不同型号便携式光谱仪间木质素近红外光谱分析模型传递研究
|
摘要
应用便携式近红外光谱仪快速检测制浆材时,如果能实现同系列不同型号仪器之间分析模型共享,将极大降低仪器建模和维护成本。为实现混合木材木质素含量的近红外分析模型从1台主机向2台不同型号从机的模型传递,收集了5种常见制浆材的82个原木样品,经粉碎预处理后分别在3台便携式光谱仪上采集其近红外光谱信号,采用差谱、光谱的平均差异和光谱间的欧氏距离等方法,定量表征了仪器间的信号差异,分析并讨论了差异产生的原因。利用偏最小二乘回归建立了样品主机近红外光谱与木质素含量的关联模型,再分别采用斜率截距、直接校正和典型相关分析算法进行主机与两台从机间的模型传递,比较了模型传递前后预测精度。结果表明,便携式光谱仪间的差异多为非线性,且不同型号从机光谱仪间差异更为复杂。尽管主机向同型号的从机模型传递效果更优,但经直接校正算法和典型相关分析算法传递后两台不同型号从机预测相关系数均大于0.98、预测相对标准偏差均大于3、预测标准偏差均小于1.1%,可实现木材木质素含量的近红外光谱分析模型在3台便携式光谱仪间的传递。该研究结果对于不同型号便携式光谱仪分析模型共享具有重要意义。
Portable near infrared spectrometer is widely applied to determine the compositions of pulping materials. The cost of modeling and maintenance for the instrument will be greatly reduced if the analytical models can be shared among different types of instruments within the same series. The transfer of near-infrared calibration model for lignin determination from the main machine to other two different types of portable sub-unit spectrometers( slave) was investigated in this study. 82 samples from 5 common wood pulps were collected and prepared by grinding pretreatment and their near-infrared spectrum signals were obtained via the three portable spectrometers. Three different methods,namely,difference spectra,average diversity of spectra and Euclidean distance between spectra were employed to quantitatively characterize the difference of spectrum signals between the three spectrometers. The reasons causing the differences between spectra from three portable spectrometers were also discussed. The near-infrared calibration model of lignin content in pulp samples was established via the master machine by partial least square regression. Then the established model was transferred from the master machine to the two slaves by algorithm of slope/bias( S/B),direct standardization( DS) and canonical correlation analysis( CCA). The accuracy of the model was evaluated by comparing the predicted value to that of the original model. The results indicated that the differences between portable spectrometers were mostly non-liner. And the difference between two types of slave spectrometers are more complicated. Although the transfer of calibration model within same type of instruments has the better performance,the prediction correlation coefficients of two different types of slaves by DS and CCA was up to 0.98,the ratio of performance to standard deviate was more than 3,and the root mean square error of prediction less than 1.1%. Therefore,the calibration model transfer between three portable near-infrared spectrometers could be achieved by both DS and CCA algorithm. This study could provide useful information for the near-infrared calibration model transfer among different types of portable spectrometers.
Portable near infrared spectrometer is widely applied to determine the compositions of pulping materials. The cost of modeling and maintenance for the instrument will be greatly reduced if the analytical models can be shared among different types of instruments within the same series. The transfer of near-infrared calibration model for lignin determination from the main machine to other two different types of portable sub-unit spectrometers( slave) was investigated in this study. 82 samples from 5 common wood pulps were collected and prepared by grinding pretreatment and their near-infrared spectrum signals were obtained via the three portable spectrometers. Three different methods,namely,difference spectra,average diversity of spectra and Euclidean distance between spectra were employed to quantitatively characterize the difference of spectrum signals between the three spectrometers. The reasons causing the differences between spectra from three portable spectrometers were also discussed. The near-infrared calibration model of lignin content in pulp samples was established via the master machine by partial least square regression. Then the established model was transferred from the master machine to the two slaves by algorithm of slope/bias( S/B),direct standardization( DS) and canonical correlation analysis( CCA). The accuracy of the model was evaluated by comparing the predicted value to that of the original model. The results indicated that the differences between portable spectrometers were mostly non-liner. And the difference between two types of slave spectrometers are more complicated. Although the transfer of calibration model within same type of instruments has the better performance,the prediction correlation coefficients of two different types of slaves by DS and CCA was up to 0.98,the ratio of performance to standard deviate was more than 3,and the root mean square error of prediction less than 1.1%. Therefore,the calibration model transfer between three portable near-infrared spectrometers could be achieved by both DS and CCA algorithm. This study could provide useful information for the near-infrared calibration model transfer among different types of portable spectrometers.
引文
[1]文甲龙,陈天影,孙润仓.生物质木质素分离和结构研究方法进展[J].林业工程学报,2017,2(5):76-84. DOI:10.13360/j.issn.2096-1359.2017.05.014.WEN J L, CHEN T Y, SUN R C. Research progress on separation and structural analysis of lignin in lignocellulosic biomass[J]. Journal of Forestry Engineering,2017,2(5):76-84.
[2]贺文明,薛崇昀,聂怡,等.近红外光谱法快速测定木材纤维素、戊聚糖和木质素含量的研究[J].中国造纸学报,2010,25(3):9-12.HE W M,XUE C Y,NIE Y,et al. Rapid prediction of wood cellulose,pentosan and klason lignin contents using near infrared spectroscopy[J]. Transactions of China Pulp and Paper,2010,25(3):9-12.
[3]褚小立.化学计量学方法与分子光谱分析技术[M].北京:化学工业出版社,2011.CHU X L. Molecular spectroscopy analytical technology combined with chemometrics and its applications[M]. Beijing:Chemical Industry Press,2011.
[4]张进,蔡文生,邵学广.近红外光谱模型转移新算法[J].化学进展,2017,29(8):902-910. DOI:10.7536/PC170416.ZHANG J,CAI W S,SHAO X G. New algorithms for calibration transfer in near infrared spectroscopy[J]. Progress in Chemistry,2017,29(8):902-910.
[5]杨辉华,张晓凤,樊永显,等.基于一元线性回归的近红外光谱模型传递研究[J].分析化学,2014,42(9):1229-1234.DOI:10.11895/j.issn.0253-3820.140290.YANG H H,ZHANG X F,FAN Y X,et al. Near infrared spectroscopic model transfer based on simple linear regression[J].Chinese Journal of Analytical Chemistry, 2014, 42(9):1229-1234.
[6]BIN J,LI X,FAN W,et al. Calibration transfer of near-infrared spectroscopy by canonical correlation analysis coupled with wavelet transform[J]. Analyst,2017,142(12):2229-2238.DOI:10.1039/c7an00280g.
[7]MARCHESINI G,SERVA L,GARBIN E,et al. Near-infrared calibration transfer for undried whole maize plant between laboratory and on-site spectrometers[J]. Italian Journal of Animal Science,2018, 17(1):66-72. DOI:10. 1080/1828051x.2017.1345660.
[8]杨浩,熊智新,陈通.苹果可溶性固形物含量的近红外模型传递研究[J].分析试验室,2018,37(2):163-167. DOI:10.13595/j.cnki.issn1000-0720.2018.0031.YANG H,XIONG Z X,CHEN T. Study on near-infrared calibration model transfer for soluble solid content in apple[J]. Chinese Journal of Analysis Laboratory,2018,37(2):163-167.
[9]刘贤,韩鲁佳,杨增玲,等.基于正交信号校正的秸秆青贮饲料粗蛋白近红外分析模型传递方法[J].分析化学,2012,40(4):596-601. DOI:10.3724/SP.J.1096.2012.10682.LIU X,HAN L J,YANG Z L,et al. Application of orthogonal signal correction for near infrared spectroscopic calibration transfer of crude protein of straw silage[J]. Chinese Journal of Analytical Chemistry,2012,40(4):596-601.
[10]丁柯,张月敬,沈广辉,等.蛋白饲料原料粗蛋白含量近红外光谱模型转移研究[J].光谱学与光谱分析,2016,36(5):1334-1339. DOI:10.3964/j.issn.1000-0593(2016)05-1334-06.DING K,ZHANG Y J,SHEN G H,et al. Calibration transfer of near infrared spectrometric models for crude protein of protein feed materials[J]. Spectroscopy and Spectral Analysis,2016,36(5):1334-1339.
[11]李天瑞,刘翠玲,位丽娜,等.斜率截距校正算法在食用油酸值和过氧化值上的近红外光谱模型转移的研究[J].中国粮油学报,2018,33(1):118-124. DOI:10.3969/j.issn.1003-0174.2018.01.021.LI T R,LIU C L,WEI L N,et al. Near-infrared spectral model transfer of acid value and peroxide value of edible oil by slope intercept correction algorithm[J]. Journal of the Chinese Cereals and Oils Association,2018,33(1):118-124.
[12]SWIERENGA H,HAANSTRA W G,DE WEIJER A P,et al.Comparison of two different approaches toward model transferability in NIR spectroscopy[J]. Applied Spectroscopy,1998,52(1):7-16. DOI:10.1366/0003702981942528.
[13]ZHENG K Y,FAN W,WU T,et al. Application of canonical correlation analysis(CCA)to correct NIR spectra measured on different instruments and in different measurement dates[J].Computers and Applied Chemistry,2013,30(3):246-250.DOI:10.11719/com.app.chem20130305.
[14]张小超,吴静珠,徐云.近红外光谱分析技术及其在现代农业中的应用[M].北京:电子工业出版社,2012.
[15]MALLEY D F, MCCLURE C, MARTIN P D, et al.Compositional analysis of cattle manure during composting using a field-portable near-infrared spectrometer[J]. Communications in Soil Science and Plant Analysis,2005,36(4/5/6):455-475.DOI:10.1081/css-200043187.
[16]GALVAO R,ARAUJO M,JOSE G,et al. A method for calibration and validation subset partitioning[J]. Talanta,2005,67(4):736-740. DOI:10.1016/j.talanta.2005.03.025.
[17]潘冉冉,骆一凡,王昌,等.高光谱成像的油菜和杂草分类方法[J].光谱学与光谱分析,2017,37(11):3567-3572. DOI:10.3964/j.issn.1000-0593(2017)11-3567-06.PAN R R,LUO Y F,WANG C,et al. Classifications of oilseed rape and weeds based on hyperspectral imaging[J]. Spectroscopy and Spectral Analysis,2017,37(11):3567-3572.
[18]熊智新,武中臣,陈朝霞,等.利用多元散射校正提高油菜籽脂肪酸近红外分析精度[J].光谱实验室,2007,24(5):953-958. DOI:10.3969/j.issn.1004-8138.2007.05.047.XIONG Z X,WU Z C,CHEN Z X,et al. Application of MSC to the improvement of analyzing accuracy of fatty acid in rapeseed by near-infrared spectroscopy[J]. Chinese Journal of Spectroscopy Laboratory,2007,24(5):953-958.
[19]胡昌勤,冯艳春.近红外光谱法快速分析药品[M].北京:化学工业出版社,2010.
[2]贺文明,薛崇昀,聂怡,等.近红外光谱法快速测定木材纤维素、戊聚糖和木质素含量的研究[J].中国造纸学报,2010,25(3):9-12.HE W M,XUE C Y,NIE Y,et al. Rapid prediction of wood cellulose,pentosan and klason lignin contents using near infrared spectroscopy[J]. Transactions of China Pulp and Paper,2010,25(3):9-12.
[3]褚小立.化学计量学方法与分子光谱分析技术[M].北京:化学工业出版社,2011.CHU X L. Molecular spectroscopy analytical technology combined with chemometrics and its applications[M]. Beijing:Chemical Industry Press,2011.
[4]张进,蔡文生,邵学广.近红外光谱模型转移新算法[J].化学进展,2017,29(8):902-910. DOI:10.7536/PC170416.ZHANG J,CAI W S,SHAO X G. New algorithms for calibration transfer in near infrared spectroscopy[J]. Progress in Chemistry,2017,29(8):902-910.
[5]杨辉华,张晓凤,樊永显,等.基于一元线性回归的近红外光谱模型传递研究[J].分析化学,2014,42(9):1229-1234.DOI:10.11895/j.issn.0253-3820.140290.YANG H H,ZHANG X F,FAN Y X,et al. Near infrared spectroscopic model transfer based on simple linear regression[J].Chinese Journal of Analytical Chemistry, 2014, 42(9):1229-1234.
[6]BIN J,LI X,FAN W,et al. Calibration transfer of near-infrared spectroscopy by canonical correlation analysis coupled with wavelet transform[J]. Analyst,2017,142(12):2229-2238.DOI:10.1039/c7an00280g.
[7]MARCHESINI G,SERVA L,GARBIN E,et al. Near-infrared calibration transfer for undried whole maize plant between laboratory and on-site spectrometers[J]. Italian Journal of Animal Science,2018, 17(1):66-72. DOI:10. 1080/1828051x.2017.1345660.
[8]杨浩,熊智新,陈通.苹果可溶性固形物含量的近红外模型传递研究[J].分析试验室,2018,37(2):163-167. DOI:10.13595/j.cnki.issn1000-0720.2018.0031.YANG H,XIONG Z X,CHEN T. Study on near-infrared calibration model transfer for soluble solid content in apple[J]. Chinese Journal of Analysis Laboratory,2018,37(2):163-167.
[9]刘贤,韩鲁佳,杨增玲,等.基于正交信号校正的秸秆青贮饲料粗蛋白近红外分析模型传递方法[J].分析化学,2012,40(4):596-601. DOI:10.3724/SP.J.1096.2012.10682.LIU X,HAN L J,YANG Z L,et al. Application of orthogonal signal correction for near infrared spectroscopic calibration transfer of crude protein of straw silage[J]. Chinese Journal of Analytical Chemistry,2012,40(4):596-601.
[10]丁柯,张月敬,沈广辉,等.蛋白饲料原料粗蛋白含量近红外光谱模型转移研究[J].光谱学与光谱分析,2016,36(5):1334-1339. DOI:10.3964/j.issn.1000-0593(2016)05-1334-06.DING K,ZHANG Y J,SHEN G H,et al. Calibration transfer of near infrared spectrometric models for crude protein of protein feed materials[J]. Spectroscopy and Spectral Analysis,2016,36(5):1334-1339.
[11]李天瑞,刘翠玲,位丽娜,等.斜率截距校正算法在食用油酸值和过氧化值上的近红外光谱模型转移的研究[J].中国粮油学报,2018,33(1):118-124. DOI:10.3969/j.issn.1003-0174.2018.01.021.LI T R,LIU C L,WEI L N,et al. Near-infrared spectral model transfer of acid value and peroxide value of edible oil by slope intercept correction algorithm[J]. Journal of the Chinese Cereals and Oils Association,2018,33(1):118-124.
[12]SWIERENGA H,HAANSTRA W G,DE WEIJER A P,et al.Comparison of two different approaches toward model transferability in NIR spectroscopy[J]. Applied Spectroscopy,1998,52(1):7-16. DOI:10.1366/0003702981942528.
[13]ZHENG K Y,FAN W,WU T,et al. Application of canonical correlation analysis(CCA)to correct NIR spectra measured on different instruments and in different measurement dates[J].Computers and Applied Chemistry,2013,30(3):246-250.DOI:10.11719/com.app.chem20130305.
[14]张小超,吴静珠,徐云.近红外光谱分析技术及其在现代农业中的应用[M].北京:电子工业出版社,2012.
[15]MALLEY D F, MCCLURE C, MARTIN P D, et al.Compositional analysis of cattle manure during composting using a field-portable near-infrared spectrometer[J]. Communications in Soil Science and Plant Analysis,2005,36(4/5/6):455-475.DOI:10.1081/css-200043187.
[16]GALVAO R,ARAUJO M,JOSE G,et al. A method for calibration and validation subset partitioning[J]. Talanta,2005,67(4):736-740. DOI:10.1016/j.talanta.2005.03.025.
[17]潘冉冉,骆一凡,王昌,等.高光谱成像的油菜和杂草分类方法[J].光谱学与光谱分析,2017,37(11):3567-3572. DOI:10.3964/j.issn.1000-0593(2017)11-3567-06.PAN R R,LUO Y F,WANG C,et al. Classifications of oilseed rape and weeds based on hyperspectral imaging[J]. Spectroscopy and Spectral Analysis,2017,37(11):3567-3572.
[18]熊智新,武中臣,陈朝霞,等.利用多元散射校正提高油菜籽脂肪酸近红外分析精度[J].光谱实验室,2007,24(5):953-958. DOI:10.3969/j.issn.1004-8138.2007.05.047.XIONG Z X,WU Z C,CHEN Z X,et al. Application of MSC to the improvement of analyzing accuracy of fatty acid in rapeseed by near-infrared spectroscopy[J]. Chinese Journal of Spectroscopy Laboratory,2007,24(5):953-958.
[19]胡昌勤,冯艳春.近红外光谱法快速分析药品[M].北京:化学工业出版社,2010.