光谱学与光谱分析
光譜學與光譜分析
광보학여광보분석
SPECTROSCOPY AND SPECTRAL ANALYSIS
2014年
10期
2863-2867
,共5页
王亚红%董大明%周萍%郑文刚%叶松%王文重
王亞紅%董大明%週萍%鄭文剛%葉鬆%王文重
왕아홍%동대명%주평%정문강%협송%왕문중
近红外光谱%光程%预处理%偏最小二乘
近紅外光譜%光程%預處理%偏最小二乘
근홍외광보%광정%예처리%편최소이승
Near infrared spectroscopy%Optical path%Pretreatment%Partial least squares
以蔗糖溶液为研究对象,利用近红外光谱分别测量4,5和6 m m光程下不同浓度蔗糖溶液的透反射光谱,研究采用矢量归一化、基线偏移校正、多元散射校正、标准正态变量变换、一阶导数5种预处理方法消除光程差异的影响,并结合PLS方法建立校正集模型。与原始光谱的PLS模型相比,五种预处理方法均对模型的预测精度有不同程度的提高,其中,多元散射校正结合 PLS方法建立的模型最优,使原始光谱的主成分数 PC 由6下降为3,决定系数 R2由0.891278提高到0.987535,交互验证决定系数 R2CV由0.888374提高到0.983343,校正标准偏差 RM S EC由1.704%下降到0.89%,交互验证的校正标准偏差RMSECV由1.827%下降到1.05%,预测集样本的相关系数由0.95089上升到0.97622,预测标准偏差由0.01436下降为0.01。结果表明,五种预处理方法中,多元散射校正法能够消除光程差异的干扰,提高模型的预测精度,改善稳定性。
以蔗糖溶液為研究對象,利用近紅外光譜分彆測量4,5和6 m m光程下不同濃度蔗糖溶液的透反射光譜,研究採用矢量歸一化、基線偏移校正、多元散射校正、標準正態變量變換、一階導數5種預處理方法消除光程差異的影響,併結閤PLS方法建立校正集模型。與原始光譜的PLS模型相比,五種預處理方法均對模型的預測精度有不同程度的提高,其中,多元散射校正結閤 PLS方法建立的模型最優,使原始光譜的主成分數 PC 由6下降為3,決定繫數 R2由0.891278提高到0.987535,交互驗證決定繫數 R2CV由0.888374提高到0.983343,校正標準偏差 RM S EC由1.704%下降到0.89%,交互驗證的校正標準偏差RMSECV由1.827%下降到1.05%,預測集樣本的相關繫數由0.95089上升到0.97622,預測標準偏差由0.01436下降為0.01。結果錶明,五種預處理方法中,多元散射校正法能夠消除光程差異的榦擾,提高模型的預測精度,改善穩定性。
이자당용액위연구대상,이용근홍외광보분별측량4,5화6 m m광정하불동농도자당용액적투반사광보,연구채용시량귀일화、기선편이교정、다원산사교정、표준정태변량변환、일계도수5충예처리방법소제광정차이적영향,병결합PLS방법건립교정집모형。여원시광보적PLS모형상비,오충예처리방법균대모형적예측정도유불동정도적제고,기중,다원산사교정결합 PLS방법건립적모형최우,사원시광보적주성분수 PC 유6하강위3,결정계수 R2유0.891278제고도0.987535,교호험증결정계수 R2CV유0.888374제고도0.983343,교정표준편차 RM S EC유1.704%하강도0.89%,교호험증적교정표준편차RMSECV유1.827%하강도1.05%,예측집양본적상관계수유0.95089상승도0.97622,예측표준편차유0.01436하강위0.01。결과표명,오충예처리방법중,다원산사교정법능구소제광정차이적간우,제고모형적예측정도,개선은정성。
Based on sucrose solution as the research object ,this paper measured the trans-reflective spectrum of sucrose solution of different concentration by the technique of near infrared spectrum in three optical path (4 ,5 ,6 mm) .Five kinds of pretreat-ment method (vector normalization ,baseline offset correction ,multiplicative scatter correction ,standard normal variate trans-formation ,a derivative) were used to eliminate the influence of the optical path difference ,and to establish model of the calibra-tion set in combination with the PLS (Partial Least Squares)method .Five kinds of pretreatment method could restrain the inter-ference of light path in varying degrees .Compared with the PLS model of original spectra ,the model of multiple scattering cor-rection combined with PLS method is the optimal model .The results of quantitative analysis of original spectra :the number of principal component PC=6 ,the determination coefficient R2 =0.891 278 ,the determination coefficient of cross validation R2CV =0.888 374 ,root mean square error of calibration RMSEC = 1.704% ,root mean square error of cross validation RMSECV =1.827% ;The results of quantitative analysis of spectra after MSC pretreatment :the number of principal component PC=3 ,the determination coefficient R2 =0.987 535 ,the determination coefficient of cross validation R2CV =0.983 343 ,root mean square er-ror of calibration RMSEC=0.89% ,root mean square error of cross validation RMSECV =1.05% .The correlation coefficient of the prediction set is as much as 0.976 22 .root mean square error of prediction is 0.01 ,lesser than 0.014 36 .The results show that the MSC can eliminate the influence of optical path difference ,improve the prediction precision and improve the stability .