Rapid Determination of Peroxide Value and Acid Value in Rice Bran Oil Using Near Infrared Spectroscopy Coupled with Partial Least Squares Regression

Near infrared spectroscopy was applied to the rapid determination of peroxide value and acid value in rice bran oil. Using chemical titration to determine the content of peroxide value and acid value in rice bran oil as a reference method, using near infrared spectroscopy analysis technology combined with partial least squares regression, established the near infrared quantitative analysis model. The correlation coefficients of the established peroxide value and acid value quantitative analysis models were 99.11% and 99.72%, the standard deviation of prediction were 0.436 7% and 0.044 78%, and the cross validation standard deviation were 0.452 2% and 0.046 69%. The calibration model was independently verified externally with 78 samples. The absolute errors of peroxide value and acid value were –0.78%~0.82% and –0.142%~0.139%. The relative errors of peroxide value and acid value were –13.21%~14.44% and –14.85%~12.46%. The established quantitative analysis model and method is simple, rapid and accurate in the content determination of peroxide value and acid value in rice bran oil.