The Journal of Light Scattering, Volume. 37, Issue 2, 221(2025)

Research on the detection method of sports drink composition based on the VMD-PLSR algorithm and Raman spectroscopy

PAN Qiji1、*, ZHENG Ranhua2, and ZHOU Zuojie3
Author Affiliations
  • 1School of Physical Education, Yichun University, Yichun Jiangxi 336000, China
  • 2Dalian University of Technology, Liaoning Dalian 116024, China
  • 3College of Food Science and Engineering, Jiangxi Agricultural University, Nanchang, Jiangxi 330045, China
  • show less

    Detecting the ingredients of sports drinks has essential application value in protecting consumer health and safety and ensuring the production quality of enterprises. With the rapid expansion of the sports drink market, consumers are increasingly concerned about the safety and authenticity of beverage ingredients. Therefore, there is an urgent need for an efficient detection method to meet market demand and protect consumer rights. This study aims to develop a Raman spectroscopy technique based on the combination of Variational Mode Decomposition (VMD) and Partial Least Squares Regression (PLSR) algorithms for rapid and accurate detection of ingredients in sports drinks. Firstly, spectral data of various sports drink samples were obtained through Raman spectroscopy technology. Raman spectroscopy is often affected by noise, baseline drift, and other interference factors, leading to a decrease in the accuracy of analysis results. To solve this problem, this article introduces the VMD algorithm to decompose complex signals into multiple modalities, effectively extracting essential features from the signal and reducing noise interference in the analysis results. The PLSR algorithm was used to model the spectral features processed by VMD, and the predictive ability and stability of the model were evaluated through cross-validation and external validation. The R2 and RMSE of the test set were 0.9759 and 0.0024, respectively, and the R2 in the low-concentration application testing stage was 0.8901. The results indicate that the VMD-PLSR method has significantly better accuracy and robustness in component detection than traditional regression analysis methods. In summary, the VMD-PLSR algorithm and Raman spectroscopy-based method for detecting ingredients in sports drinks have demonstrated outstanding potential for application, providing an efficient and reliable technical means for quality control and ingredient analysis of sports drinks. Future work will focus on optimizing algorithm parameters and expanding the application of this method in detecting other food and beverage ingredients further to improve the efficiency and accuracy of food safety detection.

    Tools

    Get Citation

    Copy Citation Text

    PAN Qiji, ZHENG Ranhua, ZHOU Zuojie. Research on the detection method of sports drink composition based on the VMD-PLSR algorithm and Raman spectroscopy[J]. The Journal of Light Scattering, 2025, 37(2): 221

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Oct. 29, 2024

    Accepted: Jul. 31, 2025

    Published Online: Jul. 31, 2025

    The Author Email: PAN Qiji (panqiji90@163.com)

    DOI:10.13883/j.issn1004-5929.202502009

    Topics