Infrared and Laser Engineering, Volume. 51, Issue 4, 20210320(2022)

Turbulence warning based on convolutional neural network by lidar

Zibo Zhuang1, Yueheng Qiu2, Jiaquan Lin2, and Delong Song2
Author Affiliations
  • 1College of Flight Technology, Civil Aviation University of China, Tianjin 300300, China
  • 2College of Electronic Information and Automation, Civil Aviation University of China, Tianjin 300300, China
  • show less
    Figures & Tables(16)
    Schematic diagram of Doppler lidar
    Schematic diagram of part of the data
    Network input layer eddy current dissipation rate image
    Output result graph of the first convolutional layer
    Output of the second convolutional layer
    Diagram convolutional neural network structuream
    Diagram of CNN training model
    Diagram of convolutional neural network training process
    Schematic diagram of the relationship between decreasing learning rate and network accuracy
    Schematic diagram of loss function during training
    Original data image of two false positives
    Diagram of the judgment of turbulence by two methods
    Schematic diagram of two early warning methods hitting turbulence
    • Table 1. Relevant parameters of lidar

      View table
      View in Article

      Table 1. Relevant parameters of lidar

      ParameterValue
      Wavelength/nm1550
      Sampling interval/ns2.5
      Laser pulse width/ns200
      Pulse repetition frequency/kHz10
      Accumulated pulse number5000
      Range resolution/m30
      Maximum detection distance/km6
    • Table 2. Turbulence warning statistics of the two methods

      View table
      View in Article

      Table 2. Turbulence warning statistics of the two methods

      DateCNN(L/M/H)Vsf(L/M/H)Times of false positives
      2016.11.25HL1
      2016.11.26HH0
      2016.11.27HH0
      2016.11.28LL0
      2016.11.29HH0
      2016.11.30MM1
      2016.12.01MM0
      2016.12.02LH1
      2016.12.03HH0
      2016.12.04HH0
      2016.12.05MM0
      2016.12.06LL0
      2016.12.07MM0
      2016.12.08HH0
      2016.12.09LL0
      2016.12.10LL0
      2016.12.11HL1
      2016.12.12HH0
      2016.12.13HH0
      2016.12.14HH0
    • Table 3. Judgment results made by two methods on 15 sets of unit reports

      View table
      View in Article

      Table 3. Judgment results made by two methods on 15 sets of unit reports

      DateCNN Alarm statistics(T/F)Classified statistics(T/F)Hog-SVM Alarm statistics(T/F)Classified statistics(T/F)
      2016.07.20TTTT
      2016.09.03TTTT
      2016.09.07FFFF
      2016.09.10TTTT
      2016.09.11TTTT
      2016.09.20TTTT
      2016.09.24TTFF
      2016.10.15TFFF
      2016.11.06TTFT
      2017.04.13FFTT
      2017.04.17TTTT
      2017.05.06TTTT
      2017.05.09TTTT
      2017.05.13TTFF
      2017.05.14TTFF
    Tools

    Get Citation

    Copy Citation Text

    Zibo Zhuang, Yueheng Qiu, Jiaquan Lin, Delong Song. Turbulence warning based on convolutional neural network by lidar[J]. Infrared and Laser Engineering, 2022, 51(4): 20210320

    Download Citation

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

    Category: Lasers & Laser optics

    Received: May. 19, 2021

    Accepted: --

    Published Online: May. 18, 2022

    The Author Email:

    DOI:10.3788/IRLA20210320

    Topics