The Journal of Light Scattering, Volume. 37, Issue 2, 292(2025)
Identification of Adulteration in Extra Virgin Olive Oil based on UV Laser-induced Fluorescence Spectroscopy Combined with FA-BPNN Algorithm
Olive oil is the most popular healthy oil for Mediterranean residents, and domestic people have gradually begun to accept this high-quality natural oil. However, adulteration and shoddy are the main problems the domestic olive oil consumption market faces, especially the extra virgin olive oil. This paper used the artificial intelligence algorithm combined with fluorescence spectroscopy to detect the adulterated concentration of extra virgin olive oil. UV laser-induced fluorescence spectroscopy detected the fluorescence signals of corn oil, rice oil, extra virgin olive oil, and their mixed oils. A factor analysis algorithm carried out the dimension reduction and feature extraction of high-dimensional fluorescence spectrum data, and then the concentration detection of extra virgin olive oil adulterated by corn oil and rice oil was realized by combining it with a back-propagation neural network algorithm. The determination coefficients of factor analysis combined with the back-propagation neural network algorithm model were 0.9738 and 0.9658, and the root mean square errors were 0.0031 and 0.0040. That is, the method proposed in this paper has better prediction performance. In a word, this paper verifies that UV laser-induced fluorescence spectroscopy combined with an artificial intelligence algorithm can predict the concentration of corn oil and rice oil adulterated with extra virgin olive oil, which provides a reference methodology for relevant researchers and law enforcement departments.
Get Citation
Copy Citation Text
ZHANG Xiquan, ZHOU Dexu, ZONG Weikai. Identification of Adulteration in Extra Virgin Olive Oil based on UV Laser-induced Fluorescence Spectroscopy Combined with FA-BPNN Algorithm[J]. The Journal of Light Scattering, 2025, 37(2): 292
Category:
Received: Jul. 14, 2024
Accepted: Jul. 31, 2025
Published Online: Jul. 31, 2025
The Author Email: ZHANG Xiquan (zhangxiquan83@163.com)