• Spectroscopy and Spectral Analysis
  • Vol. 41, Issue 9, 2789 (2021)
Jin-ge MA1、*, Qiao-ling YANG2、2;, Xiao-jun DENG1、1; *;, Yi-yin SHI1、1;, Shu-qing GU1、1;, Chao-min ZHAO1、1;, Yong-ai YU3、3;, and Feng ZHANG4、4;
Author Affiliations
  • 11. Technical Center for Animal Plant and Food Inspection and Quarantine, Shanghai Customs, Shanghai 200135, China
  • 22. School of Life Sciences, Shanghai University, Shanghai 200444, China
  • 33. Shanghai Oceanhood Opto-electronics Tech Co., LTD., Shanghai 201201, China
  • 44. Chinese Academy of Inspection and Quarantine, Beijing 100176, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2021)09-2789-06 Cite this Article
    Jin-ge MA, Qiao-ling YANG, Xiao-jun DENG, Yi-yin SHI, Shu-qing GU, Chao-min ZHAO, Yong-ai YU, Feng ZHANG. On-Site Rapid and Non-Destructive Identification Method for Imported Bulk Olive Oil Quality Based on Portable Raman Spectroscopy[J]. Spectroscopy and Spectral Analysis, 2021, 41(9): 2789 Copy Citation Text show less
    Trend chart of quality indicators of olive oil heating(a): Change in acid value; (b): Change in peroxide value; (c): Change in linolenic acid
    Fig. 1. Trend chart of quality indicators of olive oil heating
    (a): Change in acid value; (b): Change in peroxide value; (c): Change in linolenic acid
    Raman spectra of olive oil under different processing conditions
    Fig. 2. Raman spectra of olive oil under different processing conditions
    The relationship between the predicted value and the actual value of the calibration set(a): The acid value correction set of the PLS method; (b): The acid value correction set of the VAPLS method;(c): Peroxide value correction set of PLS method; (d): Peroxide value correction set of VAPLS metion;(e): Linolenic acid correction set of PLS metion; (f): Linolenic acid correction set of VAPLS metion
    Fig. 3. The relationship between the predicted value and the actual value of the calibration set
    (a): The acid value correction set of the PLS method; (b): The acid value correction set of the VAPLS method;(c): Peroxide value correction set of PLS method; (d): Peroxide value correction set of VAPLS metion;(e): Linolenic acid correction set of PLS metion; (f): Linolenic acid correction set of VAPLS metion
    方法品质指标校正集验证集
    R2RMSE相对误差范围/%R2RMSE相对误差范围/%
    酸价0.992 70.025-9.33~10.120.939 60.073 3-15.50~26.14
    PLS过氧化值0.917 70.187 4-11.25~14.280.702 70.341 8-25.81~19.61
    亚麻酸0.998 99.919 3-3.33~7.040.890 389.877 5-18.70~139.77
    酸价0.992 70.025 5-14.00~10.040.844 60.128 1-105.30~22.53
    VAPLS过氧化值0.911 30.258 4-17.00~11.420.904 30.235 714.80~21.30
    亚麻酸0.952 065.416 4-60.42~42.850.809 232.454 9-34.61~172.03
    Table 1. Model evaluation results under full spectrum
    方法品质指标校正集验证集
    R2RMSE相对误差范围/%R2RMSE相对误差范围/%
    酸价0.998 80.013 0-5.36~4.100.993 10.019 8-8.45~7.41
    PLS过氧化值0.995 40.060 8-3.46~4.500.989 80.100 9-7.43~6.35
    亚麻酸0.998 213.334 1-9.92~9.340.983 837.407 4-12.42~8.19
    酸价0.995 70.022 0-5.43~4.600.992 50.023 5-4.81~4.37
    VAPLS过氧化值0.998 60.030 9-2.04~3.410.994 60.076 6-5.00~3.42
    亚麻酸0.992 919.767 6-4.51~4.460.994 225.074 4-3.90~4.35
    Table 2. Model evaluation results after spectral preprocessing
    样品酸值/(mg·g-1)
    实测值预测值
    (PLS)
    相对误差
    /%
    预测值
    (VAPLS)
    相对误差
    /%
    10.724 80.686 3-5.320.690 7-4.71
    20.588 50.544 2-7.540.608 83.44
    30.389 10.383 7-1.400.378 3-2.78
    40.528 40.601 913.910.508 7-3.73
    50.679 80.670 0-1.440.657 3-3.32
    60.522 70.544 94.250.545 84.42
    70.538 90.514 7-4.510.567 75.34
    RMSE0.041 30.025 8
    Table 3. Evaluation results of acid value model
    样品过氧化值/[g·(100 g)-1]
    实测值预测值
    (PLS)
    相对误差
    /%
    预测值
    (VAPLS)
    相对误差
    /%
    15.224 05.570 26.635.111 4-2.16
    25.427 24.965 5-8.515.196 9-4.24
    32.206 32.259 12.402.189 6-0.76
    45.917 05.846 5-1.196.036 62.02
    57.383 97.616 33.157.606 13.01
    65.299 25.142 2-2.965.350 90.98
    74.363 34.592 45.254.501 43.17
    RMSE0.305 90.162 8
    Table 4. Model evaluation results of peroxide value
    样品亚麻酸/[mg·(100 g)-1]
    实测值预测值
    (PLS)
    相对误差
    /%
    预测值
    (VAPLS)
    相对误差
    /%
    1869.681890.1112.35891.4252.50
    2890.074929.0704.38943.2825.99
    3881.528869.104-1.41913.6613.65
    4837.253869.9833.91878.4044.91
    5856.288917.4507.14888.4113.75
    6985.232925.638-6.05974.550-1.08
    7886.793910.6292.69890.0810.37
    RMSE23.298 517.064 1
    Table 5. Model evaluation results of linolenic acid
    Jin-ge MA, Qiao-ling YANG, Xiao-jun DENG, Yi-yin SHI, Shu-qing GU, Chao-min ZHAO, Yong-ai YU, Feng ZHANG. On-Site Rapid and Non-Destructive Identification Method for Imported Bulk Olive Oil Quality Based on Portable Raman Spectroscopy[J]. Spectroscopy and Spectral Analysis, 2021, 41(9): 2789
    Download Citation