Spectroscopy and Spectral Analysis, Volume. 40, Issue 12, 3901(2020)

Research on On-Line Classification System of Aluminum Alloy for Laser-Induced Breakdown Spectrum

Jia LIU1,1, Xue-jing SHEN1,1, Peng XU1,1, Fei-peng CUI1,1, Xiao-xia SHI1,1, Xiao-peng LI1,1, and Hai-zhou WANG1,1
Author Affiliations
  • 1[in Chinese]
  • 11. Central Iron and Steel Research Institute, Beijing 100081, China
  • show less
    Figures & Tables(8)
    Sample of aluminum alloy(a): Aluminum alloy standard sample;(b): Actual sample of aluminum alloy
    Schematic diagram of Sorting-LIBS system
    Fast identification and location of transmission materials(a): Material continuous tracking identification location image;(b): Identification and positioning algorithm flow
    System identification and detection accuracy test
    LIBS full spectrum of three series of aluminum alloy
    LIBS signals of three series of aluminum alloy characteristic lines
    Comparison of the accuracy of different characteristic variable models in predicting training set(a): Gaussian intensity normalization training set; (b): Gaussian intensity training set
    • Table 1. Classification model performance comparison

      View table
      View in Article

      Table 1. Classification model performance comparison

      Accuracy/%Accuracy
      /%
      KappaModeling
      time
      7xxx2xxxA356
      MGD99.4398.7799.7099.150.986 167 ms
      SVM98.2199.9299.8699.390.990 0640.71 s
    Tools

    Get Citation

    Copy Citation Text

    Jia LIU, Xue-jing SHEN, Peng XU, Fei-peng CUI, Xiao-xia SHI, Xiao-peng LI, Hai-zhou WANG. Research on On-Line Classification System of Aluminum Alloy for Laser-Induced Breakdown Spectrum[J]. Spectroscopy and Spectral Analysis, 2020, 40(12): 3901

    Download Citation

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

    Category: Research Articles

    Received: Oct. 15, 2019

    Accepted: --

    Published Online: Jun. 18, 2021

    The Author Email:

    DOI:10.3964/j.issn.1000-0593(2020)12-3901-05

    Topics