【作者】 闫春华； 王占梅； 阮方琦； 张天龙； 李华；

【Author】 Chunhua Yan;Zhanmei Wang;Fangqi Ruan;Tianlong Zhang;Hua Li;Institution of Analytical Science,College of Chemistry & Materials Science,Northwest University;College of Chemistry and Chemical Engineering,Xi’an Shiyou University;

【机构】 西北大学化学与材料科学学院分析科学研究所； 西安石油大学化学与化工学院；

【摘要】 利用激光诱导击穿光谱(LIBS)技术结合N最近邻(N3)算法对酸性铁矿石、半自熔性铁矿石、自熔性铁矿石、碱性铁矿石四种铁矿石样品进行识别。在最优化的模型参数下,以归一化的LIBS光谱(400-600 nm)作为输入变量分别构建了N3和KNN模型用于铁矿石的识别,并比较了两种方法对未知铁矿石样品分类预测的准确性。结果表明,N3模型的分类精度为100%,其预测能力优于KNN模型。LIBS技术结合N3算法是一种高效的铁矿石样品分类方法,并且有助于实现铁矿石的快速、现场和在线分析。更多还原

【Abstract】 LIBS coupled with N3 method was developed for classification and identification of four types of iron ore(acid iron ore,seiili-self fluxing iron ore,self-fluxing iron ore,alkaline iron ore).The parameters included spectral pretreatment methods,variables selection and the model parameter a were optimized at the same time by 5-fold CV.The region of 400-600 nm with normalized by maximum integrated intensity were used to construct the N3 and KNN models.The N3 and KNN models were evaluated and applied to discriminate iron ore.The classification accuracy is 100%for N3 model,which shows a better predictive capabilities than the KNN model.Therefore,LIBS technique combined with N3 would be a promising method for real-time online,rapid analysis of iron ore.更多还原