Influence of temperature on visible and near-infrared spectra and the predictive ability of multivariate models

When vibrational spectra are measured on- or in-line for process analytical or control purposes, the spectra may fluctuate in response due to fluctuations in environmental conditions, such as temperature or humidity that must be taken into consideration when developing calibration models. In this paper, the influence of temperature fluctuations on visible and near-infrared (Vis/NIR) spectra and their effect on the predictive power of calibration models, partial least squares (PLS), principal component regression (PCR) and stepwise multiple linear regression (SMLR) was studied. The sample was peach. Soluble solids content in peach was detected. The results shows influence of temperature on Vis/NIR spectra of the peach exists. The overall results sufficiently demonstrate that the performances of the same method, PLS, PCR or SMLR are similar, no matter what the data are at different temperatures.