Fragmentation Calculation And Analysis For Blast Muck Piles In Open-pit Copper Mines Based On Multi-scale And Dimensional Features Of 3D Laser Point Cloud Data

With the rapid development of China’s economy,the demand for copper resources is also increasing quickly.The production efficiency of mines directly determines the supply capacity of copper resources in our country.Against the backdrop of the country’s advocacy of developing green and intelligent mines,the energy utilization efficiency and post-blast environmental impact during the mine blasting process cannot be ignored.Improving the effectiveness of open-pit mine bench blasting as much as possible has a positive effect on increasing the economic benefits of mines and reducing production costs.The fragment distribution of blast muck piles is one of the important indicators of mine blasting effectiveness,and studying the fragment distribution of blast muck piles is of great significance to mine production.Three-dimensional laser point clouds have the advantages of high precision and multiple features.This paper proposes a calculation method for the fragment of blast muck piles based on the multi-scale and dimensional features of three-dimensional laser point clouds,combined with existing segmentation algorithms for three-dimensional laser point clouds.The specific research results are as follows:(1)Point cloud segmentation of blast muck piles based on multi-scale and dimensional features: Firstly,the Euclidean segmentation algorithm was used to recognize the fragment in the point cloud of blast muck piles.However,the noise between the fragment point clouds severely affects the fragment recognition results of the Euclidean segmentation algorithm.Therefore,this paper proposes to define the local dimension of the blasting point cloud by using the multi-scale and dimensional features of the point cloud,identify the noise points through the multi-scale and dimensional features of the point cloud,and improve the clustering segmentation basis of the European segmentation algorithm.Experiments show that the recognition accuracy of the fragment in the point cloud of third blast muck piles was improved by 19.27%,and the comprehensive evaluation score of the recognition results was improved by 10.25%after the Euclidean segmentation algorithm was improved.Finally,this paper also discusses the threshold settings of the improved Euclidean algorithm and found that the optimal ore recognition accuracy was achieved when the Euclidean distance threshold of the improved Euclidean segmentation algorithm was set to 0.02 meters,and the normal vector angle threshold was set to 25 degrees.(2)Local optimization of fragment recognition results through parameter adaptive adjustment: Because the stacking form of ore in different blasting areas is different,the improved Euclidean segmentation algorithm resulted in local recognition errors after replacing the blasting data.In order to improve the data adaptability of the improved Euclidean segmentation algorithm,this paper proposes a parameter adaptive adjustment method,which automatically judges the fragment point cloud with local errors and improves the algorithm parameter threshold in a targeted manner to achieve local optimal segmentation.Finally,through the qualitative analysis of precision evaluation indicators,the feasibility of local optimization through parameter adaptive adjustment was verified.It was found that the accuracy of fragment recognition for the 11 th and17th blast muck piles increased by 7.84% and 10.28%,respectively,and the comprehensive evaluation score increased by 2.77% and 4.51%,respectively,after local optimization.(3)Fragmentation calculation and evaluation of blast muck piles: Firstly,the definition and statistical method of blast muck piles were determined.Then,based on the fragment size results identified and calculated by the improved Euclidean algorithm in this paper,the fragment size information,grading distribution information,and average size of fragment in the 3rd,11 th,and 17 th blast muck piles were respectively obtained.The fragment calculation result of the three blast muck piles was represented by the average size of the fragment: the third blast muck piles was 0.14 meters,the 11 th blast muck piles was 0.15 meters,and the 17 th blast muck piles was 0.16 meters.Finally,based on the beneficiation process and crushing equipment parameters of Dexing Copper Mine,the fragment evaluation indicators of blast muck piles were determined as the large block rate and crushing cost.Combined with the fragmentation grading distribution information of the blast muck piles,a comprehensive evaluation score was given to the 3rd,11 th,and 17 th blast muck piles.It was found that all the three blasting piles performed well.When the total score of the comprehensive evaluation was 4points,the 3rd blast muck piles scored 3.189 points,the 11 th blast muck piles scored3.034 points,and the 17 th blast muck piles scored 2.949 points.Based on the data of Dexing Copper Mine,this paper proposes the calculation method for the fragment of blast muck piles based on the multi-scale and dimensional features of three-dimensional laser point clouds to calculate and evaluate the 3rd,11 th,and 17 th blast muck piles of Dexing Copper mine.The effectiveness and feasibility of the method in this paper are verified,which provides a reference for the research in the field of blasting production of open pit copper mine.