and safety. Due to the lack of accurate prediction models, traditional blasting technology is difficult to predict key indicators such as over-excavation, under-excavation and residual rate of blast marks after blasting, resulting in difficulty in controlling the blasting effect.

[0003] In summary, in traditional directional focused blasting, the selection of parameters such as the explosive deployment position and the borehole diameter mainly relies on experience and lacks precise scientific basis. This leads to unstable blasting effects, serious waste of resources, and potential safety hazards. Summary of the invention

[0004] The purpose of the present invention is to propose a directional focused blasting optimization method based on big data analysis to solve the problems existing in the above-mentioned prior art. By analyzing a large amount of data generated by the drilling machine during the drilling process, the blasting effect can be predicted more accurately, thereby optimizing the blasting parameters. This method can improve blasting efficiency, reduce resource waste, and improve the safety of the blasting process.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A directional focused blasting optimization method based on big data analysis, comprising:

[0007] Obtain drilling parameters of drilling machine during tunnel blasting;

[0008] Inverting rock properties using the drilling parameters;

[0009] Determining blasting parameters based on the rock characteristics;

[0010] Directed focused energy blasting is performed based on the blasting parameters, and the blasting effect is evaluated.

[0011] Optionally, the drilling parameters include: rotation speed, torque, thrust, footage, vibration, and vibration speed of the drilling machine during the drilling process.

[0012] Optionally, using the drilling parameters, inverting rock properties includes:

[0013] Conducting experimental tests on rock samples to obtain rock properties of the rock samples; wherein the rock properties include: rock strength, hardness and integrity;

[0014] Collect drilling parameters of the drilling machine in the rock sample;

[0015] constructing a first data set based on rock properties of the rock sample and drilling parameters in the rock sample;

[0016] Preprocessing the first data set;

[0017] Using the preprocessed first data set to train and verify the preset fB-GPR model, to obtain a rock inversion model;

[0018] Based on the rock inversion model, rock properties are inverted.

[0019] Optionally, performing an experimental test on a rock sample to obtain rock properties of the rock sample includes:

[0020] Collect preset rock samples according to engineering requirements and geological conditions;

[0021] The rock samples are subjected to rock strength test, rock hardness test and rock integrity test respectively.

[0022] Optionally, training the fB-GPR model includes:

[0023] The model was trained using a supervised learning algorithm, with drilling machine parameters as input and physical properties of the rock as output;

[0024] The k-fold cross validation method was used to evaluate the performance of the model;

[0025] Use Bayesian optimization methods to optimize model hyperparameters.

[0026] Optionally, based on the rock characteristics, determining blasting parameters includes:

[0027] Based on the rock characteristics, for each blasting point, the propagation of the explosion energy in different rock media is simulated by a numerical simulation method to obtain the temporal and spatial distribution of the explosion energy in the rock medium;

[0028] An optimization algorithm is adopted, and the focusing degree of the explosion energy in the blasting target area is taken as the optimization target, and the optimal blasting parameters of each blasting point are determined through iterative search; wherein the blasting parameters include: explosive deployment position, blast hole diameter, and blast hole spacing;

[0029] constructing a second data set based on the rock characteristics and the optimal blasting parameters of each blasting point;

[0030] Using the second data set to train and verify a preset support vector machine to obtain a blasting prediction model;

[0031] The blasting prediction model is used to predict blasting parameters.

[0032] Optionally, determining the optimal blasting parameters of each blasting point through iterative search includes:

[0033] Based on the rock characteristics and the position and shape of the blasting target area, a three-dimensional space model is established, and the three-dimensional space model is divided into a plurality of grid units;

[0034] For each blasting point, the particle swarm optimization algorithm is used to establish an optimization model with the explosive deployment position, blast hole diameter, and blast hole spacing as optimization variables. The degree of focusing of the explosion energy in the target area is taken as the optimization target, and the optimal blasting parameter combination of each blasting point is determined through iterative search.

[0035] During the iteration process, according to the current blasting parameter combination, the finite element method is used to simulate the explosion process and calculate the distribution of explosion energy in each grid unit; the energy distribution of the grid units in the target area is used as an evaluation index of the focusing degree, and it is mapped to the optimization target value through the fitness function; according to the optimization target value, the particle swarm algorithm is used to update the parameters of the blasting point, and a new parameter combination is obtained to enter the next round of iteration; when the number of iterations reaches the preset threshold or the optimization target value converges, the explosive deployment position, blasthole diameter, and blasthole spacing of each blasting point are output.

[0036] The beneficial effects of the present invention are:

[0037] The present invention first obtains the drilling parameters of the drilling machine during the tunnel blasting process; then uses the drilling parameters to invert the rock characteristics; reduces the safety hazards in the blasting process by predicting the strength, hardness and integrity of the rock and optimizing the blasting parameters; secondly, determines the blasting parameters based on the rock characteristics; improves the blasting efficiency and reduces the waste of resources by optimizing the explosive deployment position, blasthole diameter and blasthole spacing; finally, performs directional focused blasting based on the blasting parameters and evaluates the blasting effect; further improves the blasting effect and ensures the quality of the project by evaluating the blasting effect, such as over-excavation, under-excavation, and residual rate of blast marks. This embodiment realizes data-driven optimization of blasting parameters through big data analysis and machine learning models, and improves the scientificity and accuracy of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 The present invention is a flowchart of a directional focused blasting optimization method based on big data analysis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] like Figure 1 As shown, this embodiment proposes a directional focused blasting optimization method based on big data analysis, including:

[0043] Obtain drilling parameters of drilling machine during tunnel blasting;

[0044] Inverting rock properties using the drilling parameters;

[0045] Determining blasting parameters based on the rock characteristics;

[0046] Directed focused energy blasting is performed based on the blasting parameters, and the blasting effect is evaluated.

[0047] Specifically, this embodiment uses a machine learning model to utilize parameters such as the rotation speed, torque, thrust, footage, vibration, and vibration velocity of the drilling machine during tunnel blasting to invert the strength, hardness, and integrity of the rock, and then determines the optimal parameters such as the explosive deployment position, blasthole diameter, and blasthole spacing, and evaluates the blasting effect, such as over-excavation, under-excavation, and blast mark residual rate.

[0048] Furthermore, the drilling parameters include: rotation speed, torque, thrust, footage, vibration, and vibration speed of the drilling machine during the drilling process.

[0049] Furthermore, using the drilling parameters, inverting rock properties includes:

[0050] Conducting experimental tests on rock samples to obtain rock properties of the rock samples; wherein the rock properties include: rock strength, hardness and integrity;

[0051] Collect drilling parameters of the drilling machine in the rock sample;

[0052] constructing a first data set based on rock properties of the rock sample and drilling parameters in the rock sample;

[0053] Preprocessing the first data set;

[0054] The fB-GPR model is trained and verified using the preprocessed first data set to obtain a rock inversion model;

[0055] Based on the rock inversion model, rock properties are inverted.

[0056] Among them, the fB-GPR (Fully Bayesian Gaussian Process Regression) model is a method based on Bayesian Gaussian Process Regression (GPR), which combines Bayesian theory with Markov Chain Monte Carlo simulation (MCMC) and is used to estimate the uniaxial compressive strength (UCS) of rock; the fB-GPR model can perform well in dealing with regression problems, especially when dealing with nonlinear relationships in data, by combining Gaussian process regression, Bayesian theory and MCMC simulation. The model shows higher accuracy and robustness, and can accurately select models even when the UCS measurement error is large; compared with other models such as Akaike Information Criterion, Bayesian Information Criterion, Bias Information Criterion, Kullback Information Criterion, etc., the fB-GPR method has a higher accuracy in selecting the optimal model, and the prediction results are more consistent with the actual situation.

[0057] Furthermore, experimental testing is performed on the rock sample to obtain the rock properties of the rock sample, including:

[0058] Collect preset rock samples according to engineering requirements and geological conditions;

[0059] The rock samples are subjected to rock strength test, rock hardness test and rock integrity test respectively.

[0060] Specifically, in this embodiment, sample collection: according to engineering requirements and geological conditions, appropriate rock samples are selected for collection, ensuring that the samples are representative and cover different rock types and weathering degrees.

[0061] Sample preparation: Process the collected rock samples into standard specimens. For example, for uniaxial compressive strength tests, it is usually necessary to process the rock into cylindrical or cubic specimens; according to the provisions of the "Rock Physical and Mechanical Properties Test Procedure", prepare specimens of different sizes and shapes to meet the needs of different tests.

[0062] Rock strength test: Use a rock mechanics testing machine to perform uniaxial compressive strength tests to measure the strength of rock under uniaxial pressure. Tests are conducted in accordance with DZ / T 0276.18-2015 standard; for tensile strength and flexural strength tests, special devices can be used, such as the Brazilian splitting method for tensile strength tests.

[0063] Rock hardness test: Different hardness test methods are used to evaluate the hardness of rocks, such as Mohs hardness, indentation hardness, Rockwell hardness, Vickers hardness, etc. Refer to ASTM D5873-2014 standard and SNI 13-6581-2001 standard for hardness test.

[0064] Rock integrity testing: Use sonic or ultrasonic testing to assess rock integrity. These methods measure the propagation speed and reflection characteristics of sound waves in rock to infer cracks and fractures in the rock.

[0065] Record all test data, including loading rate, maximum load, failure load, etc. Analyze the test results and calculate the average strength, hardness and integrity index of the rock.

[0066] Further, preprocessing the first data set includes:

[0067] The first data set is cleaned to process missing values ​​and outliers;

[0068] Normalize the cleaned data.

[0069] Specifically, in this embodiment, data preprocessing includes:

[0070] Clean the data and handle missing values ​​and outliers.

[0071] Normalize or standardize the data to make data of different dimensions comparable.

[0072] Feature engineering, extracting features that are useful to the model, such as extracting trends, periodicity, and other features from time series data.

[0073] Furthermore, training the fB-GPR model includes:

[0074] The fB-GPR model is trained using a supervised learning algorithm, taking the drilling machine parameters as input and the physical properties of the rock as output;

[0075] The k-fold cross validation method was used to evaluate the performance of the model;

[0076] Use Bayesian optimization methods to optimize model hyperparameters.

[0077] After the model is trained, the accuracy of the model is evaluated using the dataset using appropriate evaluation metrics such as mean square error (MSE), mean absolute error (MAE), coefficient of determination (R 2 ), etc. Analyze the generalization ability of the model to ensure that the model performs well on unseen data.

[0078] Further, based on the rock characteristics, determining blasting parameters includes:

[0079] Based on the rock characteristics, for each blasting point, the propagation of the explosion energy in different rock media is simulated by a numerical simulation method to obtain the temporal and spatial distribution of the explosion energy in the rock medium;

[0080] An optimization algorithm is adopted, and the focusing degree of the explosion energy in the blasting target area is taken as the optimization target, and the optimal blasting parameters of each blasting point are determined through iterative search; wherein the blasting parameters include: explosive deployment position, blast hole diameter, and blast hole spacing;

[0081] constructing a second data set based on the rock characteristics and the optimal blasting parameters of each blasting point;

[0082] Using the second data set to train and verify the support vector machine to obtain a blasting prediction model;

[0083] The blasting prediction model is used to predict blasting parameters.

[0084] Furthermore, the optimal blasting parameters for each blasting point are determined by iterative search, including:

[0085] Based on the rock characteristics and the position and shape of the blasting target area, a three-dimensional space model is established, and the three-dimensional space model is divided into a plurality of grid units;

[0086] For each blasting point, the particle swarm optimization algorithm is used to establish an optimization model with the explosive deployment position, blast hole diameter, and blast hole spacing as optimization variables. The degree of focusing of the explosion energy in the target area is taken as the optimization target, and the optimal blasting parameter combination of each blasting point is determined through iterative search.

[0087] In the iterative process, according to the current blasting parameter combination, the finite element method is used to simulate the explosion process and calculate the distribution of the explosion energy in each grid unit. The energy distribution of the grid units in the target area is used as the evaluation index of the focusing degree, and it is mapped to the optimization target value through the fitness function; according to the optimization target value, the particle swarm algorithm is used to update the parameters of the blasting point, and a new parameter combination is obtained to enter the next round of iteration; when the number of iterations reaches the preset threshold or the optimization target value converges, the explosive deployment position, blasthole diameter, and blasthole spacing of each blasting point are output.

[0088] Specifically, in this embodiment, for example, according to the three-dimensional space model of the target area, the hexahedral meshing method is used to divide the model into 50×50×50 grid units. For the preset 10 blasting points, the explosive deployment position (x, y, z), blasthole diameter D, and blasthole spacing Z of each blasting point are used as optimization variables to establish a particle swarm optimization model. The fitness function f(x, y, z, D, Z) is defined to represent the degree of focusing of the explosion energy in the target area, and the goal is to maximize the f value. Initialize 50 particles, each particle contains the parameters of 10 blasting points, and randomly generate the initial position and speed. In each iteration, according to the current position of the particle, the finite element method is used to simulate the explosion process, calculate the distribution of the explosion energy in each grid unit, and take the average energy value of the grid unit in the target area as the f value. According to the f value, the speed and position of the particle are updated, and the historical optimal solution and the global optimal solution of each particle are recorded at the same time. When the number of iterations reaches 1000 or the global optimal solution of 50 consecutive iterations changes less than 0.1, the iteration is terminated and the global optimal solution is output as the optimal parameter combination of each blasting point. Through numerical simulation verification, the optimized blasting parameters can increase the average energy in the target area by 30%, make the energy distribution more uniform, and achieve effective focusing of the explosion energy. This embodiment can improve blasting efficiency and reduce resource waste by optimizing the explosive deployment position, blast hole diameter and blast hole spacing.

[0089] This embodiment uses the blasting parameters predicted by the model to carry out actual blasting, collects effect data after blasting, and evaluates the blasting effect, such as overbreak, underbreak, and residual rate of blast marks. The optimized blasting parameters and blasting effect evaluation results are output in the form of a report for reference and implementation by blasting engineers. Based on the field verification results, the model parameters are continuously adjusted and optimized.

[0090] This embodiment first obtains the drilling parameters of the drilling machine during the tunnel blasting process; then uses the drilling parameters to invert the rock characteristics; reduces the safety hazards in the blasting process by predicting the strength, hardness and integrity of the rock and optimizing the blasting parameters; secondly, determines the blasting parameters based on the rock characteristics; improves the blasting efficiency and reduces the waste of resources by optimizing the explosive deployment position, blasthole diameter and blasthole spacing; finally, performs directional focused blasting based on the blasting parameters and evaluates the blasting effect; further improves the blasting effect and ensures the quality of the project by evaluating the blasting effect, such as over-excavation, under-excavation, and residual rate of blast marks. This embodiment uses big data analysis and machine learning models to achieve data-driven optimization of blasting parameters and improve the scientificity and accuracy of decision-making.

[0091] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A directional focused blasting optimization method based on big data analysis, characterized in that: include: Obtain drilling parameters of drilling machine during tunnel blasting; Inverting rock properties using the drilling parameters; Determining blasting parameters based on the rock characteristics; Directed focused energy blasting is performed based on the blasting parameters, and the blasting effect is evaluated.

2. The directional focused blasting optimization method based on big data analysis according to claim 1 is characterized in that: The drilling parameters include: the rotation speed, torque, thrust, penetration, vibration, and vibration speed of the drilling machine during the drilling process.

3. The directional focused blasting optimization method based on big data analysis according to claim 1 is characterized in that: Using the drilling parameters, the rock properties are inverted including: Performing experimental tests on rock samples to obtain rock properties of the rock samples; wherein the rock properties include: rock strength, hardness and integrity; Collect drilling parameters of the drilling machine in the rock sample; constructing a first data set based on rock properties of the rock sample and drilling parameters in the rock sample; Preprocessing the first data set; The pre-processed first data set is used to train and verify the preset full Bayesian Gaussian process regression model to obtain a rock inversion model; Based on the rock inversion model, rock properties are inverted.

4. The directional focused blasting optimization method based on big data analysis according to claim 3 is characterized in that: Experimental testing is performed on rock samples to obtain rock properties of the rock samples including: Collect preset rock samples according to engineering requirements and geological conditions; The rock samples are subjected to rock strength test, rock hardness test and rock integrity test respectively.

5. The directional focused blasting optimization method based on big data analysis according to claim 1 is characterized in that: Training the preset full Bayesian Gaussian process regression model includes: The model was trained using a supervised learning algorithm, with drilling machine parameters as input and physical properties of the rock as output; The k-fold cross validation method was used to evaluate the performance of the model; Use Bayesian optimization methods to optimize model hyperparameters.

6. The directional focused blasting optimization method based on big data analysis according to claim 4 is characterized in that: Based on the rock characteristics, blasting parameters are determined including: Based on the rock characteristics, for each blasting point, the propagation of the explosion energy in different rock media is simulated by a numerical simulation method to obtain the temporal and spatial distribution of the explosion energy in the rock medium; An optimization algorithm is adopted, and the focusing degree of the explosion energy in the blasting target area is taken as the optimization target, and the optimal blasting parameters of each blasting point are determined through iterative search; wherein the blasting parameters include: explosive deployment position, blast hole diameter, and blast hole spacing; constructing a second data set based on the rock characteristics and the optimal blasting parameters of each blasting point; Using the second data set to train and verify a preset support vector machine to obtain a blasting prediction model; The blasting prediction model is used to predict blasting parameters.

7. The directional focused blasting optimization method based on big data analysis according to claim 6 is characterized in that: The optimal blasting parameters for each blasting point determined by iterative search include: Based on the rock characteristics and the position and shape of the blasting target area, a three-dimensional space model is established, and the three-dimensional space model is divided into a plurality of grid units; For each blasting point, the particle swarm optimization algorithm is used to establish an optimization model with the explosive deployment position, blast hole diameter, and blast hole spacing as optimization variables. The degree of focusing of the explosion energy in the target area is taken as the optimization target, and the optimal blasting parameter combination of each blasting point is determined through iterative search. During the iteration process, according to the current blasting parameter combination, the finite element method is used to simulate the explosion process and calculate the distribution of explosion energy in each grid unit; the energy distribution of the grid units in the target area is used as an evaluation index of the focusing degree, and it is mapped to the optimization target value through the fitness function; according to the optimization target value, the particle swarm algorithm is used to update the parameters of the blasting point, and a new parameter combination is obtained to enter the next round of iteration; when the number of iterations reaches the preset threshold or the optimization target value converges, the explosive deployment position, blasthole diameter, and blasthole spacing of each blasting point are output.

Images