华东理工大学 化学与分子工程学院,上海 200237
栾绍嵘,博士,高级工程师,研究方向:光谱、色谱技术及其应用,E-mail:srluan@ecust.edu.cn
扫 描 看 全 文
张站鸽,倪力军,张立国等.基于MC-UVE-VIP两步波长筛选实现近红外光谱模型的无标样传递[J].分析测试学报,2023,42(02):204-209.
ZHANG Zhan-ge,NI Li-jun,ZHANG Li-guo,et al.Standard Free Transfer of Near Infrared Spectral Model by Two-step Wavelength Screening Method Based on MC-UVE-VIP Algorithm[J].Journal of Instrumental Analysis,2023,42(02):204-209.
张站鸽,倪力军,张立国等.基于MC-UVE-VIP两步波长筛选实现近红外光谱模型的无标样传递[J].分析测试学报,2023,42(02):204-209. DOI: 10.19969/j.fxcsxb.22101501.
ZHANG Zhan-ge,NI Li-jun,ZHANG Li-guo,et al.Standard Free Transfer of Near Infrared Spectral Model by Two-step Wavelength Screening Method Based on MC-UVE-VIP Algorithm[J].Journal of Instrumental Analysis,2023,42(02):204-209. DOI: 10.19969/j.fxcsxb.22101501.
该文将蒙特卡洛-无变量信息消除(MC-UVE)算法和变量重要性投影(VIP)算法结合,挑选出重要、有信息的波长变量,建立了MC-UVE-VIP两步波长筛选方法。该法首先采用MC-UVE筛选出稳定性参数大于某一阈值(,M,threshold,)的有信息波长集合,U,UVE,,然后采用VIP算法从,U,UVE,中筛选出VIP参数大于,U,UVE,中所有波长VIP均值的波长,作为重要、有信息的波长集合,U,UVE-VIP,。基于,U,UVE-VIP,建立玉米中蛋白质含量的偏最小二乘回归(PLSR)近红外光谱预测模型,模型的潜变量个数根据累计贡献率大于99.9%确定。该模型变量少、稳健、可解释性强、运算速度快,其预测两台从机样品蛋白质的平均相对误差(MARE)分别为1.64%与1.88%,均小于MC-UVE模型的从机MARE(5.40%与5.19%)和VIP模型的从机MARE(6.23%与7.16%)。因此,基于MC-UVE-VIP两步波长筛选法所建立的玉米蛋白质含量近红外光谱模型可直接传递到从机,无需进行模型或从机光谱校正。该模型传递性能优于单纯基于MC-UVE或VIP算法筛选波长所建模型及全波长模型。
Previous studies have shown that selecting suitable wavelengths to establish a near-infrared(NIR) spectral model can realize the transfer of calibration model without standard samples.Using the corn spectra of three near-infrared instruments and the protein in corn as the case study,a two-step wavelength selection method of MC-UVE-VIP is proposed in this paper by combining Monte Carlo-uninformative variable elimination(MC-UVE) method with variable importance in projection(VIP) method.The method selects important and informative wavelength variables to build partial least square(PLS) calibration model for protein in corn.Firstly,the method uses MC-UVE to screen the set of ,U,UVE, consisting of informative wavelengths,whose stability parameter for screening wavelengths are greater than a threshold(,M,threshold,).Then,the VIP algorithm is used to screen the important and informative wavelength set of ,U,UVE-VIP,,whose VIP parameter is greater than the average value of VIP of all wavelengths in the ,U,UVE,.A PLS near-infrared spectral model for predicting protein in corn is established based on ,U,UVE-VIP,.The number of latent variables of the PLS model is determined according to the principle that accumulative contribution rate of the first latent variables should be higher or equal to 99.9%.The model is of few variables,robust,good at interpretability,and fast at calculation.The mean absolute relative error(MARE) of predicting protein in the samples tested on the secondaries is 1.64% and 1.88%,respectively.While the MARE of the secondary samples predicted by MC-UVE and VIP models are 5.19%-5.40% and 6.23%-7.16%,respectively.Therefore,the near-infrared spectral model of corn protein based on the MC-UVE-VIP two-step wavelength screening method can be directly transferred to the secondary instruments without correcting model or secondary samples spectra.The transfer performance of the model is better than that of the model based on single MC-UVE or single VIP wavelength selection algorithm,and is better than full wavelength model.The adoptability of the two-step wavelength selection method of MC-UVE-VIP for the other data set needs to be inspected further.
两步波长筛选蒙特卡洛-无变量信息消除(MC-UVE)变量重要性投影(VIP)无标样模型转移玉米蛋白质含量预测
two-step wavelength screeningMonte Carlo-uninformative variable elimination(MC-UVE)variable importance in projection(VIP)model transfer without standard samplesprediction of corn protein
Xu Q L,Guo L Y,Du K,Shan B M,Zhang F K.J. Instrum. Anal. (徐啟蕾,郭鲁钰,杜康,单宝明,张方坤.分析测试学报),2022,41(8):1229-1234,1241.
Ni L J,Wang N N,Zhang L G,Luan S R.J. Instrum. Anal. (倪力军,王南南,张立国,栾绍嵘.分析测试学报),2017,36(5):607-613.
Geng Y R,Shen H C,Ni H F,Wang H,Wu J Z,Zhang L L,Li Y S,He W M,Chen Y,Liu X S.J. Instrum. Anal. 耿莹蕊,沈欢超,倪鸿飞,王辉,吴继忠,张立立,李永生,何文苗,陈勇,刘雪松.分析测试学报),2022,41(7):1066-1071.
Ni L J,Zhang L G,Xie J,Luo J Q.Anal. Chim. Acta,2009,633(1):43-50.
Ni L J,Han M Y,Luan S R,Zhang L G.Spectrochim. Acta A,2019,206:350-358.
Ni L J,Han M Y,Zhang L G,Luan S R.Chin. J. Anal. Chem. (倪力军,韩明月,张立国,栾绍嵘.分析化学),2018,46(10):1660-1668.
Zhang L G,Li Y Q,Ni L J,Ge J.Spectrochim. Acta A,2020,227:117647.
Hong S J,Huang W,Zhang L G,Ge J,Ni L J,Luan S R.J. Instrum. Anal. (洪士军,黄雯,张立国,葛炯,倪力军,栾绍嵘.分析测试学报),2020,39(10):1260-1266.
Ni L J,Chen H X,Hong S J,Zhang L G,Luan S R.Spectrochim. Acta A,2021,260:119802.
Niazi A,Leardi R.J. Chemom.,2012,26:345-351.
Valber E D A,Adriano D A G,David D D S F,Héctor C G,Roberto K H G,Mario C U A.Talanta,2018,181:38-43.
Farrés M,Platikanov S,Tskovski S,Tauler R.J. Chemom.,2015,29(10):528-536.
Tong P J,Du Y P,Zheng K Y,Wu T,Wang J J.Chemom. Intell. Lab. Syst.,2015,143:40-48.
Ni L J,Xiao L X,Zhang L G,Luan S R.J. Instrum. Anal. (倪力军,肖丽霞,张立国,栾绍嵘.分析测试学报),2018,37(5):539-546.
Wang Y D,Veltkamp D J,Kowalski B R.Anal. Chem.,1991,63:2750-2756.
Cai W S,Li Y K,Shao X G.Chemom. Intell. Lab. Syst.,2008,90:188-194.
Tahir M,Kristian H L,Lars S,Solve S.Intell. Lab. Syst.,1996,118:62-69.
Xu Z P,Fan S,Cheng W M,Liu J,Zhang P F,Yang Y,Xu C,Liu B M,Liu J,Wang Q,Wu Y.Spectrochim. Acta A,2020,230:118053.
Fu J S,Yu H D,Chen Z,Yun Y H.Infrared Phys. Technol.,2022,125:104231.
Zhang Y Y,Chen W H,Tang Z M,Gu J,Mo L N,Chen H Z.J. Instrum. Anal. (张优优,陈伟豪,唐志敏,辜洁,莫丽娜,陈华舟.分析测试学报),2020,39(11):1392-1397.
Yu H D,Yun Y H,Zhang W M,Chen H M,Liu D L,Zhong Q P,Chen W X,Chen W J.Spectrochim. Acta A,2020,224:117376.
Li Y Q,Hong S J,Huang W,Zhang L G,Ge J,Luan S R,Ni L J.J. Instrum. Anal. 李永琪,洪士军,黄雯,张立国,葛炯,栾绍嵘,倪力军.分析测试学报),2020,39(10):1231-1238.
Galvão R K H,Araujo M C U,José G E,Pontes M J C,Silva E C,Saldanha T C B.Talanta,2005,67(4):736-740.
Workman J.Practical Guide to Interpretive Near-Infrared Spectroscopy.Chu X L,Xu Y P,Tian G Y,transl. Beijing:Chemical Industry Press(沃克曼·文依.近红外光谱解析实用指南.褚小立,许育鹏,田高友,译.北京:化学工业出版社),2009:219-233.
0
浏览量
5
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构