稀土镀层永磁废料色选识别技术研究  

Study on color separation and identification technology for rare-earth permanent magnet waste

作　　者：王丽婷 陶天一 曹宏斌 孙峙[1,2,3,4] WANG LiTing;TAO TianYi;CAO HongBin;SUN Zhi(School of Chemical Engineering,University of Chinese Academy of Sciences,Beijing 100049,China;National Engineering Research Center of Green Recycling for Strategic Metal Resources,Beijing 100190,China;Chemistry&Chemical Engineering Data Center,Chinese Academy of Sciences,Beijing 100190,China;Beijing Engineering Research Centre of Process Pollution Control,Beijing 100190,China)

机构地区：[1]中国科学院大学化学工程学院,北京100049 [2]战略金属资源绿色循环利用国家工程研究中心,北京100190 [3]中国科学院化学化工科学数据中心,北京100190 [4]北京市过程污染控制工程技术研究中心,北京100190

出　　处：《中国科学：技术科学》2024年第11期2156-2166,共11页Scientia Sinica(Technologica)

基　　金：国家重点研发计划“稀土新材料”重点专项(编号:2022YFB3504300,2022YFB3504301);中国科学院化学化工科学数据中心能力建设(编号:WX145XQ07-12)资助项目。

摘　　要：稀土废料具有种类多、成分杂等显著特征,高效分类分级获取低杂质含量的中间物料是提高稀土回收效率、降低全过程碳排放的关键.本研究以最普遍的含镀层稀土永磁废料为对象,系统研究了光强、角度等因素对废料识别效率的影响,进一步提出基于多维度光强动态自主补偿的方法,开发出针对典型稀土废料的限定环境要素的阈值优化算法.研究结果表明,在光强范围82~161 lux内,模型相关系数高达R^(2)=0.97,分选效率可达100%.本研究可以为稀土废料高效分选富集和低碳循环提供新思路.Rare-earth waste exhibits significant diversity in types and complex compositions.Efficient classification and grading of rare-earthwaste produce intermediate materials with reduced impurity content,which is essential for improving recycling efficiency andlowering carbon emissions.This study focuses on the general rare-earth permanent magnet waste,NdFeB,in different coatings andinvestigates efficiency factors such as light intensity and capture angle.The study proposes a novel method based onmultidimensional light-intensity dynamic self-compensation.A threshold optimization algorithm is developed to mitigateenvironmental factors for typical rare-earth waste.The research findings demonstrate that within the light-intensity range of82–161 lux,the model achieves a high correlation coefficient of up to R^(2)=0.97,with sorting efficiency reaching 100%.This studyprovides new insights into the efficient sorting and enrichment of rare-earth waste and promotes low-carbon recycling efforts.

关 键 词：稀土永磁废料 镀层钕铁硼 机器视觉识别 循环利用 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TM273[自动化与计算机技术—计算机科学与技术]