• Spectroscopy and Spectral Analysis
  • Vol. 40, Issue 12, 3789 (2020)
Yao-yao CUI1、1、*, De-ming KONG1、1, Ling-fu KONG1、1, Shu-tao WANG1、1, and Hui-chao SHI1、1
Author Affiliations
  • 1[in Chinese]
  • 11. School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2020)12-3789-06 Cite this Article
    Yao-yao CUI, De-ming KONG, Ling-fu KONG, Shu-tao WANG, Hui-chao SHI. An Oil Identification Method Based on Reconstructed 3D Fluorescence Spectra Combined With Partial Least Squares Discriminant Analysis[J]. Spectroscopy and Spectral Analysis, 2020, 40(12): 3789 Copy Citation Text show less
    Original fluorescence spectrum of a gasoline sample
    Fig. 1. Original fluorescence spectrum of a gasoline sample
    Fluorescence spectrum of gasoline removal scattering
    Fig. 2. Fluorescence spectrum of gasoline removal scattering
    Fluorescence spectrum of normalized gasoline samples
    Fig. 3. Fluorescence spectrum of normalized gasoline samples
    Identification of abnormal samples
    Fig. 4. Identification of abnormal samples
    Residual figure of components
    Fig. 5. Residual figure of components
    3D fluorescence spectrum, reconstruction 3D fluorescence spectrum and residual distribution of gasoline samples
    Fig. 6. 3D fluorescence spectrum, reconstruction 3D fluorescence spectrum and residual distribution of gasoline samples
    Selection of LVs
    Fig. 7. Selection of LVs
    The first 3 LVs scores of oil samples
    Fig. 8. The first 3 LVs scores of oil samples
    PLS-DA modeling and classification results
    Fig. 9. PLS-DA modeling and classification results
    SampleConcentration/(mg·mL-1)
    Q0.10.20.512
    C0.10.20.512
    H0.10.20.512
    R0.10.20.512
    Table 1. Oil samples concentration
    Evaluation indexEEMa-PLS-DAConstEEMb-PLS-DA
    CHQRCHQR
    CC training/%100100100100100100100100
    CC Test/%100506020100100100100
    Accuracy/%5041.6730-16.67100100100100
    Sensitivity/%41.67400-200100100100100
    Specificity/%100506020100100100100
    F-score/%58.8244.44044.44100100100100
    Table 2. PLS-DA modeling and classification evaluation results of 3D fluorescence spectra
    Yao-yao CUI, De-ming KONG, Ling-fu KONG, Shu-tao WANG, Hui-chao SHI. An Oil Identification Method Based on Reconstructed 3D Fluorescence Spectra Combined With Partial Least Squares Discriminant Analysis[J]. Spectroscopy and Spectral Analysis, 2020, 40(12): 3789
    Download Citation