基于神经网络集成-X射线荧光光谱法的铁矿石中全铁含量测定
|
摘要
为了探索人工智能在铁矿石品质快速检验中的应用,研究了机器学习算法与化学计量学和X射线荧光光谱仪(XRF)相结合快速测定铁矿石中全铁含量的方法。收集来自于不同产地的,主要物相为赤铁矿、褐铁矿、磁铁矿、针铁矿和多物相混和结构的铁矿石样品共1 098个作为样本集。采用X射线荧光光谱仪对铁矿石样品熔片进行扫描,扫描后的光谱图提取数据点后作为神经网络的输入,以全铁含量作为输出结果。然后依据X射线衍射(XRD)得到的物相结构优化自组织(SOM)网络,并对全部样本的XRF图谱进行分类,对分类后的每一个子集分别采用反向传播(BP)和径向基函数(RBF)网络建立回归子模型,对各子模型的预测结果进行整合,最终建立基于集成神经网络和X射线荧光光谱法的铁矿石中全铁含量预测模型。方法模型建立后,不需要额外标准物质建立校准曲线,能够实现对未知样品的分类和输出全铁含量结果。
In order to explore the application of artificial intelligence in rapid quality test of iron ore,the rapid determination method of total iron content in iron ore was studied by machine learning algorithm,chemometrics and X-ray fluorescence spectrometry(XRF).Total 1 098 iron ore samples from different regions(the main phases were hematite,limonite,magnetite,goethite and multi-phase mixture structure)were used as the sample set.The fuse pieces of iron ore samples were scanned by X-ray fluorescence spectrometer.The data points extracted from the spectra were used as the input of neural network,while the content of total ion was used as the output result.Then,the self-organization mapping(SOM)networks were optimized according to the phase structure obtained by X-ray diffraction(XRD).The XRF spectra of all samples were classified.The regression submodel of each subset was established based on back propagation(BP)and radial basis function(RBF)network.The prediction results of each submodel were integrated.Finally the prediction model of total iron content in iron ore based on neural network integration and XRF was established.The model experience certificate test results indicated that the accuracy was up to89.1%.After model establishment,the additional standard substances were not required to establish calibration curve for the classification and total iron content output of unknown samples.
In order to explore the application of artificial intelligence in rapid quality test of iron ore,the rapid determination method of total iron content in iron ore was studied by machine learning algorithm,chemometrics and X-ray fluorescence spectrometry(XRF).Total 1 098 iron ore samples from different regions(the main phases were hematite,limonite,magnetite,goethite and multi-phase mixture structure)were used as the sample set.The fuse pieces of iron ore samples were scanned by X-ray fluorescence spectrometer.The data points extracted from the spectra were used as the input of neural network,while the content of total ion was used as the output result.Then,the self-organization mapping(SOM)networks were optimized according to the phase structure obtained by X-ray diffraction(XRD).The XRF spectra of all samples were classified.The regression submodel of each subset was established based on back propagation(BP)and radial basis function(RBF)network.The prediction results of each submodel were integrated.Finally the prediction model of total iron content in iron ore based on neural network integration and XRF was established.The model experience certificate test results indicated that the accuracy was up to89.1%.After model establishment,the additional standard substances were not required to establish calibration curve for the classification and total iron content output of unknown samples.
引文
[1]赵树宝.三氯化钛还原-高锰酸钾无汞滴定法测定铁矿石中全铁量[J].冶金分析,2010,30(1):77-80.ZHAO Shu-bao.Mercury-free titration of total iron in iron ore with potassium permanganate after titanium tricolored reduction[J].Metallurgical Analysis,2010,30(1):77-80.
[2]廖海平,应海松,付冉冉.电位滴定法测定铁矿石中全铁的微波溶样法[J].冶金分析,2007,27(11):54-58.LIAO Hai-ping,YING Hai-song,FU Ran-ran.Microwave digestion of iron ore samples for determination of total iron by potentiometric titration[J].Metallurgical Analysis,2007,27(11):54-58.
[3]张莉娟,徐铁民,李小莉,等.X射线荧光光谱法测定富含硫砷钒铁矿石中的主次量元素[J].岩矿测试,2011,30(6):772-776.ZHANG Li-juan,XU Tie-min,LI Xiao-li,et al.Quantification of major and minor components in iron ores with sulfur,arsenic and vanadium by X-ray fluorescence spectrometry[J].Rock and Mineral Analysis,2011,30(6):772-776.
[4]杨峰,杨秀玖,刘伟洪,等.熔融制样-波长色散X射线荧光光谱法测定铁矿石中12种主次成分[J].分析实验室,2015,32(3):351-355.YANG Feng,YANG Xiu-jiu,LIU Wei-hong,et al.Analysis of 12major and trace components in iron ore by the wavelength dispersive X-ray fluorescence spectrometry with fusion sample preparation[J].Chinese Journal of Analysis Laboratory,2015,32(3):351-355.
[5]廖海平,付冉冉,任春生,等.X射线荧光光谱法测定铁矿石中全铁及18个次量成分[J].冶金分析,2011,31(5):36-40.LIAO Hai-ping,FU Ran-ran,REN Chun-sheng,et al.Determination of total iron and eighteen minor components in iron ore by X-ray fluorescence spectrometry[J].Metallurgical Analysis,2011,31(5):36-40.
[6]李小莉.X射线荧光光谱法测定铁矿中铁等多种元素[J].岩矿测试,2008,27(3):229-231.LI Xiao-li.Determination of multi-elements in iron ores by X-ray fluorescence spectrometry[J].Rock and Mineral Analysis,2008,27(3):229-231.
[7]KATAOKA Y,HOMMA H,KOHNO H.X射线荧光分析法测定铁矿石中全铁[J].冶金分析,2011,31(7):18-21.KATAOKA Y,HOMMA H,KOHNO H.Determination of total iron in iron ores using X-ray fluorescence analysis[J].Metallurgical Analysis,2011,31(7):18-21.
[8]廖海平,付冉冉,任春生,等.熔融制样-X射线荧光光谱法测定硫铁矿中主次成分[J].冶金分析,2014,34(12):29-32.LIAO Hai-ping,FU Ran-ran,REN Chun-sheng,et al.Determination of major and minor components in pyrite by X-ray fluorescence spectrometry with fusion sample preparation[J].Metallurgical Analysis,2014,34(12):29-32.
[9]曾江萍,吴磊,李小莉,等.较低稀释比熔融制样X射线荧光光谱法分析铬铁矿[J].岩矿测试,2013,32(6):915-919.ZENG Jiang-ping,WU Lei,LI Xiao-li,et al.Determination of chromite by X-ray fluorescence spectrometry with sample preparation of a lower-dilution fusion[J].Rock and Mineral Analysis,2013,32(6):915-919.
[10]应海松,廖海平,任春生.铁矿石检测技术及其新进展[J].金属矿山,2007,37(7):10-12,32.YING Hai-song,LIAO Hai-ping,REN Chun-sheng.Testing techniques for iron ore and their new advances[J].Metal Mine,2007,37(7):10-12,32.
[11]王谦,应晓浒,张建波.X射线荧光光谱分析样品烧增量的影响及校正[J].光谱学与光谱分析,2011,31(9):2574-2577.WANG Qian,YING Xiao-hu,ZHANG Jian-bo.The influence of the gain on ignition and correction in X-ray fluorescence spectrometry[J].Spectroscopy and Spectral Analysis,2011,31(9):2574-2577.
[12]Noemi N,Patricio G,Ronei J,et al.Multivariate calibrations for the SR-TXRF determination of trace concentrations of lead and arsenic in the presence of bromine[J].X-Ray Spectrometry,2006,35(1):79-84.
[13]Jahan B,Mohammad K,Nayereh A.Multivariate curve resolution alternating least squares in the quantitative determination of sulfur using overlapped S(Kα)-Mo(Lα)emission peaks by wavelength dispersive X-ray fluorescence spectrometry[J].X-Ray Spectrometry,2015,44(2):75-80.
[14]Jahan B,Mohammad K,Nayereh A.Using chemometric methods for overlap correction of sodium-zinc spectral lines generated by wavelength dispersive X-ray fluorescence in mineral samples[J].X-Ray Spectrometry,2014,43(3):131-137.
[15]LUO Li-qiang.An algorithm combining neural networks withfundamentalparameters[J]. X-Ray Spectrometry,2002,31(4):332-338.
[16]周志华,陈世福.神经网络集成[J].计算机学报,2002,25(1):1-8.ZHOU Zhi-hua,CHEN Shi-fu.Neural network ensemble[J].Chinese J.Computers,2002,25(1):1-8.
[17]ISO 2596-1:2006Iron ores-Determination of total ironsPart 1:Titrimetric method after tin(II)chloride reduction[S].Geneva:International Organization for Standardization,2006.
[18]中华人民共和国国家质量监督检验检疫总局.SN/T0832—1999进出口铁矿石中铁、硅、钙、锰、铝、钛、镁和磷的测定波长色散X射线荧光光谱法[S].北京:中国标准出版社,2002.
[19]雷萌,李明,徐志彬.遗传神经网络在近红外光谱煤质分析中的应用研究[J].工矿自动化,2012(2):41-44.LEI Meng,LI Ming,XU Zhi-bin.Application of genetic neural network in coal quality analysis with near-infrared spectroscopy[J].Industry and Mine Automation,2012(2):41-44.
[2]廖海平,应海松,付冉冉.电位滴定法测定铁矿石中全铁的微波溶样法[J].冶金分析,2007,27(11):54-58.LIAO Hai-ping,YING Hai-song,FU Ran-ran.Microwave digestion of iron ore samples for determination of total iron by potentiometric titration[J].Metallurgical Analysis,2007,27(11):54-58.
[3]张莉娟,徐铁民,李小莉,等.X射线荧光光谱法测定富含硫砷钒铁矿石中的主次量元素[J].岩矿测试,2011,30(6):772-776.ZHANG Li-juan,XU Tie-min,LI Xiao-li,et al.Quantification of major and minor components in iron ores with sulfur,arsenic and vanadium by X-ray fluorescence spectrometry[J].Rock and Mineral Analysis,2011,30(6):772-776.
[4]杨峰,杨秀玖,刘伟洪,等.熔融制样-波长色散X射线荧光光谱法测定铁矿石中12种主次成分[J].分析实验室,2015,32(3):351-355.YANG Feng,YANG Xiu-jiu,LIU Wei-hong,et al.Analysis of 12major and trace components in iron ore by the wavelength dispersive X-ray fluorescence spectrometry with fusion sample preparation[J].Chinese Journal of Analysis Laboratory,2015,32(3):351-355.
[5]廖海平,付冉冉,任春生,等.X射线荧光光谱法测定铁矿石中全铁及18个次量成分[J].冶金分析,2011,31(5):36-40.LIAO Hai-ping,FU Ran-ran,REN Chun-sheng,et al.Determination of total iron and eighteen minor components in iron ore by X-ray fluorescence spectrometry[J].Metallurgical Analysis,2011,31(5):36-40.
[6]李小莉.X射线荧光光谱法测定铁矿中铁等多种元素[J].岩矿测试,2008,27(3):229-231.LI Xiao-li.Determination of multi-elements in iron ores by X-ray fluorescence spectrometry[J].Rock and Mineral Analysis,2008,27(3):229-231.
[7]KATAOKA Y,HOMMA H,KOHNO H.X射线荧光分析法测定铁矿石中全铁[J].冶金分析,2011,31(7):18-21.KATAOKA Y,HOMMA H,KOHNO H.Determination of total iron in iron ores using X-ray fluorescence analysis[J].Metallurgical Analysis,2011,31(7):18-21.
[8]廖海平,付冉冉,任春生,等.熔融制样-X射线荧光光谱法测定硫铁矿中主次成分[J].冶金分析,2014,34(12):29-32.LIAO Hai-ping,FU Ran-ran,REN Chun-sheng,et al.Determination of major and minor components in pyrite by X-ray fluorescence spectrometry with fusion sample preparation[J].Metallurgical Analysis,2014,34(12):29-32.
[9]曾江萍,吴磊,李小莉,等.较低稀释比熔融制样X射线荧光光谱法分析铬铁矿[J].岩矿测试,2013,32(6):915-919.ZENG Jiang-ping,WU Lei,LI Xiao-li,et al.Determination of chromite by X-ray fluorescence spectrometry with sample preparation of a lower-dilution fusion[J].Rock and Mineral Analysis,2013,32(6):915-919.
[10]应海松,廖海平,任春生.铁矿石检测技术及其新进展[J].金属矿山,2007,37(7):10-12,32.YING Hai-song,LIAO Hai-ping,REN Chun-sheng.Testing techniques for iron ore and their new advances[J].Metal Mine,2007,37(7):10-12,32.
[11]王谦,应晓浒,张建波.X射线荧光光谱分析样品烧增量的影响及校正[J].光谱学与光谱分析,2011,31(9):2574-2577.WANG Qian,YING Xiao-hu,ZHANG Jian-bo.The influence of the gain on ignition and correction in X-ray fluorescence spectrometry[J].Spectroscopy and Spectral Analysis,2011,31(9):2574-2577.
[12]Noemi N,Patricio G,Ronei J,et al.Multivariate calibrations for the SR-TXRF determination of trace concentrations of lead and arsenic in the presence of bromine[J].X-Ray Spectrometry,2006,35(1):79-84.
[13]Jahan B,Mohammad K,Nayereh A.Multivariate curve resolution alternating least squares in the quantitative determination of sulfur using overlapped S(Kα)-Mo(Lα)emission peaks by wavelength dispersive X-ray fluorescence spectrometry[J].X-Ray Spectrometry,2015,44(2):75-80.
[14]Jahan B,Mohammad K,Nayereh A.Using chemometric methods for overlap correction of sodium-zinc spectral lines generated by wavelength dispersive X-ray fluorescence in mineral samples[J].X-Ray Spectrometry,2014,43(3):131-137.
[15]LUO Li-qiang.An algorithm combining neural networks withfundamentalparameters[J]. X-Ray Spectrometry,2002,31(4):332-338.
[16]周志华,陈世福.神经网络集成[J].计算机学报,2002,25(1):1-8.ZHOU Zhi-hua,CHEN Shi-fu.Neural network ensemble[J].Chinese J.Computers,2002,25(1):1-8.
[17]ISO 2596-1:2006Iron ores-Determination of total ironsPart 1:Titrimetric method after tin(II)chloride reduction[S].Geneva:International Organization for Standardization,2006.
[18]中华人民共和国国家质量监督检验检疫总局.SN/T0832—1999进出口铁矿石中铁、硅、钙、锰、铝、钛、镁和磷的测定波长色散X射线荧光光谱法[S].北京:中国标准出版社,2002.
[19]雷萌,李明,徐志彬.遗传神经网络在近红外光谱煤质分析中的应用研究[J].工矿自动化,2012(2):41-44.LEI Meng,LI Ming,XU Zhi-bin.Application of genetic neural network in coal quality analysis with near-infrared spectroscopy[J].Industry and Mine Automation,2012(2):41-44.