A method for determining the charge quantity of tunneling blasting

Abstract

The application provides a tunneling blasting charge quantity determination method, which comprises the following steps: 1) determining the explosive unit consumption based on an improved neural network model of KNN algorithm; 2) determining the total charge quantity of tunneling blasting according to the volume of the rock mass of tunneling blasting; 3) calculating the charge quantity of peripheral holes; 4) calculating the charge quantity of cutting holes; and 5) calculating the charge quantity of auxiliary holes according to the total charge quantity of tunneling blasting, the charge quantity of peripheral holes and the charge quantity of cutting holes. The application provides a tunneling blasting charge quantity determination method, which can quickly determine the charge quantity of different blast holes, improve the utilization rate of blast holes, the blasting effect and the intelligence and refinement of tunneling blasting design.

Description

Technical Field

[0001] This invention belongs to the field of blasting engineering, and specifically relates to a method for determining the amount of explosive charge in tunneling blasting. Background Technology

[0002] Tunneling blasting is a common construction method in excavation projects such as tunnel construction and mining. Due to the complexity of the geological environment of the construction site, blasting parameters have a significant impact on project quality and construction safety. Excessive charge can easily lead to over-excavation, rock collapse, damage to anchor bolts and linings, and increased hazards such as blasting vibration, flyrock, and dust. Insufficient charge will result in insufficient fragmentation of the surrounding rock, hindering lining construction, reducing construction efficiency, and increasing construction costs. Currently, the determination of charge amounts mostly relies on experience, human judgment based on geological survey data, or on-site test blasts. This method lacks universality and is difficult to guarantee sufficient rock fragmentation and construction safety in complex geological environments.

[0003] In recent years, with the gradual application of artificial intelligence in blasting engineering, intelligent prediction and optimization of blasting effects have been gradually realized by establishing mapping models between charge quantity and geological conditions, and between blasting parameters and blasting responses. However, neural network models are quite sensitive to the distribution of training samples and are easily affected by small and outlier samples, resulting in large errors in charge quantity prediction. Summary of the Invention

[0004] This invention addresses the problems existing in the prior art by providing a method for determining the charge quantity in tunneling blasting. This method uses the KNN algorithm to select samples based on feature similarity in the sample space, and selects samples with composite predicted target working conditions as learning samples for the neural network model. This overcomes the shortcomings of relying on traditional empirical formulas and single neural network models, and realizes intelligent determination and dynamic adjustment of the charge quantity, thereby providing a more scientific and reasonable data basis for tunneling blasting.

[0005] To solve the above technical problems, the present invention provides the following technical solution: a method for determining the amount of explosive charge in tunneling blasting, comprising the following steps:

[0006] S1. Based on the KNN algorithm, determine the training sample that is closest to the predicted target;

[0007] S2. Using various influencing factors as input and the unit consumption of explosives for tunneling blasting as output, construct and train a model for determining the amount of explosives used in tunneling blasting; use this model to obtain the unit consumption of explosives for tunneling blasting that meets the preset accuracy requirements.

[0008] S3. Based on the unit consumption of explosives for tunneling blasting in step S3, determine the total charge amount for tunneling blasting according to the volume of rock mass in the tunneling blasting.

[0009] S4. Based on the unit consumption of blasting explosives in step S3, determine the amount of explosives to be charged in the surrounding holes.

[0010] S5. Determine the charge amount for the slotted hole;

[0011] S6. Calculate the charge amount for auxiliary holes based on the total charge amount for tunneling blasting, the charge amount for peripheral holes, and the charge amount for slotting holes.

[0012] Furthermore, the aforementioned step S1 includes the following sub-steps:

[0013] S1.1 Based on historical blasting records and surveys, collect geological and blasting design information for different blasting projects, and construct a sample library for predicting explosive consumption per unit; establish a set of influencing factors on explosive consumption per unit in tunneling blasting, and construct a sample dataset;

[0014] S1.2 Normalize the sample dataset. And generate feature vectors from the normalized sample data. In the formula It is the first Group 1 The data after column normalization It is the first Group 1 The data in the column, , They are the first The largest and smallest data in the column, It is the first Feature vectors generated from grouped normalized data. For the first Grouped normalized data;

[0015] S1.3, Establish the relationship between the sample to be tested and the first... Euclidean distance between groups of samples The Euclidean distance is used to determine the distance between the sample data and the target data, and the nearest sample data to the target data is retrieved as the training data sample. The sample to be tested and the first Euclidean distance between groups of samples It is a feature vector generated from the target data set. Weighting coefficients for each influencing factor;

[0016] S1.4. Based on the relationship between explosive consumption and engineering geological conditions and blasting design parameters, preliminary selection of weights is made;

[0017] The engineering geological conditions mentioned include the integrity coefficient of the rock mass, the elastic modulus of the rock mass, the tensile strength of the rock mass, the cross-sectional area, and the tunneling depth. The blasting design parameters include the borehole diameter, borehole depth, borehole spacing, and row spacing.

[0018] S1.5. Retrieve the training sample in the sample dataset that is closest to the prediction target, i.e., the sample with the smallest Euclidean distance.

[0019] Furthermore, the aforementioned step S2 includes the following steps:

[0020] S2.1 Determine the number of neurons in the input layer, the number of neurons in the hidden layer, and the number of hidden layers, and simultaneously assign connection weights between each neuron. Hidden layer threshold Input layer threshold Perform initialization and set the learning rate. and excitation function ;

[0021] S2.2 Calculate the hidden layer output: based on the input value Connection weights between the input layer and the hidden layer and hidden layer threshold Through formula Calculate the output value of the hidden layer. ,

[0022] S2.3 Calculate the output of the output layer: based on the output values ​​of the hidden layers. Connection weights between hidden layers and output layers and output layer threshold The predicted output value of the neural network is calculated as follows: ;

[0023] S2.4 Calculation Error: Based on the predicted output value and the expected output value Y of the neural network, calculate the prediction error: ;

[0024] S2.5 Update weights: Update and correct each weight based on the error calculated in step S2.4. ;

[0025] S2.6, Update Thresholds: Update and correct each threshold according to the error. ;

[0026] S2.7 Determine if training is complete. Recalculate the output error based on the updated weights and check if the output result meets expectations. If the accuracy is not met, repeat steps S2.2 to S2.6 until the output meets the accuracy requirements for the unit consumption of tunneling blasting explosives. .

[0027] Furthermore, the aforementioned step S3 includes the following sub-steps:

[0028] S3.1 Obtain the volume of the blasted rock mass during tunneling. ;

[0029] S3.2 Calculate the total charge amount for tunneling blasting based on the volume of the blasted rock mass. .

[0030] Furthermore, the aforementioned step S4 includes the following sub-steps:

[0031] S4.1 Obtain the linear charge density of the surrounding holes according to the blasting design. ;

[0032] S4.2 Obtain the length of the surrounding holes ;

[0033] S4.3 Calculate the charge amount for a single peripheral hole. In the formula for The length of each peripheral hole;

[0034] S4.4 Calculate the total charge amount for the peripheral holes. .

[0035] Furthermore, the aforementioned step S5 includes the following sub-steps:

[0036] S5.1 Calculate the rock mass volume within the excavation area. In the formula The area of ​​the cut section is... The length of the cut hole;

[0037] S5.2 Determine the charge density of the cut hole according to the blasting design. ;

[0038] S5.3 Calculate the charge amount for a single cut hole ;

[0039] S5.4 Calculate the total charge amount in the cut hole. .

[0040] Furthermore, the aforementioned step S6 includes the following sub-steps:

[0041] S6.1 Calculate the total charge amount of the auxiliary holes based on the total charge amount and the total charge amount of the peripheral holes and the slotted holes. ;

[0042] S6.2 Determine the charge amount for a single auxiliary hole based on the number of auxiliary holes, n. .

[0043] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described in the present invention.

[0044] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described in the present invention.

[0045] Compared with the prior art, the beneficial technical effects of the present invention using the above technical solution are as follows:

[0046] Before using a neural network model for learning, this invention employs the KNN algorithm to select learning samples. This allows for the retrieval of learning samples that meet the prediction target based on the current working conditions. When modifying the weights and thresholds of the neural network model, four different parameters are used to control the increments of the weights and thresholds, which can more accurately predict the explosive consumption per unit of the current working conditions. Attached Figure Description

[0047] Figure 1 This is a flowchart of the present invention.

[0048] Figure 2 This is a schematic diagram of the structure of the prediction model for the unit consumption of explosives in tunneling blasting. Detailed Implementation

[0049] To better understand the technical content of the present invention, specific embodiments are described below in conjunction with the accompanying drawings.

[0050] In this invention, various aspects of the invention are described with reference to the accompanying drawings, which illustrate numerous illustrative embodiments. Embodiments of the invention are not limited to those shown in the drawings. It should be understood that the invention is implemented through any of the various concepts and embodiments described above, as well as the concepts and embodiments described in detail below, because the concepts and embodiments disclosed in this invention are not limited to any particular implementation. Furthermore, some aspects of the invention disclosed can be used alone or in any suitable combination with other aspects of the invention disclosed.

[0051] like Figure 1-2 As shown in the figure, this embodiment provides a method for determining the amount of explosive charge in tunneling blasting, including the following steps:

[0052] S1. Determine the training samples closest to the predicted target based on the KNN (K-Nearest Neighbor) algorithm:

[0053] S1.1 Based on historical blasting records and surveys, collect geological and blasting design information for different blasting projects, and construct a sample library for predicting explosive consumption per unit.

[0054] S1.2 Establish a set of factors affecting the unit consumption of explosives in tunneling blasting, and construct a sample dataset;

[0055] S1.3 Normalize the sample dataset. And generate feature vectors from the normalized sample data. In the formula It is the first Group 1 The data after column normalization It is the first Group 1 The data in the column, , They are the first The largest and smallest data in the column, It is the first Feature vectors generated from grouped normalized data. For the first Grouped normalized data;

[0056] S1.4, Establish the relationship between the sample to be tested and the first... Euclidean distance between groups of samples The Euclidean distance is used to determine the distance between the sample data and the target data, and the nearest sample data to the target data is retrieved as the training data sample. The sample to be tested and the first Euclidean distance between groups of samples It is a feature vector generated from the target data set. Weighting coefficients for each influencing factor;

[0057] S1.5. Based on the relationship between explosive consumption and various influencing factors, make preliminary selection of weights;

[0058] S1.6 Retrieve the training sample in the sample dataset that is closest to the prediction target, i.e. the sample with the smaller Euclidean distance;

[0059] S2. Determination of explosive consumption based on an improved neural network model:

[0060] like Figure 2 The tunneling blasting explosive consumption prediction model structure shown in step S2 includes the following steps:

[0061] S2.1 Determine the number of neurons in the input layer, the number of neurons in the hidden layer, and the number of hidden layers, and simultaneously assign connection weights between each neuron. Hidden layer threshold Input layer threshold Perform initialization and set the learning rate. and excitation function Schematic diagram of the prediction model for explosive consumption per unit of tunneling blasting

[0062] S2.2 Calculate the hidden layer output: based on the input value Connection weights between the input layer and the hidden layer and hidden layer threshold Through formula Calculate the output value of the hidden layer. ,

[0063] S2.3 Calculate the output of the output layer: based on the output values ​​of the hidden layers. Connection weights between hidden layers and output layers and output layer threshold The predicted output value of the neural network is calculated as follows: ;

[0064] S2.4 Calculation Error: Based on the predicted output value and the expected output value Y of the neural network, calculate the prediction error: ;

[0065] S2.5 Update weights: Update and correct each weight based on the error calculated in step S2.4. ;

[0066] S2.6, Update Thresholds: Update and correct each threshold according to the error. ;

[0067] S2.7 Determine if training is complete. Recalculate the output error based on the updated weights and check if the output result meets expectations. If the accuracy is not met, repeat steps S2.2 to S2.6 until the output meets the accuracy requirements for the unit consumption of tunneling blasting explosives. .

[0068] S3. Determine the total charge amount for tunneling blasting based on the volume of the rock mass being blasted.

[0069] S3.1 Obtain the volume of the blasted rock mass during tunneling. ;

[0070] S3.2 Calculate the total charge amount for tunneling blasting based on the volume of the blasted rock mass. .

[0071] S4. Determine the charge amount for the peripheral holes:

[0072] S4.1 Obtain the linear charge density of the surrounding holes according to the blasting design. ;

[0073] S4.2 Obtain the length of the surrounding holes ;

[0074] S4.3 Calculate the charge amount for a single peripheral hole. In the formula for The length of each peripheral hole;

[0075] S4.4 Calculate the total charge amount for the peripheral holes. ;

[0076] S5. Determine the charge amount for the cut hole:

[0077] S5.1 Calculate the rock mass volume within the excavation area. In the formula The area of ​​the cut section is... The length of the cut hole;

[0078] S5.2 Determine the charge density of the cut hole based on the blasting design. ;

[0079] S5.3 Calculate the charge amount for a single cut hole ;

[0080] S5.4 Calculate the total charge amount in the cut holes. .

[0081] S6. Calculate the auxiliary hole charge based on the total charge of the tunneling blasting, the charge of the peripheral holes, and the charge of the cut holes:

[0082] S6.1 Calculate the total charge amount of the auxiliary holes based on the total charge amount and the total charge amount of the peripheral holes and the slotted holes. ;

[0083] S6.2 Determine the charge amount for a single auxiliary hole based on the number of auxiliary holes, n. .

[0084] Another aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described in the embodiments.

[0085] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods described in the embodiments.

[0086] While the present invention has been described above with reference to preferred embodiments, it is not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A method for determining the amount of explosive charge in tunneling blasting, characterized in that, The steps are as follows: S1. Based on the KNN algorithm, the similarity of training samples of historical tunneling and blasting data is calculated to determine the training sample set that is most similar to the predicted target working condition; S2. Using the influencing factors of the amount of explosive charge in tunneling blasting as input and the unit consumption of explosive charge in tunneling blasting as output, construct and train a model for determining the amount of explosive charge in tunneling blasting; use this model to obtain the unit consumption of explosive charge in tunneling blasting that meets the preset accuracy requirements; S3. Based on the unit consumption of explosives for tunneling blasting in step S3, determine the total charge amount for tunneling blasting according to the volume of rock mass in the tunneling blasting. S4. Based on the unit consumption of blasting explosives in step S3, determine the amount of explosives to be charged in the surrounding holes. S5. Determine the charge amount for the slotted hole; S6. Calculate the charge amount for auxiliary holes based on the total charge amount for tunneling blasting, the charge amount for peripheral holes, and the charge amount for slotting holes.

2. The method for determining the amount of explosive charge in tunneling blasting according to claim 1, characterized in that, Step S1 includes the following sub-steps: S1.1 Based on historical blasting records and surveys, collect geological and blasting design information for different blasting projects, and construct a sample library for predicting explosive consumption per unit; establish a set of influencing factors on explosive consumption per unit in tunneling blasting, and construct a sample dataset; S1.2 Normalize the sample dataset. And generate feature vectors from the normalized sample data. In the formula It is the first Group 1 The data after column normalization It is the first Group 1 The data in the column, , They are the first The largest and smallest data in the column, It is the first Feature vectors generated from grouped normalized data. For the first Grouped normalized data; S1.3, Establish the relationship between the sample to be tested and the first... Euclidean distance between groups of samples The Euclidean distance is used to determine the distance between the sample data and the target data, and the nearest sample data to the target data is retrieved as the training data sample. The sample to be tested and the first Euclidean distance between groups of samples It is a feature vector generated from the target data set. Weighting coefficients for each influencing factor; S1.

4. Based on the relationship between explosive consumption and engineering geological conditions and blasting design parameters, preliminary selection of weights is made; The engineering geological conditions mentioned include the integrity coefficient of the rock mass, the elastic modulus of the rock mass, the tensile strength of the rock mass, the cross-sectional area, and the tunneling depth. The blasting design parameters include the borehole diameter, borehole depth, borehole spacing, and row spacing. S1.

5. Retrieve the training sample in the sample dataset that is closest to the prediction target, i.e., the sample with the smallest Euclidean distance.

3. The method for determining the amount of explosive charge in tunneling blasting according to claim 1, characterized in that, Step S2 includes the following steps: S2.1 Determine the number of neurons in the input layer, the number of neurons in the hidden layer, and the number of hidden layers, and simultaneously assign connection weights between each neuron. Hidden layer threshold Input layer threshold Perform initialization and set the learning rate. and excitation function ; S2.2 Calculate the hidden layer output: based on the input value Connection weights between the input layer and the hidden layer and hidden layer threshold Through formula Calculate the output value of the hidden layer. , S2.3 Calculate the output of the output layer: based on the output values ​​of the hidden layers. Connection weights between hidden layers and output layers and output layer threshold The predicted output value of the neural network is calculated as follows: ; S2.4 Calculation Error: Based on the predicted output value and the expected output value Y of the neural network, calculate the prediction error: ; S2.5 Update weights: Update and correct each weight based on the error calculated in step S2.

4. ; S2.6, Update Thresholds: Update and correct each threshold according to the error. ; S2.7 Determine if training is complete. Recalculate the output error based on the updated weights and check if the output result meets expectations. If the accuracy is not met, repeat steps S2.2 to S2.6 until the output meets the accuracy requirements for the unit consumption of tunneling blasting explosives. .

4. The method for determining the amount of explosive charge in tunneling blasting according to claim 1, characterized in that, Step S3 includes the following sub-steps: S3.1 Obtain the volume of the blasted rock mass during tunneling. ; S3.2 Calculate the total charge amount for tunneling blasting based on the volume of the blasted rock mass. .

5. The method for determining the amount of explosive charge in tunneling blasting according to claim 1, characterized in that, Step S4 includes the following sub-steps: S4.1 Obtain the linear charge density of the surrounding holes according to the blasting design. ; S4.2 Obtain the length of the surrounding holes. ; S4.3 Calculate the charge amount for a single peripheral hole. In the formula for The length of each peripheral hole; S4.4 Calculate the total charge amount for the peripheral holes. .

6. The method for determining the amount of explosive charge in tunneling blasting according to claim 1, characterized in that, Step S5 includes the following sub-steps: S5.1 Calculate the rock mass volume within the excavation area. In the formula The area of ​​the cut section is... The length of the cut hole; S5.2 Determine the charge density of the cut hole according to the blasting design. ; S5.3 Calculate the charge amount for a single cut hole ; S5.4 Calculate the total charge amount in the cut hole. .

7. The method for determining the amount of explosive charge in tunneling blasting as described in claim 1, characterized in that, Step S6 includes the following sub-steps: S6.1 Calculate the total charge amount of the auxiliary holes based on the total charge amount and the total charge amount of the peripheral holes and the slotted holes. ; S6.2 Determine the charge amount for a single auxiliary hole based on the number of auxiliary holes, n. .

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images