Acidity analysis of iron ore based on laser-induced breakdown spectroscopy coupled with self-absorption correction and partial least squares
DU Yao1, LI Mao-gang1, WANG Ping2, FENG Yao-zhou2, ZHANG Tian-long2, LI Hua*1,2
1. College of Chemistry and Chemical Engineering, Xi′an Shiyou University, Xi′an 710065, China; 2. Key Laboratoryof Synthetic and Natural Functional Molecular Chemistry of Ministry of Education, College of Chemistry andMaterial Science, Northwest University, Xi′an 710127, China
摘要 铁矿石冶炼过程中对其酸度进行准确把控将对铁矿石利用率及冶炼过程产生严重影响。因此,亟需一种铁矿石酸度快速准确分析方法。实验基于激光诱导击穿光谱(Laser-induced breakdown spectroscopy,LIBS)技术结合偏最小二乘回归(Partial least squares regression,PLSR)方法成功地提出了一种铁矿石酸度快速定量分析方法。首先,采集了20组铁矿石样品的LIBS光谱数据,并采用美国国家标准与技术研究院(National Institute of Standards and Technology,NIST)数据库对铁矿石的LIBS特征谱线进行标定。然后,采用内参考线自吸收修正(Internal reference for self-absorption correction,IRSAC)和5折交叉验证分别对光谱数据以及PLSR模型潜变量(Latent variables,LVs)进行优化。最后,基于优化后的光谱数据以及LVs构建了PLSR模型用于预测集铁矿石酸度的分析。结果表明,该模型具有较好的预测性能,其预测集决定系数(R2p)为0.9784,均方根误差(RMSEP)为2.916%。说明LIBS结合自吸收修正和PLSR法为铁矿石酸度的快速定量分析提供了一种可行的方法。
Abstract：Accurately control the acidity of iron ore in the smelting process will have a serious impact on the utilization rate of iron ore and the smelting process. Therefore, it is urgent to develop a rapid and accurate method for the analysis of iron ore acidity. A rapid quantitative analysis method of iron ore acidity was proposed based on laser-induced breakdown spectroscopy (LIBS) and partial least squares regression (PLSR). Firstly, LIBS spectral data of 20 iron ore samples were collected. Secondly, the LIBS characteristic peak of iron ore was calibrated based on national institute of standards and technology (NIST) database. Then, the spectra data and latent variables (LVs) of PLSR model were optimized by internal reference for self-absorption correction (IRSAC) and 5-fold cross validation (5-fold CV). Finally, a PLSR model was constructed based on the optimized spectral data and LVs to predict the acidity of iron ore of prediction set, and the obtained results showed that the PLSR model had good prediction performance with determination coefficient (R2P) of 0.9784 and the root-mean-square error (RMSEP) of 2.916%. It indicated that the research of LIBS coupled with self-absorption correction and PLSR provided a reliable method for rapid quantitative analysis of iron ore acidity.
杜瑶, 李茂刚, 王萍, 冯耀州, 张天龙, 李华. 激光诱导击穿光谱技术结合自吸收修正和偏最小二乘法的铁矿石酸度分析[J]. 冶金分析, 2020, 40(12): 105-111.
DU Yao, LI Mao-gang, WANG Ping, FENG Yao-zhou, ZHANG Tian-long, LI Hua. Acidity analysis of iron ore based on laser-induced breakdown spectroscopy coupled with self-absorption correction and partial least squares. , 2020, 40(12): 105-111.
Ding Y,Yan F,Yang G,et al.Quantitative analysis of sinters using laser-induced breakdown spectroscopy (LIBS) coupled with kernel-based extreme learning machine (K-ELM)[J].Analytical Methods,2018,10(9):1074-1079.
[2]
Grant K L,George P L,O′Neill J A.Quantitative elemental analysis of iron ore by laser-induced breakdown spectroscopy[J].Applied Spectroscopy,1991,45(4):701-705.
[3]
郑国经.铁矿石化学分析方法标准及实验室能力验证[J].冶金分析,2015,35(2):37-44.ZHENG Guo-jing.Chemical analysis method standards and proficiency testing of laboratory for iron ore[J].Metallurgical Analysis,2015,35(2):37-44.
[4]
陈贺海,荣德福,付冉冉,等.微波消解-电感耦合等离子体质谱法测定铁矿石中15个稀土元素[J].岩矿测试,2013,32(5):702-708.CHEN He-hai,RONG De-fu,FU Ran-ran,et al.Determination of fifteen rare-earth elements in iron ores using inductively coupled plasma mass spectrometry with microwave digestion[J].Rock and Mineral Analysis,2013,32(5):702-708.
[5]
钟奇秀,赵天卓,李欣,等.多元素LIBS分析的标准化交叉验证及其优化[J].光谱学与光谱分析,2020,40(2):622-627.ZHONG Qi-xiu,ZHAO Tian-zhuo,LI Xin,et al.Standardized cross-validation and its optimization for multi-element LIBS analysis[J].Spectroscopy and Spectral Analysis,2020,40(2):622-627.
[6]
Wang P,Li N,Yan C,et al.Rapid quantitative analysis of the acidity of iron ore by laser-induced breakdown spectroscopy (LIBS) technique coupled with variable importance measurement-random forest (VIM-RF)[J].Analytical Methods,2019,11(27):3419-3428.
[7]
Zeng Q,Guo L,Li X,et al.Laser-induced breakdown spectroscopy using laser pulses delivered by optical fibers for analyzing Mn and Ti elements in pig iron[J].Journal of Analytical Atomic Spectrometry,2015,30(2):403-409.
[8]
Yang Y W,Hao Y J,Zhang L L,Ren L.Application of scikit and keras libraries for the classification of iron ore data acquired by laser-induced breakdown spectroscopy (LIBS)[J].Sensors,2020,20(5):1393-1393.
[9]
Hahn D W,Omenetto N.Laser-induced breakdown spectroscopy (LIBS),part II:review of instrumental and methodological approaches to material analysis and applications to different fields[J].Applied Spectroscopy,2012,66(4):347-419.
[10]
薛彬,刘生润,杨建峰.用于火星表面生命信息探测的激光拉曼技术进展[J].深空探测学报,2019,6(5):503-512.XUE Bin,LIU Sheng-run,YANG Jian-feng.Advancements in detection of life information on mars with raman laser spectroscopy[J].Journal of Deep Space Exploration,2019,6(5):503-512.
[11]
张鹏,孙兰香,于海斌,等.Voigt函数拟合的LIBS谱线自吸收校正方法在磷矿镁元素分析中的应用[J].光谱学与光谱分析,2020,40(1):266-270.ZHANG Peng,SUN Lan-xiang,YU Hai-bin,et al.A LIBS spectral self-absorption correction method using voigt profile fitting for the application of magnesium analysis in phosphorus ore[J].Spectroscopy and Spectral Analysis,2020,40(1):266-270.
[12]
Sun L X,Yu H.Correction of self-absorption effect in calibration-free laser-induced breakdown spectroscopy by an internal reference method[J].Talanta,2009,79(2):388-395.
[13]
Wold S,Albano C,Dunn M,et al.Pattern recognition:finding and using regularities in multivariate data[J].Proceedings of Food Research and Data Analysis,1983:146-189.
[14]
Yaroshchyk P,Death D L,Spencer S J.Quantitative measurements of loss on ignition in iron ore using laser-induced breakdown spectroscopy and partial least squares regression analysis[J].Applied spectroscopy,2010,64(12):1335-1341.
