Reliability detection method, device, medium and product for engineering structure at the edge of unloading area

Abstract

The disclosed embodiments disclose a method, device, medium, and product for reliability detection of edge engineering structures in unloading areas. The method includes: constructing a classification model; obtaining first point cloud data of a target unloading area, and obtaining a first cross-sectional view based on the first point cloud data; classifying the edge engineering structure based on the first cross-sectional view and the classification model to obtain a first type; when the first type is a preset type, determining that the edge engineering structure is reliable; when the first type is not a preset type, obtaining the confidence of the edge engineering structure; and determining whether the edge engineering structure is reliable based on the confidence. This method can complete the reliability detection of the edge engineering structure of the unloading area through the classification model and the confidence, and effectively ensure the stability and safety of the unloading area operation.

Description

Technical Field

[0001] The present disclosure relates to the field of mining engineering technology, and in particular to a method, device, medium and product for detecting the reliability of an engineering structure at the edge of an unloading zone. Background Art

[0002] Unmanned driving in mines is a key step in the intelligent construction of energy mines. It involves technologies such as artificial intelligence, cloud computing, Internet of Vehicles, intelligent driving and high-precision maps. These technologies are deeply integrated into modern coal and other energy mining, enabling mine cars to carry out continuous operation of the entire process of mining, transportation and drainage in open-pit mines.

[0003] However, when the mine car is unloading in the unloading area, the edge engineering structure may change due to irregular unloading operations, affecting the safety and efficiency of normal unloading. Therefore, the reliability detection of the edge engineering structure is very important. Among them, the unloading area is the area where the mine car is unloading, and the edge engineering structure is the soil structure composed of soil, stones and other materials dropped from the ore body during the unloading operation. Summary of the invention

[0004] In view of this, the embodiments of the present disclosure provide a method, device, medium, and product for reliability detection of engineering structures at the edge of an unloading area, which can complete reliability detection of engineering structures at the edge of an unloading area through classification models and confidence levels, and effectively ensure the stability and safety of operations in the unloading area.

[0005] In a first aspect, the embodiment of the present disclosure provides a method for detecting the reliability of an engineering structure at the edge of an unloading zone, which adopts the following technical solution:

[0006] Build a classification model;

[0007] Acquire first point cloud data of a target unloading area, and obtain a first cross-sectional view based on the first point cloud data;

[0008] Classifying the edge engineering structure based on the first cross-sectional view and the classification model to obtain a first type;

[0009] When the first type is a preset type, determining that the edge engineering structure is reliable;

[0010] When the first type is not a preset type, obtaining the confidence level of the edge engineering structure;

[0011] determining whether the edge engineering structure is reliable based on the confidence level;

[0012] The first profile view is a cross-sectional view of the edge engineering structure of the target unloading area, and the edge engineering structure is a soil structure composed of soil and rocks dropped from the ore body during the unloading operation.

[0013] Optionally, the constructing of the classification model includes:

[0014] Construct a virtual unloading area based on the real unloading area;

[0015] obtaining second point cloud data based on the real unloading area and the virtual unloading area;

[0016] Obtaining a plurality of second cross-sectional views based on the second point cloud data;

[0017] Classifying and labeling the plurality of second cross-sectional views to obtain a plurality of third cross-sectional views;

[0018] The VIT neural network model is trained based on the multiple third profile images to obtain the classification model.

[0019] Optionally, the classifying and labeling the plurality of second cross-sectional views to obtain a plurality of third cross-sectional views includes:

[0020] determining a first corner point and a second corner point of the second cross-sectional view;

[0021] Obtaining a height value and a slope angle of the second profile based on the first corner point and the second corner point;

[0022] classifying the second profile image based on the height value and the slope angle of the second profile image;

[0023] The second cross-sectional view is labeled according to the classification result to obtain the third cross-sectional view.

[0024] Optionally, a calculation formula for the height value of the second cross-sectional image is as follows:

[0025] D z =ΔZ=|Z2-Z1|;

[0026] Among them, D z is the height value of the second cross-sectional view; ΔZ is the height displacement between the second corner point and the first corner point; Z2 is the height value of the second corner point; Z1 is the height value of the first corner point.

[0027] Optionally, the calculation formula of the slope angle of the second cross-sectional view is as follows:

[0028]

[0029] Among them, γ is the slope angle of the second profile; ΔX is the lateral displacement between the second corner point and the first corner point, ΔX=|X2-X1|, X2 is the lateral coordinate value of the second corner point, and X1 is the lateral coordinate value of the first corner point; ΔY is the longitudinal displacement between the second corner point and the first corner point, ΔY=|Y2-Y1|, Y2 is the longitudinal coordinate value of the second corner point, and Y1 is the longitudinal coordinate value of the first corner point.

[0030] Optionally, the confidence calculation formula is as follows:

[0031]

[0032] Wherein, ZX is the confidence level; Δe1 is the excess volume of the first cross-sectional image; Δe2 is the missing volume of the first cross-sectional image; L X is the length of the edge engineering structure; L Y is the width of the edge engineering structure; L Z is the predicted height of the edge engineering structure; L N is the standard height of the edge engineering structure; ρ is the slope of the edge engineering structure, and β is the soil condition.

[0033] Optionally, determining whether the edge engineering structure is reliable based on the confidence level includes:

[0034] Determining whether the confidence level is greater than a preset confidence threshold;

[0035] If so, it is determined that the edge engineering structure is reliable;

[0036] If not, it is determined that the edge engineering structure is unreliable.

[0037] In a second aspect, the embodiment of the present disclosure further provides a reliability detection system for the engineering structure at the edge of the unloading area, which adopts the following technical solution:

[0038] Model building module, used to build classification models;

[0039] A data acquisition module, used to acquire first point cloud data of a target unloading area, and obtain a first cross-sectional view based on the first point cloud data;

[0040] A structure classification module, used for classifying the edge engineering structure based on the first cross-sectional view and the classification model to obtain a first type;

[0041] A first determination module, configured to determine that the edge engineering structure is reliable when the first type is a preset type;

[0042] A confidence acquisition module, used for acquiring the confidence of the edge engineering structure when the first type is not a preset type;

[0043] A second determination module, configured to determine whether the edge engineering structure is reliable based on the confidence level;

[0044] The first profile view is a cross-sectional view of the edge engineering structure of the target unloading area, and the edge engineering structure is a soil structure composed of soil and rocks dropped from the ore body during the unloading operation.

[0045] In a third aspect, the embodiments of the present disclosure further provide a computer device, which adopts the following technical solution:

[0046] The computer device comprises:

[0047] at least one processor; and,

[0048] a memory communicatively connected to the at least one processor; wherein,

[0049] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any of the above-mentioned methods for detecting the reliability of the unloading area edge engineering structure.

[0050] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute any of the above-mentioned methods for detecting the reliability of the unloading zone edge engineering structure.

[0051] In a fifth aspect, an embodiment of the present disclosure further provides a computer program product, including a computer program / instruction, which implements the steps of any of the above methods when executed by a processor.

[0052] The reliability detection method of the edge engineering structure of the unloading area provided by the embodiment of the present disclosure obtains a first cross-sectional view through first point cloud data. The first cross-sectional view can reflect the characteristics of the edge engineering structure. Therefore, by analyzing the first cross-sectional view through a classification model, the edge engineering structure of the target unloading area can be automatically classified to obtain a first type; if the first type is a preset type, it can be directly judged that the edge engineering structure is reliable. If the first type is not the preset type, it is necessary to further obtain the confidence of the edge engineering structure, and use the confidence to complete further judgment of the reliability, so as to realize real-time monitoring and feedback of the edge engineering structure, which helps to timely discover and deal with potential problems and improve the safety of unloading operations.

[0053] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following preferred embodiments are specifically cited and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, 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 disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 A schematic flow chart of a reliability detection method for an unloading zone edge engineering structure provided in an embodiment of the present disclosure;

[0056] Figure 2 A schematic diagram of a process for constructing a classification model provided in an embodiment of the present disclosure;

[0057] Figure 3 A schematic diagram of a first cross-sectional view and a fourth cross-sectional view provided for an embodiment of the present disclosure;

[0058] Figure 4 A principle block diagram of a reliability detection system for an engineering structure at the edge of an unloading area provided in an embodiment of the present disclosure;

[0059] Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0061] It should be clear that the following embodiments of the present disclosure are described by specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other in the absence of conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0062] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0063] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0064] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0065] Reference Figure 1 The present disclosure provides a method for detecting the reliability of an engineering structure at the edge of an unloading zone, comprising the following steps:

[0066] S1: Build a classification model.

[0067] S2: Acquire first point cloud data of the target unloading area, and obtain a first cross-sectional view based on the first point cloud data.

[0068] Among them, the target unloading point refers to the area where the mine car performs unloading operations; the first profile view is a cross-sectional view of the edge engineering structure of the target unloading area. In the present disclosure, a cross-section is set along the XZ direction; the edge engineering structure is also called a retaining wall, which is an earth structure composed of soil, stones and other materials dropped from the ore body during the unloading operation. The staff will repair the edge engineering structure from time to time during the operation.

[0069] S3: Classify the edge engineering structure based on the first cross-sectional view and the classification model to obtain a first type.

[0070] S4: Determine whether the first type is a preset type; if so, execute S5; if not, execute S6.

[0071] S5: Determine that the edge engineering structure is reliable.

[0072] S6: Obtain confidence of edge engineering structure.

[0073] S7: Determine whether the confidence is greater than a preset confidence threshold; if so, execute S5; if not, execute S8.

[0074] S8: Determine that the edge engineering structure is unreliable, repair the edge engineering structure or replace the target unloading area.

[0075] After repairing the edge engineering structure, the first point cloud data of the target unloading area is reacquired, and the reliability of the edge engineering structure is judged again based on the first point cloud data. Only when it is determined that the edge engineering structure of the target unloading area is reliable can the safe unloading operation continue in the target unloading area.

[0076] The reliability detection method of the edge engineering structure of the unloading area disclosed in the present invention obtains the first cross-sectional view through the first point cloud data. The first cross-sectional view can reflect the characteristics of the edge engineering structure. Therefore, the first cross-sectional view is analyzed by the classification model, and the edge engineering structure of the target unloading area can be automatically classified. The classification result determines whether it is reliable or needs further judgment, and the confidence of the edge engineering structure is obtained. The confidence is used to complete further judgment, and the real-time monitoring and feedback of the edge engineering structure is realized, which helps to timely discover and deal with potential problems, improve the safety and reliability of the edge engineering structure, and effectively avoid unnecessary maintenance and repair costs. In addition, timely replacement of the target unloading area can also reduce potential safety risks and losses.

[0077] In summary, the present disclosure applies machine learning to improve the accuracy of classification and evaluation, and subsequently combines confidence to complete further discrimination, which makes the entire process more efficient and reduces the need for manual intervention. The method is applicable to various types of unloading zone edge engineering structures, including soil structures composed of soil, stones, etc. Therefore, it can be applied to different types of mines and engineering scenarios, and has strong versatility and applicability.

[0078] Reference Figure 2 The schematic diagram of the process of building a classification model is shown. In S1, "building a classification model" includes the following steps:

[0079] S21: constructing a virtual unloading area based on the real unloading area;

[0080] S22: obtaining second point cloud data based on the real unloading area and the virtual unloading area;

[0081] S23: obtaining a plurality of second cross-sectional views based on the second point cloud data;

[0082] S24: classifying and labeling the plurality of second cross-sectional views to obtain a plurality of third cross-sectional views;

[0083] S25: Training the VIT neural network model based on the multiple third profile images to obtain a classification model.

[0084] In S21, a parallel perception system is constructed in accordance with the parallel system theory. In this system, multiple real unloading areas are selected as the basis for training the VIT neural network model. The virtual scene of the open-pit mine unloading area is reproduced with reference to the real unloading area, and the unstructured characteristics of the unloading area are restored. The virtual mine car equipped with a virtual radar collects data of various artificial open-pit mine unloading area scenes, and parallel calculations and experiments are performed to construct a virtual unloading area. The point cloud of the virtual unloading area is obtained by using the virtual radar, which is recorded as the second point cloud data, wherein the virtual radar is constructed using the pose and fitting data.

[0085] In S22, the millimeter wave radar is used to scan the actual unloading area to obtain a point cloud of the actual unloading area, which is also recorded as second point cloud data.

[0086] In S23, grid averaging, statistical filtering, height value replacement, and plane fitting are performed on the second point cloud data to obtain a second cross-sectional view.

[0087] Among them, the “grid average” specifically includes: uniformly gridding the three-dimensional space of the real unloading area and / or the virtual unloading area, calculating the coordinate average of all points contained in each unit grid, which is equivalent to compressing all points in each unit grid into one point to obtain a down-sampled point cloud, and recording the down-sampled point cloud as the third point cloud data. The specific formula is as follows:

[0088] P3={p1,p2,…,p m ,…,p M};

[0089]

[0090]

[0091]

[0092]

[0093] Where P3 is the set of the third point cloud data; m is the mth point in the third point cloud data, m is the number of the points in the third point cloud data, and M is the maximum number of the points in the third point cloud data; x m ,y m , z m are the coordinates of the mth point in the third point cloud data. n i is the number of points contained in the i-th unit grid (here refers to the points in the second point cloud data), i is the number of the unit grid, and the i-th unit grid adopts gi Indicates that g i ∈G, G is the set of unit grids, G={g1, g2, …, g i , …, g I}, I is the maximum number of unit grids; x n ,y n 、z n are the coordinates of the nth point in the second point cloud data, n is the number of the point in the second point cloud data, and the nth point in the second point cloud data adopts p n Indicates that p n =(x n ,y n , z n ), p n ∈P, P2 is the set of the second point cloud data, P2={p1, p2, …, p n ,…,p N}, N is the maximum number of points in the second point cloud data.

[0094] Among them, “statistical filtering” specifically includes: using the K-nearest neighbors algorithm to obtain the K nearest neighbor points of each point in the third point cloud data, specifically referring to the points with p m The K points closest to each other. Get p m The sum of the square differences between the points and its K nearest neighbors is recorded as the first sum of square differences. Based on the first sum of square differences, the first threshold is obtained to determine whether the sum of the first square differences is greater than the first threshold. If so, p is determined. m As an outlier, p m Remove; if not, keep p m After removing all outliers, the fourth point cloud data is obtained. The fourth point cloud data is represented by P4, where P4 = {p1, p2, ..., p j ,…,p J}, p j is the jth point in the fourth point cloud data, p j =(x j ,y j , z j ), x j ,y j 、z j are the coordinates of the jth point in the fourth point cloud data, j is the number of the points in the fourth point cloud data; and J is the maximum number of the points in the fourth point cloud data.

[0095] Among them, the calculation formula for the sum of the first square differences is as follows:

[0096] l m =∑‖p m -p mk ‖2 ;

[0097] In the formula, l m For p m The sum of the first square differences between its K nearest neighbors; p mk For p m The nearest neighbor point of p, and the number of the nearest neighbor point is k; mk ∈N k (p m ), N k (p m ) is p m The set of K nearest neighbor points, N k (p m )={p 1k , p 2k ,…,p mk ,…,p JK}, K is p m The maximum number of the nearest neighbor point, that is, p' m The total number of nearest neighbors.

[0098] The calculation formula of the first threshold is as follows:

[0099] θ=μ l +Oσ l ;

[0100] Where, θ is the first threshold; μ l is the mean of the sum of the first square differences; σ l is the standard deviation of the sum of the first square differences; O is a preset positive integer, usually 3.

[0101] Based on the above, it can be seen that the present disclosure adopts a statistical method to identify abnormal points in the third point cloud data that do not match the distribution of surrounding points, and filters them, thereby achieving the effect of removing noise.

[0102] Among them, the "height value replacement" specifically includes: since any discrete grid can approximate a continuous surface, the surface is sampled, and the height field is used for each two-dimensional coordinate point (x v ,y v ) saves the corresponding vertical height value, geometrically retaining the key structural features represented by the first form of the surface. Using the vertical height value as a three-dimensional scalar of a two-dimensional grid point can restore topological information such as the local tilt of the edge engineering structure surface, and at the same time, it can achieve compression from three-dimensional geometry to two-dimensional representation, and the information quality is still sufficient to restore the surface topological structure.

[0103] The cubic grid discrete model G' is used, and the points of G' are all located at the two-dimensional grid points, G'={g v |(x v ,yv ), v = 1, 2, ... V}, g v is the vth two-dimensional grid point, (x v ,y v ) is the center coordinate of the two-dimensional grid, v is the number of the two-dimensional grid, and V is the maximum number of the two-dimensional grid. The fourth mapping of point cloud data to the two-dimensional height field can be defined as f(P”→G'), and its mapping function is f(p j )=z v , if p j Located at the two-dimensional grid point g v If the vertical height z corresponding to the two-dimensional grid point is used, v Replace z j , at this time, p j =(x j ,y j , z v ).

[0104] Therefore, by calculating the vertical height value around the center of the two-dimensional grid, the two-dimensional height field can be represented as:

[0105] H={z v |(x v ,y v ), v = 1, 2, ... V};

[0106] Where H is the two-dimensional height field representation.

[0107] Among them, "plane fitting" specifically includes: using the transverse gradient operator and the longitudinal gradient operator Determine the slight changes in terrain, and by calculating the gradient of each point in the height field, the key features of the terrain can be effectively extracted. The height field representation function H(x j ,y j ) is defined as a two-dimensional grid G". By calculating the discrete gradient of the height field, the gradient approximation at each point in the fourth point cloud data is taken:

[0108]

[0109]

[0110]

[0111] In the formula, is the gradient approximation of the jth point in the fourth point cloud data, is the lateral gradient operator of the j-th point in the fourth point cloud data; The longitudinal gradient operator of the j-th point in the fourth point cloud data; h is the mesh side length of G”.

[0112] Each point in the fourth point cloud data is regarded as a pixel, that is, C j =p j , C j The pixel represented by the jth point in the fourth point cloud data is referred to as the jth pixel. The gradient modulus of each pixel is obtained based on the gradient approximation. The gradient modulus reflects the steepness of the terrain. The expression of the gradient modulus of the jth pixel is:

[0113] Determine whether the gradient modulus of the j-th pixel is greater than the modulus threshold; if so, mark the j-th pixel as a cliff, i.e., C j =1; if not, the jth pixel is marked as a slope, i.e., C j = 0. The expression of the modulus length threshold is as follows:

[0114] T = ab;

[0115] Where T is the modulus threshold; a is an empirical parameter, usually a∈(1,3); and b is the characteristic standard deviation of the grid block of G”.

[0116] The above pixel-level classification method can be used to mark the terrain and distinguish between cliff and slope types. According to the complexity of the actual terrain, pixels can be further distinguished into types other than cliffs and slopes.

[0117] Connect adjacent pixels of the same type to form different areas. For example, connect several adjacent pixels with cliff marks to form a small cliff area. Calculate the statistical features of the pixels in the area, such as the mean and standard deviation. These features help describe the characteristics of the terrain in the area. For each pixel in the area, a plane model can be fitted using the least squares method or other fitting algorithms to represent the overall distribution of the pixels in the area. This plane model is the edge engineering structure of the real unloading area and / or the virtual unloading area. Select the xz direction as the section direction, cut the edge engineering structure along the section direction, and obtain the second section view.

[0118] In S24, a first corner point and a second corner point of the second cross-sectional view are determined, wherein the first corner point is the highest point in the second cross-sectional view and the second corner point is the lowest point in the second cross-sectional view. A height value and a slope angle of the second cross-sectional view are obtained based on the first corner point and the second corner point.

[0119] The expression of the height value of the second profile is as follows:

[0120] D z =ΔZ=|Z2-Z1|;

[0121] Where D zis the height value of the second cross-sectional view; ΔZ is the height displacement between the second corner point and the first corner point; Z2 is the height value of the second corner point; Z1 is the height value of the first corner point.

[0122] The expression for the slope angle of the second profile is as follows:

[0123]

[0124] Wherein, γ is the slope angle of the second profile; ΔX is the lateral displacement between the second corner point and the first corner point, ΔX=|X2-X1|, X2 is the lateral coordinate value of the second corner point, and X1 is the lateral coordinate value of the first corner point; ΔY is the longitudinal displacement between the second corner point and the first corner point, ΔY=|Y2-Y1|, Y2 is the longitudinal coordinate value of the second corner point, and Y1 is the longitudinal coordinate value of the first corner point.

[0125] The second profile is classified based on the height value and slope angle of the second profile. The disclosure divides the second profile into 9 categories in combination with the coal mine unloading operation, as follows:

[0126] Category 1: [D z ∈(1.2, 1.5), γ∈(20°, 45°)], that is, the height value and slope angle of the second profile are normal, indicating that the edge engineering structure is reasonable and meets the specifications, and the efficient unloading operation can be carried out normally;

[0127] Category II: [D z ∈(0, 1.2), γ∈(0°, 20°)], i.e., both the height value and the slope angle of the second profile are insufficient;

[0128] Category III: [D z ∈(0, 1.2), γ∈(20°, 45°)], that is, the height value of the second profile is insufficient, but the slope angle of the second profile is normal;

[0129] Category 4: [D z ∈(0, 1.2), γ∈(45°, ∞)], that is, the height value of the second profile is insufficient and the slope angle of the second profile is too large.

[0130] Among them, when the second cross-sectional view is of the second, third or fourth category, it means that there may be areas of lower height in the structure due to uneven distribution of deposits or too fast unloading operations, and there is a certain risk of structural instability.

[0131] Category 5: [D z ∈(1.2, 1.5), γ∈(0°, 20°)], that is, the height value of the second profile is normal, but the slope angle of the second profile is insufficient;

[0132] Category 6: [D z∈(1.2, 1.5), γ∈(45°, ∞)], that is, the height value of the second profile is normal, but the slope angle of the second profile is too large.

[0133] Among them, when the second cross-sectional view is of the fifth or sixth category, it means that the edge engineering structure of the real unloading area and / or the virtual unloading area is in the initial stage. The low slope will cause the mine car to slip uncontrollably, and the high slope structure is unstable, affecting the efficiency and safety of the unloading operation.

[0134] Category 7: D z ∈(1.5,∞),γ∈(0°,20°)], that is, the height value of the second profile is too high, and the slope angle of the second profile is insufficient;

[0135] Category 8: [D z ∈(1.5, ∞), γ∈(20°, 45°)], that is, the height value of the second profile is too high, but the slope angle of the second profile is normal;

[0136] Category 9: [D z ∈(1.5,∞),γ∈(45°,∞)], that is, the height value of the second profile is too high and the slope angle of the second profile is too large.

[0137] Among them, when the second cross-sectional view is of the seventh, eighth or ninth category, it means that too much material is piled up in the real unloading area and / or the virtual unloading area, and the structural stability is poor, which affects the efficiency and safety of the unloading operation.

[0138] It is worth noting that if combined with other energy mine unloading operations, the classification of the second profile will be different, but the classification principle is the same, so this disclosure will not be exhaustive here.

[0139] After the classification is completed, the plurality of second cross-sectional views are annotated based on the classification result to obtain a plurality of third cross-sectional views.

[0140] In S25, the classification model is optimized by the VIT neural network model, which includes a CNN convolutional neural network, a Transformer encoder, and a classification processing module (MLP Head).

[0141] The third profile is taken as input and upsampled using the CNN convolutional neural network. Upsampling can enlarge the size of the input data for more detailed feature learning and processing. In the convolution stage, different sizes of convolution kernels are used in the network to learn the long-term and short-term features of the input signal. Convolution kernels of different sizes can capture feature information at different scales, which helps to improve the learning ability of terrain features. In order to enable the VIT neural network model to fully learn the area of ​​interest in the signal, all convolution steps are designed to be 1, which ensures that no important feature information is missed in the convolution operation, which is conducive to improving the perception ability of the model. The output of the last convolution layer matches the input of the standard VIT method so that the processed feature data can be smoothly input into the VIT neural network model for subsequent processing and learning.

[0142] Then, the feature map output by the CNN convolutional neural network is processed in blocks to obtain image blocks. Each image block is converted into a vector of D dimensions through linear transformation, and a vector set is constructed. The vector set is input into the Transformer encoder, which is composed of N encoder blocks. In these encoder blocks, the input vector is first layer-normalized. The data obtained after layer normalization is sent to the multi-head attention mechanism for processing to capture the dependency between different positions, and Dropout processing is performed to obtain the first data, wherein the role of Dropout processing is to prevent the VIT neural network model from overfitting. Then, the vector before layer normalization and the first data are combined, and the combined data is recorded as the second data. The second data is layer-normalized and sent to the multi-layer perceptron, and Dropout processing is performed again to obtain the third data. Then, the second data and the third data are combined to obtain the fourth data, thus completing the output process of an encoder block. The fourth data is input into other encoder blocks, and the same operation is performed until the output of the last encoder block is obtained, and a learnable vector for classification is obtained. This learnable vector is extracted and input into the classification processing module, and finally the classification result is output after layer normalization, full connection, GELU activation function, and full connection in the classification processing module, and the classification result is recorded as the second type. The improved VIT neural network model disclosed in the present invention can further enhance the ability of terrain recognition and classification.

[0143] Use multiple third profiles as sample data, construct the sample data into training set, validation set and test set in a ratio of 6:2:2, initialize the parameters of the VIT neural network model, set the learning rate, number of iterations, batch size, decay coefficient, etc., use the adaptive Adam optimization algorithm as the optimizer of the VIT neural network model, and use the cross entropy loss to calculate the loss of the model, so as to optimize the VIT neural network model. Among them, the training accuracy is defined as the ratio of the number of samples correctly predicted by the VIT neural network model in the training set to the total number of samples:

[0144] Ac=(TP+TN) / (TP+FN+FP+TN);

[0145] Wherein, Ac is the accuracy of the VIT neural network model; TP (True Positive) means that a Positive judgment is made and the judgment is correct, which means the number of samples predicted by the model as positive examples and actually positive examples; TN (True Negative) means that a Negative judgment is made and the judgment is correct, which means the number of samples predicted by the model as negative examples and actually negative examples; FP (False Positive) means that a Positive judgment is made but the judgment is wrong, which means the number of samples predicted by the model as positive examples but actually negative examples; FN (False Negative) means that a Negative judgment is made but the judgment is wrong, which means the number of samples predicted by the model as negative examples but actually positive examples.

[0146] Accuracy is one of the important indicators for evaluating the performance of classification models. It describes the proportion of correct predictions made by the model. By calculating indicators such as TP, TN, FP, and FN, the prediction ability of the model on the training set can be comprehensively evaluated. During the training process, the parameters of the model (learning rate, number of iterations, batch size, decay coefficient, etc.) are constantly adjusted in the hope of improving the accuracy and generalization ability of the model.

[0147] In S2, grid averaging, statistical filtering, height value replacement, and plane fitting are performed on the first point cloud data to obtain a first cross-sectional view. The principle is the same as the principle of "obtaining multiple second cross-sectional views based on the second point cloud data", which will not be repeated here.

[0148] In S3, the first cross-sectional view is input into the classification model, and the classification model classifies the edge engineering structure according to the first cross-sectional view, outputs the classification result, and records the classification result as the first type.

[0149] In S4, the preset type is the first type.

[0150] In S6, a cross-sectional view of the edge engineering structure in an ideal state is obtained, which is recorded as a fourth cross-sectional view. The confidence of the edge engineering structure is obtained based on the fourth cross-sectional view, and the expression of the confidence is as follows:

[0151]

[0152] Where ZX is the confidence level; Δe1 is the excess volume of the first cross-sectional view, specifically the portion of the first cross-sectional view that exceeds the fourth cross-sectional view; Δe2 is the missing volume of the first cross-sectional view, specifically the portion of the first cross-sectional view that is missing compared to the fourth cross-sectional view; L X is the length of the edge engineering structure; L Y is the width of the edge engineering structure; L Z is the predicted height of the edge engineering structure; L N is the standard height of the edge engineering structure; ρ is the slope of the edge engineering structure, and β is the soil condition.

[0153] Reference Figure 3 , the dotted area is the fourth cross-sectional view, the solid area is the first cross-sectional view, the left oblique line shadow part is Δe1, and the right oblique line shadow part is Δe2, Figure 3 Also shown in L X , L Z , ρ, σ, where σ is the fitted slope of the fourth cross-sectional view.

[0154] Optionally, the influence of the parallel perception system on the reliability judgment of the edge engineering structure is obtained, and based on the influence, the design of the virtual scene of the open-pit mine unloading area, the virtual mine car, the virtual radar and the parameters of the VIT neural network model can be optimized.

[0155] Among them, the calculation formula for the influence of the parallel perception system on the reliability judgment of the edge engineering structure is as follows:

[0156]

[0157] Where MD is the influence of the parallel perception system on the reliability judgment of the edge engineering structure; M α is the influence matrix of the virtual scene; M β is the influence matrix of the virtual minecart; M γ is the influence matrix of the virtual radar; m α1 is the influence of the three-dimensional model of the virtual scene on the reliability judgment of the marginal engineering structure; m α2 is the influence of the three-dimensional model characteristics of the virtual scene on the reliability judgment of the marginal engineering structure; m β1 is the influence of the three-dimensional model of the virtual mine car on the reliability judgment of the edge engineering structure; m β2 is the influence of the three-dimensional model characteristics of the virtual mine car on the reliability judgment of the edge engineering structure; mγ1 is the influence of the virtual radar’s position and posture on the reliability judgment of the edge engineering structure; m γ2 It is the influence of the fitting data of virtual radar on the reliability judgment of edge engineering structure.

[0158] Further,

[0159] m α1 =[r α1 k α1 ];

[0160] m α2 =[r α2 k α2 ];

[0161] m β1 =[r β1 k β1 ];

[0162] m β2 =[r β2 k β2 ];

[0163] m γ1 =[r γ1 k γ1 ];

[0164] m γ2 =[r γ2 k γ2 ];

[0165] In the formula, r α1 k is the influence of changing the 3D model of the virtual scene on the accuracy of the VIT neural network model; α1 is the influence of changing the 3D model of the virtual scene on the confidence accuracy; r α2 The influence of changing the 3D model characteristics of the virtual scene on the accuracy of the VIT neural network model; k α2 The influence of changing the 3D model characteristics of the virtual scene on the confidence accuracy. β1 To change the influence of the 3D model of the virtual mine car on the accuracy of the VIT neural network model; k β1 To change the influence of the three-dimensional model of the virtual mine car on the confidence accuracy; β2 To change the influence of the three-dimensional model characteristics of the virtual mine car on the accuracy of the VIT neural network model; k β2 To change the influence of the three-dimensional model characteristics of the virtual mine car on the confidence accuracy. γ1 The influence of changing the virtual radar's posture on the accuracy of the VIT neural network model; k γ1 The influence of changing the virtual radar's position on the confidence accuracy; r γ2To change the influence of the fitting data of the virtual radar on the accuracy of the VIT neural network model; k γ2 To change the influence of the fitting data of the virtual radar on the confidence accuracy.

[0166] Further,

[0167]

[0168]

[0169]

[0170]

[0171] In the formula, r w1 is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model of the w1th group of virtual scenes (note that when this data changes, other data will remain unchanged, and the same applies to subsequent experimental data); w1 = 1, 2, ..., W1, W1 is a positive integer; k w1 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model of the virtual scene group w1; ZXB is the actual confidence; ZX w1 is the confidence obtained based on the experimental data of the three-dimensional model of the virtual scene group w1. w2 is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model characteristics of the w2th group of virtual scenes; w2 = 1, 2, ..., W2, W2 is a positive integer; k w2 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model characteristics of the w2th group of virtual scenes; ZX w2 is the confidence obtained based on the experimental data of the three-dimensional model characteristics of the w2th group of virtual scenes.

[0172] The actual confidence is obtained through the cross-sectional view of the actual edge engineering structure and the cross-sectional view of the edge engineering structure under an ideal state, and the actual edge engineering structure is constructed after manually measuring relevant data.

[0173]

[0174]

[0175]

[0176]

[0177] In the formula, r w3is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model of the w3th group of virtual mine cars; w3 = 1, 2, ..., W3, W3 is a positive integer; k w3 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model of the virtual mine car in group w3; ZX w3 is the confidence level obtained based on the experimental data of the three-dimensional model of the virtual mine car of group w3. w4 is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model characteristics of the w4th group of virtual mine cars; w4 = 1, 2, ..., W4, W4 is a positive integer; k w4 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model characteristics of the w4th group of virtual mine cars; ZX w4 is the confidence level obtained based on the experimental data of the three-dimensional model characteristics of the w4th group of virtual mine cars.

[0178]

[0179]

[0180]

[0181]

[0182] In the formula, r w5 is the accuracy of the VIT neural network model obtained based on the w5th group of virtual radar posture experimental data; w5 = 1, 2, ..., W5, W5 is a positive integer; k w5 is the accuracy of the confidence obtained based on the pose experimental data of the w5th group of virtual radars; ZX w5 is the confidence obtained based on the pose experimental data of the w5th group of virtual radars. w6 is the accuracy of the VIT neural network model obtained based on the fitting data of the w6th group of virtual radars; w6 = 1, 2, ..., W6, W6 is a positive integer; k w6 is the accuracy of the confidence obtained based on the fitting data of the w6th group of virtual radars; ZX w6 is the confidence obtained based on the fitting data of the w6th group of virtual radars.

[0183] The present invention constructs a parallel perception system by referring to the real unloading area environment, conducts computational experiments in the parallel perception system, uses the constructed virtual scene, virtual mine car, and virtual radar to obtain virtual point cloud data, and then optimizes the parameters through continuous experiments based on the virtual point cloud data, so as to prevent the edge engineering structure judgment method from being only applicable to specific unloading area scenes, reduce the limitations of the method, and by training the VIT neural network model, it can provide initial reference values ​​for parameter estimation in actual scenes and speed up the algorithm iteration speed. Finally, the algorithm designed in the artificial scene is applied to practice, and the artificial system and the designed algorithm are improved and optimized through the data obtained in the actual scene, that is, parallel execution in the real unloading area scene, so as to improve the edge engineering structure detection technology based on millimeter wave radar and improve the stability of unloading operations.

[0184] In summary, the present disclosure can realize the virtual-reality interaction between the real unloading area and the virtual unloading area scenes, significantly increase the accuracy of edge engineering structure detection, and meet the needs of mine cars to perform efficient and safe operations in the unloading area.

[0185] Reference Figure 4 The present disclosure provides a reliability detection system for an unloading zone edge engineering structure, comprising:

[0186] A model building module 101 is used to build a classification model;

[0187] A data acquisition module 102 is used to acquire first point cloud data of a target unloading area, and obtain a first cross-sectional view based on the first point cloud data;

[0188] A structure classification module 103 is used to classify the edge engineering structure based on the first cross-sectional view and the classification model to obtain a first type;

[0189] A first determination module 104, configured to determine that the edge engineering structure is reliable when the first type is a preset type;

[0190] A confidence acquisition module 105, used to acquire the confidence of the edge engineering structure when the first type is not a preset type;

[0191] A second determination module 106, configured to determine whether the edge engineering structure is reliable based on the confidence level;

[0192] Among them, the first profile view is a cross-sectional view of the edge engineering structure of the target unloading area, and the edge engineering structure is a soil structure composed of soil and stones dropped from the ore body during the unloading operation.

[0193] The various variations and specific examples of the above-mentioned method for detecting the reliability of the engineering structure at the edge of the unloading area are also applicable to the system for detecting the reliability of the engineering structure at the edge of the unloading area provided in the present invention. Through the above-mentioned detailed description of the method for detecting the reliability of the engineering structure at the edge of the unloading area, those skilled in the art can clearly know the implementation method of the system for detecting the reliability of the engineering structure at the edge of the unloading area. For the sake of brevity of the specification, it will not be described in detail here.

[0194] The computer device according to the embodiment of the present disclosure includes a memory and a processor. The memory is used to store non-temporary computer-readable instructions. Specifically, the memory may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0195] The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the computer device to perform desired functions. In one embodiment of the present disclosure, the processor is used to run the computer-readable instructions stored in the memory, so that the computer device executes all or part of the steps of the reliability detection method of the unloading zone edge engineering structure of each embodiment of the present disclosure.

[0196] Those skilled in the art should be able to understand that in order to solve the technical problem of how to obtain a good user experience, the present embodiment may also include well-known structures such as a communication bus and an interface, and these well-known structures should also be included in the protection scope of the present disclosure.

[0197] like Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure is shown, which is a schematic diagram of the structure of a computer device suitable for implementing the embodiment of the present disclosure. Figure 5 The computer device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0198] like Figure 5 As shown, the computer device may include a processor (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). In the RAM, various programs and data required for the operation of the computer device are also stored. The processor, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0199] Typically, the following devices can be connected to the I / O interface: input devices such as sensors or visual information acquisition devices; output devices such as display screens; storage devices such as tapes, hard disks, etc.; and communication devices. The communication device can allow the computer device to communicate with other devices (such as edge computing devices) wirelessly or by wire to exchange data. Figure 5 A computer device having various devices is shown, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead.

[0200] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processor, all or part of the steps of the reliability detection method of the unloading area edge engineering structure of the embodiment of the present disclosure are executed.

[0201] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0202] According to the computer-readable storage medium of the embodiment of the present disclosure, non-transitory computer-readable instructions are stored thereon. When the non-transitory computer-readable instructions are executed by the processor, all or part of the steps of the reliability detection method of the unloading zone edge engineering structure of each embodiment of the present disclosure are executed.

[0203] The above-mentioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or mobile hard disk), media with built-in rewritable non-volatile memory (e.g., memory card) and media with built-in ROM (e.g., ROM box).

[0204] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0205] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0206] In the present disclosure, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. The block diagrams of the devices, devices, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0207] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.

[0208] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0209] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.

[0210] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0211] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A reliability detection method for engineering structures at the edge of an unloading zone, characterized in that: include: Build a classification model; Acquire first point cloud data of a target unloading area, and obtain a first cross-sectional view based on the first point cloud data; Classifying the edge engineering structure based on the first cross-sectional view and the classification model to obtain a first type; When the first type is a preset type, determining that the edge engineering structure is reliable; When the first type is not a preset type, obtaining the confidence level of the edge engineering structure; determining whether the edge engineering structure is reliable based on the confidence level; The first cross-sectional view is a cross-sectional view of the edge engineering structure of the target unloading area, and the edge engineering structure is a soil structure composed of soil and rocks dropped from the ore body during the unloading operation; Wherein, constructing the classification model includes: Construct a virtual unloading area based on the real unloading area; obtaining second point cloud data based on the real unloading area and the virtual unloading area; Obtaining a plurality of second cross-sectional views based on the second point cloud data; Classifying and labeling the plurality of second cross-sectional views to obtain a plurality of third cross-sectional views; Training the VIT neural network model based on the plurality of third profile images to obtain the classification model; The reliability detection method of the unloading zone edge engineering structure also includes: Obtain the influence of the parallel perception system on the reliability judgment of the edge engineering structure, and optimize the design of the virtual scene, virtual mine car, virtual radar and the parameters of the VIT neural network model in the open-pit mine unloading area based on the influence; Among them, the calculation formula for the influence of the parallel perception system on the reliability judgment of the edge engineering structure is as follows: Where MD is the influence of the parallel perception system on the reliability judgment of the edge engineering structure; M α is the influence matrix of the virtual scene; M β is the influence matrix of the virtual minecart; M γ is the influence matrix of the virtual radar; m α1 is the influence of the three-dimensional model of the virtual scene on the reliability judgment of the marginal engineering structure; m α2 is the influence of the three-dimensional model characteristics of the virtual scene on the reliability judgment of the marginal engineering structure; m β1 is the influence of the three-dimensional model of the virtual mine car on the reliability judgment of the edge engineering structure; m β2 is the influence of the three-dimensional model characteristics of the virtual mine car on the reliability judgment of the edge engineering structure; m γ1 is the influence of the virtual radar’s position and posture on the reliability judgment of the edge engineering structure; m γ2 The influence of the fitting data of virtual radar on the reliability judgment of edge engineering structure; m α1 =[r α1 k α1 ]; m α2 =[r α2 k α2 ]; m β1 =[r β1 k β1 ]; m β2 =[r β2 k β2 ]; m γ1 =[r γ1 k γ1 ]; m γ2 =[r γ2 k γ2 ]; In the formula, r α1 k is the influence of changing the 3D model of the virtual scene on the accuracy of the VIT neural network model; α1 is the influence of changing the 3D model of the virtual scene on the confidence accuracy; r α2 The influence of changing the 3D model characteristics of the virtual scene on the accuracy of the VIT neural network model; k α2 The influence of changing the 3D model characteristics of the virtual scene on the confidence accuracy; r β1 To change the influence of the 3D model of the virtual mine car on the accuracy of the VIT neural network model; k β1 To change the influence of the three-dimensional model of the virtual mine car on the confidence accuracy; β2 To change the influence of the three-dimensional model characteristics of the virtual mine car on the accuracy of the VIT neural network model; k β2 To change the influence of the three-dimensional model characteristics of the virtual mine car on the confidence accuracy; γ1 The influence of changing the virtual radar's posture on the accuracy of the VIT neural network model; k γ1 The influence of changing the virtual radar's position on the confidence accuracy; r γ2 To change the influence of the fitting data of the virtual radar on the accuracy of the VIT neural network model; k γ2 To change the influence of the fitting data of the virtual radar on the confidence accuracy; In the formula, r w1 is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model of the w1th group of virtual scenes; w1 = 1, 2, ..., W1, W1 is a positive integer; k w1 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model of the virtual scene group w1; ZXB is the actual confidence; ZX w1 is the confidence obtained based on the experimental data of the three-dimensional model of the virtual scene group w1; r w2 is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model characteristics of the w2th group of virtual scenes; w2 = 1, 2, ..., W2, W2 is a positive integer; k w2 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model characteristics of the w2th group of virtual scenes; ZX w2 is the confidence obtained based on the experimental data of the three-dimensional model characteristics of the w2th group of virtual scenes; The actual confidence is obtained through the cross-sectional view of the actual edge engineering structure and the cross-sectional view of the edge engineering structure under an ideal state, and the actual edge engineering structure is constructed after manually measuring relevant data; In the formula, r w3 is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model of the w3th group of virtual mine cars; w3 = 1, 2, ..., W3, W3 is a positive integer; k w3 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model of the virtual mine car in group w3; ZX w3 is the confidence obtained based on the experimental data of the three-dimensional model of the virtual mine car of group w3; r w4 is the accuracy of the VIT neural network model obtained based on the experimental data of the three-dimensional model characteristics of the w4th group of virtual mine cars; w4 = 1, 2, ..., W4, W4 is a positive integer; k w4 is the accuracy of the confidence obtained based on the experimental data of the three-dimensional model characteristics of the w4th group of virtual mine cars; ZX w4 is the confidence level obtained based on the experimental data of the three-dimensional model characteristics of the w4th group of virtual mine cars; In the formula, r w5 is the accuracy of the VIT neural network model obtained based on the w5th group of virtual radar posture experimental data; w5 = 1, 2, ..., W5, W5 is a positive integer; k w5 is the accuracy of the confidence obtained based on the pose experimental data of the w5th group of virtual radars; ZX w5 is the confidence obtained based on the pose experimental data of the w5th group of virtual radars; r w6 is the accuracy of the VIT neural network model obtained based on the fitting data of the w6th group of virtual radars; w6 = 1, 2, ..., W6, W6 is a positive integer; k w6 is the accuracy of the confidence obtained based on the fitting data of the w6th group of virtual radars; ZX w6 is the confidence obtained based on the fitting data of the w6th group of virtual radars.

2. The reliability detection method of the unloading zone edge engineering structure according to claim 1 is characterized in that: The step of classifying and labeling the plurality of second cross-sectional views to obtain a plurality of third cross-sectional views comprises: determining a first corner point and a second corner point of the second cross-sectional view; Obtaining a height value and a slope angle of the second profile based on the first corner point and the second corner point; classifying the second profile image based on the height value and the slope angle of the second profile image; The second cross-sectional view is labeled according to the classification result to obtain the third cross-sectional view.

3. The reliability detection method of the unloading zone edge engineering structure according to claim 2 is characterized in that: The calculation formula of the height value of the second profile is as follows: D z =ΔZ=|Z2-Z1|; Among them, D z is the height value of the second cross-sectional view; ΔZ is the height displacement between the second corner point and the first corner point; Z2 is the height value of the second corner point; Z1 is the height value of the first corner point.

4. The reliability detection method of the unloading zone edge engineering structure according to claim 3 is characterized in that: The calculation formula of the slope angle of the second profile is as follows: Among them, γ is the slope angle of the second profile; ΔX is the lateral displacement between the second corner point and the first corner point, ΔX=|X2-X1|, X2 is the lateral coordinate value of the second corner point, and X1 is the lateral coordinate value of the first corner point; ΔY is the longitudinal displacement between the second corner point and the first corner point, ΔY=|Y2-Y1|, Y2 is the longitudinal coordinate value of the second corner point, and Y1 is the longitudinal coordinate value of the first corner point.

5. The reliability detection method of the unloading zone edge engineering structure according to claim 4 is characterized in that: The confidence calculation formula is as follows: Wherein, ZX is the confidence level; Δe1 is the excess volume of the first cross-sectional image; Δe2 is the missing volume of the first cross-sectional image; L X is the length of the edge engineering structure; L Y is the width of the edge engineering structure; L Z is the predicted height of the edge engineering structure; L N is the standard height of the edge engineering structure; ρ is the slope of the edge engineering structure, and β is the soil condition.

6. The reliability detection method of the unloading zone edge engineering structure according to claim 5 is characterized in that: The determining whether the edge engineering structure is reliable based on the confidence level includes: Determining whether the confidence level is greater than a preset confidence threshold; If so, it is determined that the edge engineering structure is reliable; If not, it is determined that the edge engineering structure is unreliable.

7. A computer device, characterized in that: The computer device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the open-pit mine unloading area safe operation method described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which are used to enable a computer to execute the open-pit mine unloading area safe operation method described in any one of claims 1-6.

9. A computer program product comprising computer instructions, characterized in that When the computer instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images