NIR spectrometric determination of quality parameters in vegetable oils using iPLS and variable selection

This paper proposes an analytical method for simultaneous near-infrared (NIR) spectrometric determination of acidity, refractive index and viscosity in four types of edible vegetable oils (corn, soya, canola and sunflower). For this purpose, a combination of spectral range selection by interval partial least squares (iPLS) and variable selection by the successive projections algorithm (SPA) is proposed to obtain simple multiple linear regression (MLR) models based on a small subset of wavenumbers. An independent set of samples was employed to evaluate the prediction ability of the resulting MLR–SPA models. As a result, correlation values of 0.94, 0.98, and 0.96 were obtained between model predictions and reference values for acidity, refractive index, and viscosity, respectively. The results show that a single calibration can be successfully performed for each parameter, without the need for developing a separate model for each vegetable oil type.