Quality evaluation of fried soybean oil base on near infrared spectroscopy. (5th October 2018)

• Record Type: Journal Article
• Authors: Liu, Yangyang; Sun, Laijun; Li, Yingsong; Zhang, Dan; Song, Ziwei; Zhang, Chen; Pan, Xuyang; Li, Jinlong; Wang, Zhiyu
• Is Part Of: Journal of food process engineering. Volume 41:Number 7(2018)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2018-10-05
• Subjects: Food industry and trade -- Periodicals; Food -- Analysis -- Periodicals; 664.005
• DOI: 10.1111/jfpe.12887
• Languages: English
• ISSNs: 0145-8876

Abstract

Abstract: Frying quality of soybean oil gives important effects on the quality of the frying food and consumer health. To evaluate the frying quality of soybean oil, this study proposed a method based on Near Infrared Spectroscopy (NIRS) to quickly identify the frying times which will give an effect on the harmful compounds. A model is established to analyze the relationship between the frying times and the corresponding spectra and the deterioration degree of soybean oil under different frying times is characterized. Ten experiments were set up and the first eight experimental spectra are used to construct a model which is predicted and evaluated by the last unmodeled two experimental spectra. Firstly, the first derivative (D1), the second derivative (D2), and standard normal variables transformation (SNV) are used to process the original spectra. Secondly, the combination of correlation coefficient, differential and standardized methods is considered to form a novel approach to obtain a model. Finally, the predictive accuracy of the frying times could reach 100% within the range of allowable error. The overall results indicated that the frying times of soybean oil can be effectively identified by the characteristic wavelengths, and thus the quality change of fried soybean oil is reflected. Practical applications: The detection of the frying times can help assess the quality of fried soybean oil. From a practical point of view, the frying times seriously affect the quality of fried soybean oil. Accurately predicting the frying times can effectively reflect the deterioration degree of fried soybean oil and thus improve the quality of fried soybean oil. In this work, the characteristic wavelengths are selected to establish the relationship between the frying times and the corresponding spectra, and the different frying times of fried soybean oil are judged according to different spectral range values. In the follow‐up study, we can use the characteristic wavelengths extracted in this study to develop a portable instrument to detect the frying times of fried soybean oil, and then determine the quality of fried soybean oil.