• Laser & Optoelectronics Progress
  • Vol. 60, Issue 17, 1712007 (2023)
Jiazhong Zheng1,2, Zhigang Li1, Feng Chen3, Zhongming Guo3..., Shu Wang1 and Min Ji1,*|Show fewer author(s)
Author Affiliations
  • 1Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, Anhui , China
  • 2University of Science and Technology of China, Hefei 230026, Anhui , China
  • 3Longyan Tobacco Industry Co., Ltd., Longyan 364030, Fujian , China
  • show less
    DOI: 10.3788/LOP222354 Cite this Article Set citation alerts
    Jiazhong Zheng, Zhigang Li, Feng Chen, Zhongming Guo, Shu Wang, Min Ji. Activated Carbon Weight Detection Based on Microwave Resonance Technology[J]. Laser & Optoelectronics Progress, 2023, 60(17): 1712007 Copy Citation Text show less
    Schematic diagram of activated carbon filter rods
    Fig. 1. Schematic diagram of activated carbon filter rods
    Schematic diagram of activated carbon weight detection device
    Fig. 2. Schematic diagram of activated carbon weight detection device
    Schematic diagram of the resonance curve and microwave amplitude change signal before and after perturbation of microwave resonator. (a) Resonance curve; (b) schematic diagram of a microwave amplitude change signal
    Fig. 3. Schematic diagram of the resonance curve and microwave amplitude change signal before and after perturbation of microwave resonator. (a) Resonance curve; (b) schematic diagram of a microwave amplitude change signal
    Data historgam
    Fig. 4. Data historgam
    Microwave signals before and after baseline correction
    Fig. 5. Microwave signals before and after baseline correction
    Flow chart of peak height and peak area acquisition
    Fig. 6. Flow chart of peak height and peak area acquisition
    Eigenvalue extraction
    Fig. 7. Eigenvalue extraction
    Support vector machine regression diagram
    Fig. 8. Support vector machine regression diagram
    Original microwave signal and baseline correction results
    Fig. 9. Original microwave signal and baseline correction results
    EigenvalueModel formula
    AM=42.803A+34.418
    HM=40.089H+33.625
    WM=55.370W+35.171
    AHM=23.5507A+17.7795H+34.6415
    AWM=33.7886A+10.2371W+36.0585
    HWM=31.1707H+13.6721W+33.7931
    AHWM=24.7973A+14.8684W+4.2529W+34.1854
    Table 1. Modeling results of activated carbon weight
    n191715131197
    RSN /dB13.494413.368013.763811.860111.387611.81426.6117
    RMSE0.026890.003350.002660.003000.002970.002670.03575
    Table 2. Filtering results with different Gaussian window lengths
    Baseline correction algorithmVariance
    ASLS0.8649
    AIRPLS0.7577
    ARPLS0.5620
    DRPLS0.3523
    Table 3. Variance of the signal after removing characteristic peaks

    method

    Modeling

    SVRPLSBP
    MAE /mgR2RSD /%MAE /mgR2RSD /%MAE /mgR2RSD /%

    Unary

    regression

    A0.79790.99241.49621.04920.99041.84621.06330.98981.8975
    H1.29600.97972.19791.32050.97972.25531.37930.98072.3374
    W3.22990.85335.90313.33090.85556.22433.47790.84536.4493
    Multiple regressionAH0.86430.99211.55690.88800.99231.60420.92740.99181.7285
    AW1.11550.98702.19671.03550.99051.81141.25600.98202.4971
    HW1.16390.98452.17771.15270.98462.09131.24670.98182.3107
    AHW0.84390.99181.63620.89500.99221.60940.97140.98631.8172
    Table 4. Comparison of prediction effects of different modeling methods
    Serial numberM1M2M3M4
    139.7556.7160.3370.01
    238.8354.1260.9272.00
    340.6654.6057.2070.70
    439.9954.1356.4270.54
    537.8858.3357.6070.27
    637.9256.5160.6771.08
    739.4153.2855.5168.86
    839.7753.1956.2968.86
    938.0955.2458.2772.17
    1038.8355.3357.1871.28
    Weight value39.255.358.970.1
    Standard deviation0.911.561.851.08
    Table 5. Repeatability of activated carbon weight detection (n=10)
    Serial numberAbsolute error /mgRelative error /%
    Location 1Location 2Location 3Location 4Location 1Location 2Location 3Location 4
    10.551.411.430.091.392.562.430.13
    20.371.182.021.900.932.133.422.72
    31.460.701.700.603.731.262.890.86
    40.791.172.480.442.022.124.210.62
    51.323.031.300.173.365.472.210.24
    61.281.211.770.983.262.203.011.39
    70.212.023.391.240.533.665.761.78
    80.572.112.611.241.443.824.431.76
    91.110.060.632.072.830.121.072.95
    100.370.031.721.180.950.052.921.69
    Weight value39.255.358.970.139.255.358.970.1
    Average value0.801.291.910.992.052.343.231.41
    Table 6. Precision results of activated carbon weight detection
    Jiazhong Zheng, Zhigang Li, Feng Chen, Zhongming Guo, Shu Wang, Min Ji. Activated Carbon Weight Detection Based on Microwave Resonance Technology[J]. Laser & Optoelectronics Progress, 2023, 60(17): 1712007
    Download Citation