基于微焦级高重频激光诱导击穿光谱的铜合金分类方法研究  

Classification of Copper Alloys Based on Microjoule High Repetition Laser-Induced Breakdown Spectra

作　　者：曲东明 张子怡 梁俊轩 廖海文 杨光[1] QU Dong-ming;ZHANG Zi-yi;LIANG Jun-xuan;LIAO Hai-wen;YANG Guang(College of Instrumentation and Electrical Engineering,Jilin University,Changchun 130026,China)

机构地区：[1]吉林大学仪器科学与电气工程学院,吉林长春130026

出　　处：《光谱学与光谱分析》2024年第11期3222-3227,共6页Spectroscopy and Spectral Analysis

基　　金：吉林省教育厅科学技术研究项目(JJKH20220993KJ);国家自然科学基金面上项目(62275099)资助。

摘　　要：针对废弃铜合金回收分类的工业应用场景,采用基于微焦级高重频激光诱导击穿光谱技术(MH-LIBS)结合人工神经网络(ANN)和支持向量机(SVM)两种机器学习算法,分别对定点模式和运动模式下采集的7种铜合金样品(H59、H62、H70、H85、H96、HPb59-1、HPb62)进行分类识别。结果显示,ANN和SVM对于定点模式下采集的铜合金能够实现100%精确分类,而对于运动模式下采集铜合金的分类精度分别为100%和99.86%。由此可见,微焦级高重频激光诱导击穿光谱系统结合机器学习算法能够实现铜合金的精细分类,适合应用于废弃铜合金的现场快速分析。For the industrial application scenario of waste copper alloy recycling and classification,two machine learning algorithms based on microjoule high-frequency laser-induced breakdown spectroscopy(MH-LIBS)combined with artificial neural network(ANN)and support vector machine(SVM)are used.Seven copper alloy samples(H59,H62,H70,H85,H96,HPb59-1,HPb62)collected in point and motion modes were classified and recognized,respectively.The results show that ANN and SVM can achieve 100%accuracy in classifying the copper alloys collected in point mode.The classification accuracy for the copper alloys collected in motion mode is 100%and 99.86%,respectively.It can be seen that the microfocus high-frequency laser-induced breakdown spectroscopy system combined with machine learning algorithms can realize the fine classification of copper alloys,which is suitable for the rapid analysis of waste copper alloys on site.

关 键 词：激光诱导击穿光谱 人工神经网络 支持向量机 铜合金 

分 类 号：O657.38[理学—分析化学]