Convolutional neural network assisted laser-induced breakdown spectroscopy for determination of calcium,magnesium, silicon and aluminum in iron ore
JIN Yue1,2, LIU Shu*2, XU Qianru2, MIN Hong2, AN Yarui*1
1. School of Materials and Chemistry,University of Shanghai for Science and Technology,Shanghai 200093,China; 2. Technical Center for Industrial Product and Raw Material Inspection and Testing of Shanghai Customs District,Shanghai 200135,China
Abstract：The quantitative analysis of calcium(Ca),magnesium(Mg),silicon(Si) and aluminum(Al) in iron ores by laser-induced breakdown spectroscopy(LIBS) is helpful for rapid evaluation of iron ore quality.However,due to the influence of laser energy fluctuation,matrix effect and spectral interference and other factors,LIBS combined with single variable quantitative analysis of Ca,Mg,Si,and Al in iron ores has the application challenges of large error and low accuracy.Multivariate analysis of the original LIBS spectra can effectively improve the quantitative performance of LIBS.In this study,a one-dimensional convolutional neural network(CNN) model based on LIBS spectra was established for the quantitative analysis of Ca (in CaO),Mg (in MgO),Si (in SiO2) and Al(in Al2O3) in iron ores.Total 628 representative samples of iron ore from 35 brands in 8 countries were collected.The determination coefficient (R2) and root mean square error (RMSE) were used to evaluate the model performance.The influence of normalization method of LIBS spectra of iron ore on model performance was compared,including the feature normalization,spectral normalization and internal standard normalization.The results showed that the normalization preprocessing had a minor impact on the contents of Mg and Al,while the spectral normalization was more suitable for the analysis of Ca content analysis,and the feature normalization was more suitable for the analysis of Si content.The model parameters had a great influence on the model performance.The number of convolution cores,the size of convolution cores and the batch size were optimized,respectively.The results showed that when the number of convolution cores was 24,the size was 50,and the batch size was 256,the predictive model for Si content achieved R2 and RMSE of 0.962 6 and 0.469 8%,respectively.When the number of convolution cores was 12,the size was 60,and the batch size was 256,the predictive model for Al content achieved R2 and RMSE of 0.949 4 and 0.132 4%,respectively.When the number of convolution cores was 24,the size was 60,and the batch size was 128,the predictive model for Ca content achieved R2 and RMSE of 0.967 0 and 0.077 6%,respectively.When the number of convolution cores was 12,the size was 60,and the batch size was 256,the predictive model for Mg content achieved R2 and RMSE of 0.999 2 and 0.075 3%,respectively.The constructed optimal models with partial least squares(PLS),support vector machine(SVM),random forest (RF) and variable importance-backpropagation-artificial neural network(VI-BP-ANN) were used for method comparison.The results showed that the CNN model exhibited better prediction performance with the lowest RMSE and the highest R2.It indicated that CNN-assisted LIBS was applicable for the determination of Ca,Mg,Si,and Al contents in iron ores.
金悦, 刘曙, 徐倩茹, 闵红, 安雅睿. 卷积神经网络辅助激光诱导击穿光谱测定铁矿石中钙镁硅铝[J]. 冶金分析, 2024, 44(10): 95-103.
JIN Yue, LIU Shu, XU Qianru, MIN Hong, AN Yarui. Convolutional neural network assisted laser-induced breakdown spectroscopy for determination of calcium,magnesium, silicon and aluminum in iron ore. , 2024, 44(10): 95-103.
[1] Wang P,Li N,Yan C H,et al.Rapid quantitative analysis of the acidity of iron ore by the laser-induced breakdown spectroscopy(LIBS) technique coupled with variable importance measures-random forests(VIM-RF)[J].Analytical Methods,2019,11(27):3419-3428. [2] 施月循,刘宏娟.高炉增加高铝铁矿冶炼的可行性探讨[J].上海金属,2001,23(1):7-10,22. SHI Yuexun,LIU Hongjuan.Discussion on the possibility of increasing the amount of aluminous iron ore for ironmaking in BF[J].Shanghai Metals,2001,23(1):7-10,22. [3] 袁祥奕,刘牡丹.高铝铁矿资源综合利用现状[J].现代矿业,2017,33(2):110-113. YUAN Xiangyi,LIU Mudan.Comprehensive utilization of high alumina iron ore resources[J].Modern Mining,2017,33(2):110-113. [4] 中华人民共和国国家质量监督检验检疫总局,中国国家标准化管理委员会.GB/T 6730.14—2017铁矿石 钙含量的测定 火焰原子吸收光谱法[S].北京:中国标准出版社,2017. [5] 中华人民共和国国家质量监督检验检疫总局,中国国家标准化管理委员会.GB/T 6730.62—2005 铁矿石 钙、硅、镁、钛、磷、锰、铝和钡含量的测定 波长色散X射线荧光光谱法[S].北京:中国标准出版社,2005. [6] 中华人民共和国国家质量监督检验检疫总局,中国国家标准化管理委员会.GB/T 6730.63—2006铁矿石 铝、钙、镁、锰、磷、硅和钛含量的测定 电感耦合等离子体发射光谱法[S].北京:中国标准出版社,2006. [7] 赵文雅,闵红,刘曙,等.人工神经网络在激光诱导击穿光谱数据分析中的应用进展[J].光谱学与光谱分析,2021,41(7):1998-2004. ZHAO Wenya,MIN Hong,LIU Shu,et al.Application progress of artificial neural network in laser-induced breakdown spectral data analysis[J].Spectroscopy and Spectral Analysis,2021,41(7):1998-2004. [8] Cao X Q,Zhang L,Wu Z C,et al.Quantitative analysis modeling for the ChemCam spectral data based on laser-induced breakdown spectroscopy using convolutional neural network [J].Plasma Science & Technology,2020,22(11):115502. [9] Su P,Liu S,Min H,et al.Accuracy improvement on quantitative analysis of the total iron content in branded iron ores by laser-induced breakdown spectroscopy combined with the double back propagation artificial neural network[J].Analytical Methods,2022,14(4):427-437. [10] Fabre C.Advances in laser-induced breakdown spectroscopy analysis for geology:A critical review[J].Spectrochimica Acta Part B:Atomic Spectroscopy,2020,166:105799. [11] Coffey P,Smith N,Lennox B,et al.Robotic arm material characterization using LIBS and Raman in a nuclear hot cell decommissioning environment[J].Journal of Hazardous Materials,2021,412:125193. [12] 杨雅雯,严承琳,徐鼎,等.激光诱导击穿光谱检测铁矿石应用进展[J].冶金分析,2020,40(12):14-20. YANG Yawen,YAN Chenglin,XU Ding,et al.Progress in the detection of iron ore by laser-induced breakdown spectroscopy[J].Metallurgical Analysis,2020,40(12):14-20. [13] 高源,孙兰香,李翔宇,等.基于LIBS在线分析烧结矿混合料成分及校正水分影响[J].中国激光,2023,50(19):198-206. GAO Yuan,SUN Lanxiang,LI Xiangyu,et al.On-line analysis of sinter mixture composition and correction of moisture influence based on laser-induced breakdown spectroscopy[J].Chinese Journal of Lasers,2023,50(19):198-206. [14] Su P,Wu X H,Li C,et al.A versatile method for quantitative analysis of total iron content in iron ore using laser-induced breakdown spectroscopy[J].Applied Spectroscopy,2023,77(2):140-150. [15] Wang Z,Afgan M S,Gu W L,et al.Recent advances in laser-induced breakdown spectroscopy quantification:From fundamental understanding to data processing[J].Trac-Trends in Analytical Chemistry,2021,143:116385. [16] Chen S,Yang Y T,Wang X,et al.Quantitative analysis method of multi-elements for logging cuttings based on LIBS technology[J].Acta Photonica Sinica,2023,52(3):0352112. [17] 谢远明,孙兰香,袁德成,等.基于互信息特征筛选偏最小二乘的激光诱导击穿光谱铁矿浆定量分析[J].冶金分析,2022,42(1):18-24. XIE Yuanming,SUN Lanxiang,YUAN Decheng,et al.Quantitative analysis of iron slurry based on laser induced breakdown spectroscopy combined with mutual infoemation feature selection partial least squares method[J].Metallurgical Analysis,2022,42(1):18-24. [18] 刘曙,金悦,苏飘,等.变量重要性-反向传播人工神经网络辅助激光诱导击穿光谱测定铁矿石中硅、铝、钙和镁含量[J].光谱学与光谱分析,2023,43(10):3132-3142. LIU Shu,JIN Yue,SU Piao,et al.Determination of calcium,magnesium,aluminum,and silicon content in iron ore using laser-induced breakdown spectroscopy assisted by variable importance-back propagation artificial networks[J].Spectroscopy and Spectral Analysis,2023,43(10):3132-3142. [19] Xu X B,Ma F,Zhou J M,et al.Applying convolutional neural networks(CNN) for end-to-end soil analysis based on laser-induced breakdown spectroscopy(LIBS) with less spectral preprocessing[J].Computers and Electronics in Agriculture,2022,199:107171. [20] Li Y J,Ma B X,Li C,et al.Accurate prediction of soluble solid content in dried Hami jujube using SWIR hyperspectral imaging with comparative analysis of models[J].Computers and Electronics in Agriculture,2022,193:106655. [21] 黄卓敏,闵红,严承琳,等.卷积神经网络辅助激光诱导击穿光谱鉴别37类铁矿石[J].中国无机分析化学,2023,13(9):975-985. HUANG Zhuomin,MIN Hong,YAN Chenglin,et al.Convolutional network-assisted laser-induced breakdown spectroscopy for indentification of 37 types of iron ores[J].Chinese Journal of Inorganic Analytical Chemistry,2023,13(9):975-985 [22] Zhao W Y,Li C,Yan C L,et al.Interpretable deep learning-assisted laser-induced breakdown spectroscopy for brand classification of iron ores[J].Analytica Chimica Acta,2021,1166:338574. [23] 刘浩东,姜喜全,牛昊,等.基于激光拉曼光谱归一化法的乙醇定量分析研究[J].光谱学与光谱分析,2023,43(12):3820-3825. LIU Haodong,JIANG Xiquan,NIU Hao,et al.Quantitative analysis of ethanol based on laser Raman spectroscopy normalization method[J].Spectroscopy and Spectral Analysis,2023,43(12):3820-3825.
