Multi-element blasting effect evaluation method and system based on point cloud data

Abstract

The invention discloses a multi-element blasting effect evaluation method and system based on point cloud data, and relates to the field of blasting effect evaluation of a drilling and blasting method.The method comprises the steps that point cloud obtained after tunnel blasting is scanned through a three-dimensional laser scanner, and classification of muck pile point cloud, tunnel face point cloud and hole body point cloud is achieved through a Canupo algorithm; the muck pile point cloud, the tunnel face point cloud and the hole body point cloud are processed, and the boulder rate, the throwing distance, the excavation footage, the blast hole utilization rate, the tunnel half hole and the over-excavation and under-excavation quantity index parameters are obtained; and obtaining a blasting effect evaluation score through the combination weight and the score of each index parameter, and obtaining blasting effect evaluation. According to the method, multiple blasting effect evaluation indexes can be accurately obtained based on the point cloud data obtained through three-dimensional laser scanning, the blasting effect is quantitatively evaluated, a large amount of manpower is saved, the blasting effect evaluation efficiency and accuracy are improved, and reference is provided for improvement of subsequent blasting design.

Description

Technical Field

[0001] The present invention relates to the field of evaluation of blasting effects by drill and blast method, and particularly to a method and system for multi - element blasting effect evaluation based on point cloud data. Background Technique

[0002] Tunnel excavation by drill and blast method refers to a construction method of excavating tunnels by drilling holes, loading explosives, and blasting rocks. Tunnel excavation by drill and blast method has the characteristics of low cost and strong geological adaptability.

[0003] Currently, the evaluation technology of blasting effects by drill and blast method faces problems such as insufficient accuracy in obtaining blasting effect evaluation parameters and great difficulty in obtaining blasting evaluation parameters. Through three - dimensional laser scanning technology, multiple blasting effect evaluation parameters can be quickly and accurately obtained, and data processing can be quickly realized to generate a blasting effect evaluation, and accordingly, the blasting design can be adjusted in real time.

[0004] Point cloud data is a data structure composed of a large number of points with coordinate information in three - dimensional space. Point cloud data can truly reflect the three - dimensional geometric shape and surface characteristics of an object, and has the characteristics of high precision and high density. Point cloud data can be obtained through three - dimensional laser scanning and other surveying and mapping technologies, and is usually used in fields such as geographic information systems and surveying and mapping, architecture and civil engineering, and cultural relics protection. Summary of the Invention

[0005] To solve the problems existing in the prior art, the purpose of the present invention is to provide a method for multi - element blasting effect evaluation based on point cloud data. The present invention can obtain multiple blasting effect evaluation indexes based on the point cloud after tunnel blasting, and proposes a method for evaluating the blasting effect based on the extracted indexes, which has the characteristics of high efficiency, intelligence, high precision, low labor cost, and quantization of blasting effect evaluation.

[0006] Specifically, the technical solution adopted by the present invention is as follows: A method for multi - element blasting effect evaluation based on point cloud data, comprising the following steps: S1. Scan the tunnel after blasting with a three - dimensional laser scanner to obtain point cloud data; classify the obtained point cloud data into face - of - tunnel - heading point cloud, muck - pile point cloud, and tunnel - body point cloud through the Canupo algorithm; S2. Obtain the excavation footage of this blasting by calculating the shortest distance between the face - of - tunnel - heading point cloud obtained from this scan and the face - of - tunnel - heading point cloud obtained from the previous blasting, and obtain the hole utilization rate through the actual blasting design and excavation footage of the project; S3. Process the muck - pile point cloud, output the muck - pile fragment size distribution data, and obtain the threshold for judging the large - block rate according to the actual situation of the project, and then obtain the large - block rate index; obtain the throwing distance according to the relationship between the muck - pile point cloud and the face - of - tunnel - heading point cloud; S4. Process the point cloud of the tunnel body, fit the semi-hole point cloud, obtain the semi-hole rate index in combination with the actual project, generate a tunnel design contour model according to the actual tunnel design, compare the obtained point cloud with the tunnel design contour model, and determine the overbreak and underbreak of the tunnel; S5. Determine the qualified values and expected values of each index parameter in combination with the tunnel blasting design and the actual project, and assign scores to the semi-hole rate, hole utilization rate, excavation footage, large block rate, throwing distance, and tunnel overbreak and underbreak indexes; evaluate the blasting effect by weighted calculation of the blasting effect evaluation score.

[0007] Further, the step S3 specifically includes the following sub-steps: S301. Use the clustering algorithm to segment the point cloud of the muck pile to obtain the segmented point cloud of the muck pile blocks; S302. Take the three-dimensional space distance between the two farthest points in each segmented point cloud of the muck pile blocks as the block size of the corresponding block; set the threshold of the large block rate according to the scale of the loading equipment used in the actual project; S303. Compare the block size of each block with the threshold to obtain the large block rate index.

[0008] Further, the step S4 specifically includes the following sub-steps: S401. Obtain the semi-hole point cloud by performing semi-circle fitting on the point cloud of the tunnel body; S402. Combine the semi-hole point cloud, measure the total length of the remaining semi-holes, compare it with the total length of the pre-set blast holes in the actual project, and obtain the semi-hole rate index; S403. Generate a tunnel design contour model according to the actual tunnel design, and take the distance difference between the obtained point cloud and the tunnel design contour as the overbreak and underbreak of the tunnel; S404. Combine the tunnel excavation footage to obtain the average linear overbreak and average linear underbreak of the tunnel.

[0009] Further, in the step S5, the specific process of score assignment is as follows: Assign scores to each extracted blasting effect evaluation parameter, determine the expected value and qualified value of each index according to the tunnel blasting design and the actual project, take the semi-hole rate, hole utilization rate, and excavation footage as positive indexes, and the large block rate, throwing distance, and tunnel overbreak and underbreak as negative indexes, and assign scores to each index. Among them, the score calculation processes of the positive index and the negative index are specifically expressed as: Positive index score assignment: ; Negative index score assignment: ; Among them, the represents each evaluation index parameter, and the Indicates the qualified values corresponding to each index parameter Is the expected value corresponding to each index parameter, and the index parameters include semi-hole rate, hole utilization rate, excavation footage, large block rate, throwing distance, and tunnel overbreak and underbreak amount indicators

[0010] Further, in the step S5, the specific process of calculating the blasting effect evaluation score by weighted calculation is as follows S501. Determine the weight of the extracted blasting effect evaluation parameters in the blasting effect evaluation model, and determine the subjective weight of each evaluation parameter in the blasting effect evaluation according to the expert evaluation method ; S502. Determine the objective weight according to the CRITIC method, and perform standardization processing on each index based on the blasting effect evaluation parameters extracted after multiple blasts to obtain standardized indexes S503. Construct a decision matrix based on the extracted large block rate, throwing distance, semi-hole rate, average linear overbreak of the tunnel, average linear underbreak of the tunnel, excavation footage, and hole utilization rate indexes ; S504. Calculate the correlation coefficient according to the standard deviation And covariance between each index, and determine the information content contained in each index S505. Determine the objective weight of each index according to the information content contained in each index ; And obtain the combined weight according to the minimum discrimination information principle combined with the objective weights of each index

[0011] Further, in the step S503, the decision matrix Specifically expressed as ; Among them, the Represents the th sample's th index value, and the Represents the total number of samples

[0012] Further, in the step S504, the calculation process of the correlation coefficient is specifically expressed as ; The calculation process of determining the information content contained in each index is specifically expressed as ; Among them, the Represents the th item index and the th item index's Represents the The covariance of the j-th index and the -th index, where the represents the standard deviation of the j-th index, and the represents the amount of information contained in the j-th index.

[0013] Furthermore, in the step S505, the calculation process of the objective weight is specifically expressed as: ; Obtain the combined weight according to the principle of minimum discrimination information , where the calculation process of the combined weight is specifically expressed as: ; where the represents the combined weight of the j-th index in the evaluation, the represents the subjective weight of the j-th index in the evaluation, and the represents the objective weight of the j-th index in the evaluation.

[0014] Furthermore, in the step S5, the calculation process of the blasting effect evaluation score is specifically expressed as: ; Based on the value, the judgment criteria for the blasting effect are: when , it is considered that the blasting effect is good; when , it is considered that the blasting effect is better; when , it is considered that the blasting effect is poor; when , it is considered that the blasting effect is very poor.

[0015] A system for multi - factor blasting effect evaluation based on point cloud data, which is implemented based on the method for multi - factor blasting effect evaluation based on point cloud data described in any one of the above, and includes: A data acquisition module, which is used to obtain the point cloud data of the tunnel after blasting through 3D laser scanning; A data pre - processing module, which is used to denoise the point cloud data and classify the muck pile point cloud, tunnel body point cloud, and heading face point cloud; An evaluation index extraction module, which is used to process the point cloud data and obtain the blasting effect evaluation parameters such as the large - block rate, throwing distance, average linear over - break and under - break of the tunnel, blasting advance, and explosive consumption per unit of explosive according to the actual engineering situation; An index parameter evaluation module, which is used to score the extracted blasting effect evaluation index parameters and determine the weight of each index in the blasting effect evaluation; The blasting effect evaluation module is used to obtain the blasting effect evaluation score based on the scores of various evaluation indicators and the weights of each indicator, and obtain the blasting effect evaluation according to the obtained blasting effect evaluation score.

[0016] The beneficial effects of the present invention are as follows: The present invention can accurately obtain a variety of blasting effect evaluation indicators at one time based on the point cloud data obtained by 3D laser scanning, and quantitatively evaluate the blasting effect, saving a large amount of manpower, improving the efficiency of blasting effect evaluation, and at the same time avoiding problems such as incomplete data collection, low accuracy, and high time cost caused by manual data collection, providing a reference for the improvement of subsequent blasting design. Description of the Drawings

[0017] Figure 1 It is a flowchart of the method of the embodiment of the present invention; Figure 2 It is a system block diagram of the embodiment of the present invention. Detailed Embodiments

[0018] The embodiments of the present invention will be described in detail below with reference to the drawings. Embodiment

[0019] As Figure 1 shown, a method for multi - element blasting effect evaluation based on point cloud data includes the following steps: A method for multi - element blasting effect evaluation based on point cloud data includes the following steps: S1. Scan the tunnel after blasting with a 3D laser scanner to obtain point cloud data; classify the obtained point cloud data into face - of - tunnel - face point cloud, muck - pile point cloud, and tunnel - body point cloud through the Canupo algorithm; S2. Obtain the excavation footage of this blasting by calculating the shortest distance between the face - of - tunnel - face point cloud obtained from this scan and the face - of - tunnel - face point cloud obtained from the previous blasting, and obtain the hole - utilization rate through the actual blasting design and excavation footage of the project; S3. Process the muck - pile point cloud, output the muck - pile fragment size distribution data, obtain the threshold for judging the large - block rate according to the actual situation of the project, and then obtain the large - block rate index; obtain the throwing distance according to the relationship between the muck - pile point cloud and the face - of - tunnel - face point cloud; S4. Process the tunnel - body point cloud, fit out the half - hole point cloud, obtain the half - hole rate index in combination with the actual situation of the project, generate a tunnel design contour model according to the actual tunnel design, compare the obtained point cloud with the tunnel design contour model, and determine the over - break and under - break of the tunnel; S5. Determine the qualified values and expected values of each index parameter in combination with tunnel blasting design and engineering practice, assign scores to the semi-hole rate, blasthole utilization rate, excavation footage, large block rate, throwing distance, and tunnel overbreak and underbreak amount indicators; calculate the blasting effect evaluation score through weighted calculation, and evaluate the blasting effect according to the evaluation score.

[0020] Specifically, the specific process principle of step S1 is as follows: The 3D laser scanner uses the Time-of-Flight (TOF) or phase difference detection technology to accurately measure the three-dimensional coordinates by emitting high-frequency laser beams to the target area and capturing the time delay or phase difference of the reflected signals. The point cloud data generated by the scan has high density and high precision, and can completely reflect the geometric shape and its characteristics of the tunnel inner surface. After the initial processing of the point cloud data, the Canupo algorithm is used for classification. The Canupo algorithm is based on the local geometric features of points, uses the multi-scale feature extraction method, and realizes the automatic classification of the point cloud by constructing a Support Vector Machine (SVM) classifier.

[0021] Exemplarily, an implementation process of automatic classification is proposed: within the neighborhood of each point, extract the multi-dimensional feature vector describing the point cloud morphology by calculating its normal vector, curvature, and statistical features of the local point distribution; by gradually expanding the neighborhood range, extract the features of the point cloud at different scales, and synthesize these features to form a high-dimensional feature space to describe the local geometric attributes of the point cloud; based on the manually labeled sample data, use the Support Vector Machine (SVM) to construct a classifier, optimize the classification hyperplane to maximize the interval between classes; apply the trained classifier to the entire point cloud data set and divide it into the face point cloud, muck pile point cloud, and tunnel body point cloud according to the geometric characteristics.

[0022] Specifically, the specific process principle of step S2 is as follows: Through the Iterative Closest Point (ICP) algorithm, the face point clouds of the two scans are accurately aligned. The ICP algorithm iteratively minimizes the Euclidean distance between the two point clouds to find the optimal rigid body transformation matrix (including translation and rotation) to achieve the spatial registration of the point cloud; for the registered point cloud, calculate the shortest distance between the current point cloud and the previous point cloud through the nearest neighbor search algorithm (such as KD-Tree). The KD-Tree structure can efficiently search for the nearest point pair in the high-dimensional space; according to the distribution of the shortest distances of all point pairs, calculate the overall average displacement or minimum displacement of the point cloud as the actual tunnel excavation footage. This method can effectively filter out local noise points and ensure the accuracy of the calculation results.

[0023] Further, step S3 specifically includes the following sub-steps: S301. The muck pile point cloud is segmented using a clustering algorithm to obtain the segmented muck block point cloud; S302. The three-dimensional spatial distance between the two farthest points in each of the segmented muck block point clouds is taken as the block size of the corresponding block; according to the scale of the loading equipment used in the actual project, a threshold value of the large block ratio is set; S303. Compare the block size of each block with the threshold value to obtain the large block ratio index.

[0024] Specifically, the muck pile point cloud is segmented by a clustering algorithm (such as DBSCAN or K-Means) to identify the independent block point cloud: Exemplarily, DBSCAN utilizes the density characteristics of the point cloud and clusters adjacent points into the same cluster by defining a distance threshold and a minimum number of points. For the muck pile point cloud with relatively uniform density, the K-Means algorithm divides the point cloud into a predefined number of blocks by iteratively optimizing the position of the cluster center.

[0025] Furthermore, the step S4 specifically includes the following sub-steps: S401. By performing semi-circle fitting on the tunnel body point cloud, obtain the semi-hole point cloud; S402. Combine the semi-hole point cloud, measure the total length of the remaining semi-holes, and compare it with the total length of the pre-set blast holes in the actual project to obtain the semi-hole rate index; S403. According to the actual tunnel design, generate a tunnel design contour model, and take the distance difference between the obtained point cloud and the tunnel design contour as the tunnel overbreak and underbreak; S404. Combine the tunnel excavation footage to obtain the average linear overbreak and the average linear underbreak of the tunnel.

[0026] Specifically, for the semi-circle fitting in the step S401, exemplarily, a fitting algorithm (such as the least squares method) is used to fit an arc to the tunnel body point cloud to identify and extract the semi-hole point cloud of the blasting residue. The fitting model optimizes the arc parameters, namely the center coordinates and the radius, by minimizing the distance error between the point cloud and the arc.

[0027] Furthermore, in the step S5, the specific process of score assignment is as follows: Assign scores to each extracted blasting effect evaluation parameter, determine the expected value and the qualified value of each index according to the tunnel blasting design and the actual project. Take the semi-hole rate, the blast hole utilization rate, and the excavation footage as positive indicators, and the large block ratio, the throwing distance, and the tunnel overbreak and underbreak as negative indicators, and assign scores to each index. Among them, the score calculation processes of the positive and negative indicators are specifically expressed as: Score assignment for positive indicators: ; Score assignment for negative indicators: ; Among them, the represents the parameters of each evaluation index, and the represents the qualified values corresponding to each index parameter, is the expected value corresponding to each index parameter. The index parameters include semi-hole rate, hole utilization rate, excavation footage, large block rate, throwing distance, and tunnel overbreak and underbreak amount indicators.

[0028] Further, in step S5, the specific process of calculating the blasting effect evaluation score through weighted calculation is as follows: S501. Determine the weights of the extracted blasting effect evaluation parameters in the blasting effect evaluation model, and determine the subjective weights of each evaluation parameter in the blasting effect evaluation according to the expert evaluation method ; S502. Determine the objective weights according to the CRITIC method, and perform standardization processing on each index based on the extracted blasting effect evaluation parameters after multiple blasts to obtain standardized indexes; S503. Construct a decision matrix based on the extracted large block rate, throwing distance, semi-hole rate, average linear overbreak of the tunnel, average linear underbreak of the tunnel, excavation footage, and hole utilization rate indicators ; S504. Calculate the correlation coefficient according to the standard deviation and covariance between each index, and determine the information content contained in each index; S505. Determine the objective weights of each index according to the information content contained in each index ; and obtain the combined weights according to the minimum discrimination information principle combined with the objective weights of each index.

[0029] Further, for step S501, in the expert evaluation method for determining each evaluation parameter, for example, the Delphi method or the Analytic Hierarchy Process (AHP) can be used to organize multiple blasting technology experts to score the importance of each index to form an expert scoring matrix.

[0030] Further, in step S503, the decision matrix is specifically expressed as: ; wherein, the represents the value of the th index of the th sample, and the represents the total number of samples.

[0031] Further, in step S504, the calculation process of the correlation coefficient is specifically expressed as: ; The calculation process for determining the information content of each index is specifically expressed as follows: ; Among them, the represents the correlation coefficient between the -th index and the -th index. The represents the covariance between the -th index and the -th index. The represents the standard deviation of the j-th index, and the represents the information content contained in the j-th index.

[0032] Furthermore, in the step S505, the calculation process of the objective weight is specifically expressed as follows: ; The combined weight is obtained according to the principle of minimum discrimination information , and the calculation process of the combined weight is specifically expressed as follows: ; Among them, the represents the combined weight of the j-th index in the evaluation, the represents the subjective weight of the j-th index in the evaluation, and the represents the objective weight of the j-th index in the evaluation.

[0033] Furthermore, in the step S5, the calculation process of the blasting effect evaluation score is specifically expressed as follows: ; Based on the value, the judgment criteria for the blasting effect are as follows: when , it is considered that the blasting effect is good; when , it is considered that the blasting effect is relatively good; when , it is considered that the blasting effect is relatively poor; when , it is considered that the blasting effect is poor.

[0034] Furthermore, as a preferred implementation manner of this embodiment, a system for multi - factor blasting effect evaluation based on point cloud data is also provided. As shown in Figure 2 , it includes: Data acquisition module: acquiring the point cloud data of the tunnel after blasting through 3D laser scanning; Data pre - processing module: denoising the point cloud data and classifying the muck pile point cloud, tunnel body point cloud, and heading face point cloud; Evaluation index extraction module: used to process point cloud data and combine with engineering practice to obtain blasting effect evaluation parameters such as large block ratio, throwing distance, average linear overbreak and underbreak of tunnel, blasting advance, specific charge, etc.; Index parameter evaluation module: assign scores to the extracted blasting effect evaluation index parameters and determine the weights of each index in the blasting effect evaluation; Blasting effect evaluation module: obtain the blasting effect evaluation score according to the scores of each evaluation index and the weights of each index, and obtain the blasting effect evaluation based on the obtained blasting effect evaluation score.

[0035] Furthermore, the specific implementation process principle of the above embodiment is as follows: Train a classifier by constructing it according to the multi-scale - dimensional feature analysis principle through the Canupo algorithm. After a certain tunnel blasting, manually segment the obtained point cloud into muck pile point cloud, face point cloud, and tunnel body point cloud samples, and train the classifier based on the samples and use this classifier for the point cloud obtained after other tunnel blasts, so as to realize the classification of muck pile body point cloud, tunnel body point cloud, and face point cloud; The muck pile point cloud is segmented by using the clustering algorithm to obtain the segmented muck pile block point cloud. The three-dimensional space distance between the two farthest points in each segmented muck pile block point cloud is used as the block size of the corresponding block; according to the scale of the loading equipment used in engineering practice, obtain the threshold of the large block ratio, and compare the block size of each block with the threshold to obtain the large block ratio index.

[0036] The tunnel body point cloud is processed by semi-circular fitting to obtain the semi-hole point cloud, measure the total length of the remaining semi-holes, compare it with the total length of the preset blast holes in engineering practice to obtain the semi-hole rate index; according to the actual tunnel design, generate a tunnel design contour model, and use the distance difference between the obtained point cloud and the tunnel design contour as the tunnel overbreak and underbreak, and combine with the tunnel excavation advance to obtain the average linear overbreak and average linear underbreak of the tunnel.

[0037] Assign scores to each extracted blasting effect evaluation parameter. The tunnel blasting design and engineering practice need to determine the expected value and qualified value of each index. The semi-hole rate, hole utilization rate, and excavation advance are positive indicators, and the large block ratio, throwing distance, and tunnel overbreak and underbreak are negative indicators; assign scores to each index. The following are the score calculation methods for positive and negative indicators: Positive indicator score assignment: ; Negative indicator score assignment: ; Among them, the represents each evaluation index parameter, and the represents the qualified value corresponding to each index parameter, To represent the expected values corresponding to various index parameters, the index parameters include the semi-hole rate, the hole utilization rate, the excavation footage, the large block rate, the throwing distance, and the tunnel overbreak and underbreak amount indicators.

[0038] Determine the weights of the extracted blasting effect evaluation parameters in the blasting effect evaluation model, and determine the subjective weight values of each evaluation parameter in the blasting effect evaluation according to the expert evaluation method ; Determine the objective weights according to the CRITIC method, and standardize each index according to the extracted blasting effect evaluation parameters after multiple blasts to obtain standardized indexes And construct a decision matrix based on the 7 indexes of the extracted large block rate, throwing distance, semi-hole rate, average linear overbreak of the tunnel, average linear underbreak of the tunnel, excavation footage, and hole utilization rate Among them, the formulas for index standardization and the decision matrix are: Positive index standardization processing: ; Reverse index standardization processing: ; Among them, the i represents the number of samples, the j represents the index of the index, the represents the positive index after standardization processing, and the represents the reverse index after standardization processing.

[0039] Decision matrix : ; Among them, the represents the value of the th sample of the th index, and the represents the total number of samples.

[0040] Calculate the correlation coefficient according to the standard deviation and covariance between each index, and determine the information content of each index. The calculation formulas for the correlation coefficient and the determined information content are: ; The calculation process for determining the information content of each index is specifically expressed as: ; Among them, the represents the correlation coefficient between the th index and the th index, and the represents the covariance between the th index and the th index. represents the standard deviation of the j-th index, and the represents the information content contained in the j-th index.

[0041] Determine the objective weights of each index according to the information content contained in each index. The calculation formula for the objective weights is: ; Obtain the combined weights according to the principle of minimum discrimination information. The calculation formula for the combined weights is: ; Combine the scores of each evaluation index with the weights to calculate the blasting effect evaluation score , and based on 's value, obtain the blasting effect evaluation. The calculation formula is: ; Based on 's value, the blasting effect judgment criterion is: when , it is considered that the blasting effect is good; when , it is considered that the blasting effect is better; when , it is considered that the blasting effect is poor; when , it is considered that the blasting effect is very poor.

[0042] Furthermore, as a preferred specific implementation manner of the above embodiment, a multi-objective evaluation method based on dynamic weights is proposed, which specifically includes the following steps: S51. Construct a multi-objective optimization model according to the scores of positive index scoring and reverse index scoring. The objective function is to maximize the scores of positive indexes and minimize the scores of reverse indexes; S52. Introduce the particle swarm optimization algorithm to dynamically adjust the weights of each index and optimize the objective function; S53. Combine the fuzzy comprehensive evaluation method, map the scores of positive indexes and reverse indexes to the fuzzy set, and calculate the comprehensive evaluation score through fuzzy reasoning; S54. Calculate the blasting effect evaluation score according to the optimized weights and the fuzzy reasoning result.

[0043] Furthermore, in the step S51, the specific construction method of the multi-objective optimization model includes: Construct the objective function of the multi-objective optimization model: Objective 1: Maximize the sum of scores of positive indexes: , where the represents the objective function of maximizing the sum of scores of positive indexes; Objective 2: Minimize the sum of scores of reverse indexes: , where the An objective function representing the minimization of the sum of the reverse index scores; Further, in the step S52, the particle swarm optimization algorithm is specifically represented as: ; Among them, the represents the velocity of particle i at time t + 1, the represents the inertia weight, the represents the velocity of particle i at time t, the represents the step size for controlling the movement of the particle towards the historical optimal position, the represents a random number for introducing randomness to the historical optimal position, the represents the step size for controlling the movement of the particle towards the global optimal position, the represents a random number for introducing randomness to the global optimal position, the represents the historical optimal position of particle i, the represents, the represents the position of particle i at time t, the represents the position of particle i at time t + 1; The particles in the particle swarm optimization algorithm represent weight vectors, that is, the weights of each index in the evaluation of the blasting effect. The position of each particle then represents a weight vector.

[0044] Further, the step S52 specifically includes the following sub-steps: S521. Randomly generate a group of particles, and the position of each particle represents a set of weights; S522. Initialize the velocity and historical optimal position of the particles; calculate the fitness value of each particle, and determine the global optimal position; S523. For each particle, calculate the new velocity and position according to the update formula; check whether the new position meets the constraint conditions, and if not, make adjustments; S524. Calculate the fitness value of the new position; update the historical optimal position and global optimal position of the particles; when the maximum number of iterations is reached or the fitness value converges, stop the iteration and output the global optimal position, that is, the optimal weight.

[0045] Further, in the S522, calculating the fitness value of each particle is specifically to evaluate the quality of the weight vector through the fitness function. Based on the normalized index, the fitness function is defined as: ; Among them, the represents the fitness function, and the represents the weight vector.

[0046] Further, the step S53 specifically includes the following sub-steps: S531. Define the fuzzy membership function for the scores of positive and negative indicators: ; Among them, the represents the membership degree of the fuzzy set A, and the and represent the boundary values of the fuzzy set; S532. Perform reasoning through the fuzzy rule base to generate a comprehensive evaluation score: S533. Use the centroid method to convert the fuzzy score into a specific value: ; Among them, the represents the defuzzified value of the i-th indicator; S534. Calculate the blasting effect evaluation score according to the weights optimized by the particle swarm optimization algorithm: ; Among them, the represents the blasting effect evaluation score.

[0047] Exemplarily, a blasting effect evaluation applying a multi-objective evaluation method based on dynamic weights is given: .

[0048] The above embodiments only represent the specific implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention.

Claims

1. A method for multivariate blasting effect evaluation based on point cloud data, characterized in that: The following steps are involved: S1. Scan the tunnel after blasting with a 3D laser scanner to obtain point cloud data; classify the obtained point cloud data into face point cloud, blast pile point cloud and tunnel body point cloud through the Canupo algorithm; S2. Obtain the excavation footage of this blasting by calculating the shortest distance between the tunnel face point cloud obtained by this scan and the tunnel face point cloud obtained by the previous blasting, and obtain the blasthole utilization rate through the actual blasting design and excavation footage of the project; S3. Process the point cloud of the blast pile, output the blast pile block distribution data, and obtain the threshold for judging the large block rate based on the actual project, and then obtain the large block rate index; obtain the throwing distance based on the relationship between the blast pile point cloud and the tunnel face point cloud; S4. Process the tunnel body point cloud, fit the half-hole point cloud, and obtain the half-hole ratio index based on the actual project. Generate the tunnel design contour model according to the actual tunnel design, compare the obtained point cloud with the tunnel design contour model, and determine the tunnel over-excavation and under-excavation amount; S5. Determine the qualified value and expected value of each parameter in combination with tunnel blasting design and engineering practice, and assign points to the semi-hole rate, blasthole utilization rate, excavation footage, large block rate, throwing distance and tunnel over-excavation and under-excavation indicators; calculate the blasting effect evaluation score through weighted calculation, and evaluate the blasting effect according to the evaluation score.

2. A method for multivariate blasting effect evaluation based on point cloud data as claimed in claim 1, characterized in that: The S3 specifically includes the following sub-steps: S301. The explosion pile point cloud is segmented using a clustering algorithm to obtain a segmented explosion pile block point cloud; S302. The three-dimensional distance between the two farthest points in the segmented point cloud of each blast pile block is the block size of the corresponding block; according to the scale of the shovel loading equipment used in the actual project, the threshold value of the large block rate is set; S303. Compare the blockiness of each block with the threshold to obtain a large block rate index.

3. The method for multivariate blasting effect evaluation based on point cloud data according to claim 1, characterized in that: The S4 specifically includes the following sub-steps: S401. A half-hole point cloud is obtained by performing semicircle fitting on the tunnel body point cloud; S402. Combined with the half-hole point cloud, the total length of the remaining half-hole is measured and compared with the total length of the actual preset blasthole in the project to obtain a half-porosity index; S403. Generate a tunnel design contour model according to the actual tunnel design, and use the distance difference between the obtained point cloud and the tunnel design contour as the tunnel over-excavation amount; S404. Based on the tunnel excavation progress, the average linear over-excavation and the average linear under-excavation of the tunnel are obtained.

4. The method for multivariate blasting effect evaluation based on point cloud data according to claim 1, characterized in that: In S5, the specific process of scoring is as follows: scoring each extracted blasting effect evaluation parameter, determining the expected value and qualified value of each indicator according to the tunnel blasting design and engineering practice, taking the semi-hole rate, blasthole utilization rate, and excavation footage as positive indicators, and the large block rate, throwing distance, and tunnel over-excavation and under-excavation as reverse indicators, scoring each indicator, wherein the score calculation process of the positive indicator and the negative indicator is specifically expressed as follows: Positive indicator scoring: ; Reverse indicator scoring: ; Among them, the Represents various evaluation index parameters, Indicates the qualified values ​​corresponding to various indicator parameters, To represent the expected value corresponding to each index parameter, the index parameters include semi-hole rate, blasthole utilization rate, excavation footage, large block rate, throwing distance and tunnel over-excavation and under-excavation index.

5. The method for multivariate blasting effect evaluation based on point cloud data according to claim 1, characterized in that: In S5, the specific process of weighted calculation of the blasting effect evaluation score is as follows: S501. Determine the weight of the extracted blasting effect evaluation parameters in the blasting effect evaluation model, and determine the subjective weight of each evaluation parameter in the blasting effect evaluation according to the expert evaluation method ; S502. Determine the objective weight according to the CRITIC method, and standardize each index to obtain a standardized index according to the blasting effect evaluation parameters extracted after multiple blastings; S503. Construct a decision matrix based on the extracted large block rate, throwing distance, half-hole rate, average linear overbreak of the tunnel, average linear underbreak of the tunnel, excavation footage, and blasthole utilization rate indicators ; S504. Based on the standard deviation between various indicators Calculate the correlation coefficient and covariance, and determine the amount of information contained in each indicator; S505. Determine the objective weight of each indicator based on the amount of information contained in each indicator ; and according to the minimum discriminant information principle, the objective weights of each indicator are combined to obtain the combined weight.

6. The method for multivariate blasting effect evaluation based on point cloud data as claimed in claim 5, characterized in that: In S503, the decision matrix Specifically expressed as: ； Among them, the Indicates The sample The value of the indicator, Represents the total number of samples.

7. The method for multivariate blasting effect evaluation based on point cloud data according to claim 1, characterized in that: In S504, the calculation process of the correlation coefficient is specifically expressed as follows: ； The calculation process for determining the amount of information contained in each indicator is specifically expressed as follows: ； Among them, the Indicates Indicators and The correlation coefficient between the indicators is Indicates Indicators and The covariance of the indicators, represents the standard deviation of the jth index, Represents the amount of information contained in the jth indicator.

8. The method for multivariate blasting effect evaluation based on point cloud data according to claim 1, characterized in that: In S505, the calculation process of the objective weight is specifically expressed as follows: ； The combined weight is obtained according to the minimum discriminant information principle , where the calculation process of the combined weight is specifically expressed as: ； Among them, the represents the combined weight of the jth indicator in the evaluation. Represents the subjective weight of the jth indicator in the evaluation. It represents the objective weight of the jth indicator in the evaluation.

9. The method for multivariate blasting effect evaluation based on point cloud data according to claim 1, characterized in that: In S5, the blasting effect evaluation score The calculation process is specifically expressed as follows: ； based on The blasting effect judgment standard of the value is: When the blasting effect is good, When When When the blasting effect is poor.

10. A system for evaluating multivariate blasting effects based on point cloud data, the system being implemented based on the method for evaluating multivariate blasting effects based on point cloud data as claimed in any one of claims 1 to 9, characterized in that: include: A data acquisition module, used to acquire point cloud data of the tunnel after blasting through three-dimensional laser scanning; The data pre-processing module is used to reduce the noise of the point cloud data and classify the blasting point cloud, tunnel point cloud, and face point cloud; Evaluation index extraction module is used to process point cloud data and obtain blasting effect evaluation parameters such as large block rate, throwing distance, average linear over-excavation and under-excavation of tunnels, blasting footage, and unit consumption of explosives based on actual engineering practice; The index parameter evaluation module is used to assign points to the extracted blasting effect evaluation index parameters and determine the weight of each index in the blasting effect evaluation; The blasting effect evaluation module is used to obtain a blasting effect evaluation score according to the score of each evaluation index and the weight of each index, and to obtain a blasting effect evaluation according to the obtained blasting effect evaluation score.

Images