面向智能光电分选的矿石粒度在线识别  

Online Recognition of Ore Particle Size for Intelligent Photoelectric Sorting Technology

作　　者：陈串串 刘祖鹏 严昊 徐阳 肖罡 CHEN Chuanchuan;LIU Zupeng;YAN Hao;XU Yang;XIAO Gang(Jiangxi Copper Technology Institute Co.,Ltd.,Nanchang 330096,China;School of Software,Tsinghua Uni-versity,Beijing 10084,China)

机构地区：[1]江西铜业技术研究院有限公司,江西南昌330096 [2]清华大学软件学院,北京100084

出　　处：《铜业工程》2025年第2期38-47,共10页Copper Engineering

基　　金：国家重点研发计划项目(2023YFC3904200),江西省自然科学基金项目(20224ACB218002),江西省重大科技研发专项项目(20232ACE01010)资助。

摘　　要：在有色金属矿山领域,智能光电分选技术是提升资源利用效率、实现绿色开发的关键技术之一,但由于矿石厚度与平均原子序数的耦合效应,该技术在大通量多粒级场景下分选决策的准确性会下降。利用非接触感知手段在线识别运动矿石的粒度,有望实现X射线透射图中矿石厚度信息的解耦,从而突破大通量多粒级矿石智能分选技术障碍。本文针对矿石粒度在线识别的应用需求,综述了基于二维图像、三维点云以及多模态数据融合技术等潜在方法,分析了现有技术的优缺点和局限性,提出了适用于矿石光电分选系统的矿石粒度在线识别的技术路线,并展望了其发展方向,为有色金属光电分选技术的创新与应用提供了理论支撑与技术路线参考。In the field of non-ferrous metal mining,high-throughput multi-grade intelligent photoelectric sorting is crucial for improving resource utilization and achieving sustainable development.However,X-ray transmission-based sorting systems face challenges due to the coupling effect of ore thickness and average atomic number,which reduces decision-making accuracy.By utilizing non-contact sensing methods to online recognize the particle size of moving ores,it is expected to decouple ore thickness information from X-ray transmission images,which serves as a critical breakthrough for intelligent sorting technology in large-scale multi-grade ore processing.Aiming at the application demands of online identification of ore particle size,this paper reviews potential feasible technical methods based on two-dimensional images,three-dimensional point clouds,and multimodal data fusion technology.Futhermore,the advantages and disadvantages of existing online ore particle size identification technologies and their respective limitations are analyzed.Finally,the technical development direction and technical route for online identification of ore particle size in ore photoelectric sorting systems are proposed.This study provides theoretical insights and technical pathways for innovation in photoelectric ore sorting technologies.

关 键 词：智能光电分选 粒度在线识别 矿石图像分割 矿石点云 多模态融合 

分 类 号：TD95[矿业工程—选矿]