Research on Measurement Method of Sucrose Concentration by Short-Wave Near-Infrared Trans-Reflective Spectroscopy

摘要

In this paper, the technology of near-infrared spectra of trans-reflective was applied to determine the feasibility of nondestructive testing of the sucrose solution in the near-infrared wavelength range of 780～1100 nm, and principal component regression (PCR) partial least squares (PLS) and support vector regression (SVR) are used to establish the near-infrared quantitative analysis model of sucrose solution. After the pretreatment of (Savitzky-Golay) 5 points smoothing and multiplicative scatter correction (MSC), the calibration model can be made. The results of quantitative analysis of PCR: the number of principal component PC is 7, the correlation coefficient R is 0.964607, root mean square error of cross validation (RMSECV) is 1.6895%; The results of quantitative analysis of PLS: the number of principal component PC, the correlation coefficient (R) and root mean square error of cross validation (RMSECV) is 4, 0.010122, 1.16% respectively. The correlation coefficient R, RMSECV by SVR algorithm is 0.841332, and 36207% respectively. To compare three kinds of models, the performance of SVR model is the worst, PCR and PLS model performance are relatively high. The prediction samples are predicted respectively by the calibration model of PCR and PLS, root mean square error of prediction RMSEP is 1.2% and 1.16% respectively. Both of model (PLS and PCR) are quite good, PLS model is simpler and its prediction accuracy is higher. The results demonstrated that short-wave near-infrared spectrometry is a valuable, rapid and nondestructive tool for the quantitative analysis of sucrose solution.