Initiation network connection method capable of autonomously identifying blast holes

Abstract

The invention discloses a detonating network line connection method capable of autonomously identifying blast holes. The detonating network line connection method comprises the following steps of data preprocessing, feature selection, relative regularity grouping, intra-group slope judgment, special point processing and autonomous line connection. The method is low in cost and easy to operate, the safety, efficiency and economical efficiency of blasting operation are improved in a mode of autonomously recognizing the blast holes, meanwhile, the traditional blast hole recognition method that technical personnel conduct visual inspection through naked eyes is achieved, and therefore the labor and automation cost is reduced. According to the method, the automation level of blasting operation is improved, the operation risk is reduced, the operation efficiency is improved, and the method has important practical application value.

Description

Technical Field

[0001] The present invention relates to a method for connecting an initiation network, and particularly to a method for connecting an initiation network that autonomously identifies blast holes, belonging to the technical field of mine blasting. Background Art

[0002] With the development of modern mining, blasting technology plays a crucial role in mine exploitation. Traditional blasting operations rely on manual positioning, drilling, charging, and initiation of blast holes. These processes not only have a high labor intensity and low efficiency but also pose significant safety risks. With the progress of automation and intelligent technologies, in the field of intelligent blasting, accurate identification of blast holes is a key step in realizing automated initiation network connection, and the accuracy of blast hole identification directly affects the blasting effect and operation safety.

[0003] Traditional methods for identifying blast holes mainly rely on manual visual inspection or simple mechanical devices. These methods have poor adaptability in complex environments and cannot achieve high-precision automated operations. In recent years, although some automated technologies have been introduced into the field of blast hole identification, such as technologies based on image recognition, laser scanning, and sensor networks, these technologies still face four major challenges in practical applications: the complex and changeable mine environment; lack of real-time performance and accuracy; inability to fully achieve autonomous operation; and high automation costs. Summary of the Invention

[0004] In order to solve the deficiencies of the above technologies, the present invention provides a method for connecting an initiation network that autonomously identifies blast holes, aiming to achieve rapid, accurate, and autonomous identification of blast holes through intelligent algorithms, and construct an efficient and reliable initiation network to improve the safety, efficiency, and economy of mine blasting operations.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for connecting an initiation network that autonomously identifies blast holes, comprising the following steps:

[0006] Step S1, data preprocessing: grouping and training the data and conducting algorithm tests;

[0007] Step S2, feature selection: making a primary slope judgment for relatively regular feature points, and finally determining the upper and lower limits of the slope applicable to the data of all blast hole positions;

[0008] Step S3, relatively regular grouping: after completing the primary slope judgment, perform data fitting, divide the blast hole data with greater connectivity into one group, set a confidence threshold, and end the loop when the threshold is exceeded, and group all blast hole positions;

[0009] Step S4, in-group slope determination: making the slopes of all points in the group lie on the linear fitting line through linear fitting;

[0010] Step S5, special point processing: Individually process the blast hole points where the group cannot be determined or disconnection occurs.

[0011] Step S6, autonomous wiring: According to the determined groups, connect the blast holes in sequence within each group to obtain the initiation network.

[0012] Preferably, in step S1, the blast hole point data of the mine is collected and sorted to form blast hole information in the geodetic coordinate system including the X coordinate and the Y coordinate, which is grouped and trained and tested by variance using the algorithm. Among them, the algorithm used is the K-NN algorithm.

[0013] Preferably, in step S1, the grouped training means that based on the K-NN algorithm, different weights are assigned to different neighboring points for classification training, and the following formula is used to calculate the distance reciprocal weight:

[0014]

[0015] Where: d(x, x i ) is the distance between the query point x and the neighboring point x i ; ω i is the weight of the i-th neighbor.

[0016] The algorithm is tested using the variance formula, and the formula is as follows:

[0017]

[0018] Where: σ 2 represents the variance; x i represents the coordinate of each blast hole information; μ represents the mean of the blast hole information; n represents the total number of blast holes.

[0019] Preferably, in step S2, the Euclidean distance formula is used to calculate the slope and distance between two points or multiple points, determine the upper and lower limits of the slope between the feature points, remove the connected lines that exceed the upper and lower limits of the slope, and only retain the same horizontal slope value, and so on, and finally determine the upper and lower limits of the slope applicable to the entire blast hole point data.

[0020] Preferably, in step S2, the feature points are selected using the neighborhood density. The neighborhood density formula estimates the density by calculating the reciprocal of the average of the nearest neighbor distances for each blast hole point information, and the formula is as follows:

[0021]

[0022] Where: D(x) is the estimated density; d(x, x i ) is the distance between the query point x and the neighboring point x i ; K is the number of the nearest neighbor points of the query point x.

[0023] The Euclidean distance formula is as follows:

[0024]

[0025] Where: L(x i , x j ) is the distance between x i and x j ; x i is the coordinate of the i-th point; x j is the coordinate of the j-th point, is the l-th eigenvalue of the coordinates of the i-th and j-th points; l is the index for summing over all features.

[0026] Preferably, in step S3, in the relative regularity grouping, "relative" means comparing with the slopes of adjacent rows and columns. When the connectivity is large, they can be divided into one group;

[0027] For the blast hole data with poor connectivity (i.e., the slope of the same row or column is lower than the slope between adjacent points, allowing the slope error to be in the range of -0.2 < k < 0.2, where k refers to the slope of adjacent points), it is necessary to recalculate the slopes of the coordinates of two or more blast holes, and the confidence threshold range is 0.65 - 0.8;

[0028] Connectivity refers to the slope between adjacent points. When the slope error of the same row or column is in the range of -0.5 < k < 0.5 (where k refers to the slope of the same row or column), they can be divided into the same group. The slope formula is:

[0029]

[0030] Where: k is the slope between the i and j coordinates; x i , y i are the coordinates of the i-th point; x j , y j are the coordinates of the j-th point.

[0031] Preferably, step S4 includes the following steps:

[0032] Step S41: After the relative regularity grouping, perform within-group slope determination for each group, sort according to the X coordinates of the blast hole positions within the group, and perform linear fitting;

[0033] Step S42: Retain the points that conform to the linear fitting within the group again. Correct the coordinates of non-horizontal blast holes, recalculate the slopes, and perform linear fitting and grouping retention again;

[0034] Correcting the coordinates of non-horizontal blast holes means rearranging the points that do not conform to the linear fitting;

[0035] Step S43: Repeat the above steps until the slopes of all points within the group are on the linear fitting line.

[0036] Preferably, step S5 includes the following steps:

[0037] Step S51: When it is impossible to determine which group an individual blast hole position belongs to based on the above steps, it is necessary to sort and group them according to the Y value and retain them;

[0038] Step S52: When there is a disconnection at an individual blast hole position, it is necessary to re-determine whether the slope formed by the two points is within the threshold range; where the threshold range refers to the upper and lower limits of the slope calculated in step S2, that is, -12 < k < 12.

[0039] Preferably, in step S6, according to the determined groups, and within the groups according to the blast hole X 1.1 、X 1.2 、X 1.3 ……; X 2.1 、X 2.2 、X 2.3 ……; X i.1 、X i.2 、X i.3 ……; Y 1.1 、Y 1.2 、Y 1.3 …… are connected in sequence to obtain the initiation network.

[0040] The initiation network refers to a network in a blasting project that connects multiple initiation elements (such as electronic detonators) through a certain connection method to achieve initiation in a predetermined order and time interval.

[0041] The present invention uses the K-NN algorithm as the basis for algorithm training and testing, calculates the slope and distance through the Euclidean distance formula, determines the upper and lower limits of the slope between feature points, removes the connected lines that exceed the upper and lower limits of the slope, and only retains the same horizontal slope value. The blast hole data with greater connectivity is divided into a group through data fitting, the Y values of special points are sorted and grouped for retention, and the slope threshold of the disconnected points is determined. Finally, the determined groups are connected in sequence to realize the autonomous recognition of the initiation network of the blast holes.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The present invention has a low cost and simple operation. By autonomously identifying the blast holes, it improves the safety, efficiency, and economy of the blasting operation, and at the same time solves the traditional method of blast hole identification by technicians through visual inspection, thereby reducing the labor and automation costs. This method not only improves the automation level of the blasting operation, but also reduces the operation risk and improves the operation efficiency, and has important practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is the overall operation flowchart of the present invention.

[0045] Figure 2 This is the schematic effect diagram of the operation of the present invention.

[0046] Figure 3 This is the schematic diagram of the initiation network for Example 1 of the present invention.

[0047] Figure 4 This is the schematic diagram of the initiation network for Example 2 of the present invention. Detailed implementation manners

[0048] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0049] As Figure 1 shown, a method for connecting the initiation network for automatically identifying blast holes includes the following steps:

[0050] S1. Data preprocessing: Collect and organize the mine blast hole position data to form blast hole information in the geodetic coordinate system including X coordinates and Y coordinates, group and train it, and test the algorithm through variance.

[0051] Among them, the algorithm used for algorithm testing is the K-NN algorithm.

[0052] Group training refers to, based on the K-NN algorithm, classifying and training by assigning different weights to different neighboring points, and calculating the distance reciprocal weight using the following formula:

[0053]

[0054] Where: d(x, x i ) is the distance between the query point x and the neighboring point x i ; ω i is the weight of the i-th neighbor.

[0055] Use the variance formula for algorithm testing as follows:

[0056]

[0057] Where: σ 2 represents the variance; x i represents the coordinate of each blast hole information; μ represents the mean of the blast hole information; n represents the total number of blast holes.

[0058] S2. Feature selection: Make a primary slope judgment on relatively regular feature points, calculate the slope and distance between two or more points using the Euclidean distance formula, and determine the upper and lower limits of the slope between the feature points. As Figure 2As shown in a, remove the connecting lines that exceed the upper and lower limits of the slope, and only retain the same horizontal slope value, and so on. Finally, determine the upper and lower limits of the slope applicable to the data of all blast hole positions;

[0059] The connecting line refers to the line segment connecting two or more points among the blast hole positions.

[0060] Relatively regular feature points refer to the points that are relatively evenly distributed or clustered in the drawn blast hole coordinate information. Feature points are selected using the nearest neighbor density. The formula for the nearest neighbor density estimates the density by calculating the reciprocal of the average of the nearest neighbor distances for each blast hole position information. The formula is as follows:

[0061]

[0062] Where: D(x) is the estimated density; d(x, x i ) is the distance between the query point x and the neighboring point x i ; K is the number of the nearest neighbor points of the query point x.

[0063] The Euclidean distance formula is as follows:

[0064]

[0065] Where: L(x i , x j ) is the distance between x i and x j ; x i is the coordinate of the i-th point; x j is the coordinate of the j-th point, is the l-th eigenvalue of the coordinates of the i-th and j-th points; l is the index for summing all features.

[0066] S3. As Figure 2 shown in b, relatively regular grouping: After completing the initial slope judgment, perform data fitting. Divide the blast hole data with greater connectivity into one group. If the blast hole data with poor connectivity (i.e., the slope of the same row or column is lower than the slope between neighboring points, allowing a slope error in the range of -0.2 < k < 0.2, where k refers to the slope of neighboring points), it is necessary to recalculate the slope of two or more coordinates of the blast hole. Set the confidence threshold to 0.65. When this threshold is exceeded, end the loop and group all blast hole positions;

[0067] Connectivity refers to the slope between neighboring points. Greater connectivity means that when the slope error of the same row or column is in the range of -0.5 < k < 0.5 (where k refers to the slope of the same row or column), they can be divided into the same group.

[0068]

[0069] Where: k is the slope between the i and j coordinates; x i and y i are the coordinates of the i-th point; x j and y j are the coordinates of the j-th point.

[0070] S4. As shown in Figure 2 c, within-group slope determination:

[0071] S41. After relatively regular grouping, within-group slope determination is performed for each group. The in-group blast hole positions are sorted according to the X coordinate and linearly fitted;

[0072] S42. The points that conform to the linear fit within the group are grouped and retained again. For the coordinates of non-horizontal blast holes, corrections are made, and the slope is recalculated, and then linearly fitted and grouped and retained again;

[0073] The correction of the coordinates of non-horizontal blast holes means that the points that do not conform to the linear fit are rearranged, and the connectivity described in S3 is recalculated and the range is restricted.

[0074] S43. Repeat the above steps until the slopes of all points within the group are on the linear fit line;

[0075] In steps S3 and S41, the relatively regular grouping is to divide multiple positions with relatively large blast hole data connectivity into one group.

[0076] S5. As shown in Figure 2 d, special point processing:

[0077] S51. When it is impossible to determine which group an individual blast hole position belongs to based on the above steps, it is necessary to sort and group and retain according to the Y value;

[0078] S52. When there is a disconnection at an individual blast hole position, it is necessary to re-determine whether the slope formed by the two points is within the threshold range.

[0079] The threshold range refers to the upper and lower limits of the slope calculated in step S2, that is, -12 < k < 12;

[0080] S6. As shown in Figure 2 e, autonomous connection: According to the determined groups, and within the groups according to the blast hole X 1.1 、X 1.2 、X 1.3 ……; X 2.1 、X 2.2 、X 2.3 ……; X i.1 、X i.2 、X i.3 ……; Y 1.1 、Y 1.2 、Y1.3 ……Connect them in sequence to achieve the autonomous recognition of the initiation network connection for blast holes.

[0081] The present invention will be further described in detail below in conjunction with embodiments.

[0082] At present, the mine environment is complex and changeable. Traditional blasting operations rely on manual positioning, drilling, charging, and initiation of blast holes to address challenges such as high labor intensity, low efficiency, and significant safety risks. Construction personnel at a certain open-pit mine only need to obtain the blast hole position information and perform data processing, and use the K-NN algorithm to complete the autonomous recognition of the initiation network connection for blast holes.

[0083] Embodiment 1

[0084] An intelligent connection initiation network method for autonomous recognition of blast holes specifically includes the following steps:

[0085] The first step: Data preprocessing:

[0086] Construction personnel at a certain open-pit mine collect the blast hole coordinate information using RTK and organize the collected coordinate data into the format of the geodetic coordinate system.

[0087] The second step: Feature selection:

[0088] Perform a primary slope judgment on regular feature points, use the Euclidean distance formula to calculate the slope and distance between two or more points, and determine the upper and lower limits of the slope between feature points: -12 < k < 2.5;

[0089] The third step: Relative regularity grouping:

[0090] Fit the blast hole data with greater connectivity, set the confidence threshold to 0.7, end the loop when the threshold is exceeded, and group all blast hole positions.

[0091] The fourth step: Intra-group slope determination:

[0092] Perform intra-group slope determination on each group, sort according to the X coordinate and perform linear fitting, retain the grouped points that meet the linear fitting, and recalculate the slope and perform linear fitting on the points that do not meet the requirements.

[0093] The fifth step: Special point processing:

[0094] For individual blast hole positions that cannot be determined to belong to a group, sort and group them according to the Y value. For disconnected blast hole positions, re-determine whether the slope of the line formed by two points is within the threshold range: -5 < k < -1.

[0095] The sixth step: Autonomous connection:

[0096] According to the determined groups, connect them in sequence according to the X and Y coordinates of the blast holes to achieve the intelligent connection initiation network for autonomous identification of blast holes as shown in Figure 3 The intelligent connection initiation network for autonomous identification of blast holes shown.

[0097] Embodiment 2

[0098] An intelligent connection initiation network method for autonomous identification of blast holes specifically includes the following steps:

[0099] The first step: Data preprocessing:

[0100] Construction workers at an open-pit mine collect the coordinate information of blast holes using RTK and organize the collected coordinate data into the format of the geodetic coordinate system.

[0101] The second step: Feature selection:

[0102] Make a preliminary slope judgment on regular feature points, use the Euclidean distance formula to calculate the slope and distance between two or more points, and determine the upper and lower limits of the slope between feature points: 2.5 < k < 5;

[0103] The third step: Relative regularity grouping:

[0104] Fit the blast hole data with greater connectivity, set the confidence threshold to 0.8, end the loop when the threshold is exceeded, and group all blast hole positions.

[0105] The fourth step: Intra-group slope determination:

[0106] Make an intra-group slope determination for each group, sort according to the X coordinate and perform linear fitting, retain the grouped points that conform to the linear fitting, and recalculate the slope and perform linear fitting on the points that do not conform.

[0107] The fifth step: Special point processing:

[0108] For individual blast hole positions that cannot be determined in groups, sort and group them according to the Y value. For disconnected blast hole positions, re-determine whether the slope of the line formed by two points is within the threshold range: 1 < k < 1.5.

[0109] The sixth step: Autonomous connection:

[0110] According to the determined groups, connect them in sequence according to the X and Y coordinates of the blast holes to achieve the intelligent connection initiation network for autonomous identification of blast holes as shown in Figure 4 The intelligent connection initiation network for autonomous identification of blast holes shown.

[0111] The above embodiments are not limitations on the present invention, and the present invention is not limited to the above examples either. Changes, modifications, additions, or substitutions made by those skilled in the art within the scope of the technical solution of the present invention also fall within the protection scope of the present invention.

Claims

1. A method for connecting a blasting network for autonomously identifying a blast hole, characterized in that: It includes the following steps: Step S1, data preprocessing: Group and train the data and conduct algorithm tests; Step S2, feature selection: Make a preliminary slope judgment for relatively regular feature points, and finally determine the upper and lower limits of the slope applicable to the data of the entire blast hole positions; Step S3, relative regularity grouping: After completing the preliminary slope judgment, perform data fitting, divide the blast hole data with greater connectivity into one group, set a confidence threshold, end the loop when the threshold is exceeded, and group all blast hole positions; Step S4, within-group slope determination: Make the slopes of all points within the group lie on the linear fitting line through linear fitting; Step S5, special point processing; Process the blast hole positions that cannot be grouped or have disconnections separately; Step S6, autonomous wiring: According to the determined groups, connect them in sequence according to the blast holes within the group to obtain the initiation network.

2. The method for connecting a blasting network for autonomously identifying a blast hole according to claim 1, characterized in that: In the said step S1, collect and organize the data of the blast hole positions in the mine to form blast hole information in the geodetic coordinate system including the X coordinate and the Y coordinate, group and train it, and conduct algorithm tests through variance. Among them, the algorithm adopted is the K-NN algorithm.

3. The method for connecting the detonation network for autonomously identifying blastholes according to claim 2 is characterized in that: In the said step S1, group training means that on the basis of the K-NN algorithm, different weights are assigned to different neighboring points for classification training, and the following formula is used to calculate the distance reciprocal weight: Where: d(x,x i ) are the query point x and the neighboring points x i The distance between i is the weight of the i-th neighbor; Use the variance formula to conduct algorithm tests, and the formula is as follows: Where: 2 represents variance; x i represents the coordinates of each blasthole information; μ represents the mean of the blasthole information; n represents the total number of blastholes.

4. The method for connecting a blasting network for autonomously identifying a blast hole according to claim 1, characterized in that: In the said step S2, use the Euclidean distance formula to calculate the slope and distance between two or more points, determine the upper and lower limits of the slope between the feature points, remove the connected lines that exceed the upper and lower limits of the slope, and only retain the same horizontal slope value, and so on, and finally determine the upper and lower limits of the slope applicable to the data of the entire blast hole positions.

5. The method for connecting the detonation network for autonomously identifying blastholes according to claim 4 is characterized in that: In the said step S2, use the neighborhood density to select feature points. The neighborhood density formula estimates the density by calculating the reciprocal of the average of the nearest neighbor distances based on the information of each blast hole position. The formula is as follows: Where: D(x) is the estimated density; d(x,x i ) are the query point x and the neighboring points x i The distance between them; K is the number of nearest neighbors of the query point x; The Euclidean distance formula is as follows: Where: L(x i ,x j ) is x i and x j The distance between i is the coordinate of the ith point; x j is the coordinate of the jth point; is the I-th eigenvalue of the coordinates of the i-th and j-th points; I is the index for summing all features.

6. The method for connecting the detonation network for autonomously identifying blastholes according to claim 1 is characterized in that: In the said step S3, in the relative regularity grouping, relative means comparing with the slopes of adjacent rows and columns. When the connectivity is greater, it can be divided into one group; For the blast hole data with poor connectivity, it is necessary to recalculate the slopes of two or more coordinates of the blast hole. The confidence threshold range is 0.65 - 0.8; Among them, the blast hole data with poor connectivity refers to: the slopes of the same row or the same column are lower than the slopes between adjacent points, and the allowable slope error is in the range of -0.2 < k < 0.2, where k refers to the slope of adjacent points; Connectivity refers to the slope between adjacent points. When the slope error of the same row or the same column is in the range of -0.5 < k < 0.5, it can be divided into the same group. k refers to the slope of the same row or the same column, and the slope formula is: Where: k is the slope between the i and j coordinates; x i ,y i is the coordinate of the ith point; x j ,y j are the coordinates of the jth point.

7. The method for connecting a blasting network for autonomously identifying a blast hole according to claim 1, characterized in that: The said step S4 includes the following steps: Step S41, after the relative regularity grouping, conduct within-group slope determination for each group, sort according to the X coordinates of the blast hole positions within the group and perform linear fitting; Step S42, the points that conform to the linear fitting within the group are grouped and retained again. The non-horizontal blast hole coordinates are corrected, and the slope is recalculated, and then grouped and retained through linear fitting again; The correction of the non-horizontal blast hole coordinates means rearranging the points that do not conform to the linear fitting; Step S43, repeat the above steps until the slopes of all points in the group are on the linear fitting line.

8. The method for connecting a blasting network for autonomously identifying a blast hole according to claim 1, characterized in that: The step S5 comprises the following steps: Step S51: When individual blasthole points cannot be determined based on the above steps to which group the points belong, they need to be sorted according to the Y value and retained in groups; Step S52: When a disconnection occurs at a certain blasthole point, it is necessary to re-determine whether the slope of the straight line formed by the two points is within the threshold range; wherein the threshold range refers to the upper and lower limits of the slope calculated in step S2, i.e. -12 <k<12。 9. The method for connecting a blasting network for autonomously identifying a blast hole according to claim 1, characterized in that: In step S6, according to the determined groups, the groups are divided into groups according to the blasthole X. 1.1 , X 1.2 , X 1.3 ……;X 2.1 , X 2.2 , X 2.3 ……;X i.1 , X i.2 , X i.3 ……; Y 1.1 , Y 1.2 , Y 1.3 ...connect them one by one to get the detonation network.

Images