NUCLEAR TECHNIQUES, Volume. 47, Issue 4, 040403(2024)

Neutron/gamma (n/γ) discrimination method based on KPCA-MPA-ELM

Wanping HU1, Guiyu ZHANG1,2,3、*, Yunlong ZHANG1, Xianguo TUO1,2, and Hulin LI4
Author Affiliations
  • 1School of Automation & Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China
  • 2Artificial Intelligence Key Laboratory of Sichuan Province, Sichuan University of Science & Engineering, Yibin 644000, China
  • 3School of Information Engineering, Southwest University of Science and Technology, Mianyang 621010, China
  • 4Chengdu Huilite Automation Technology Co., Ltd., Chengdu 610000, China
  • show less
    Figures & Tables(14)
    Experimental flowchart
    Pulse samples of neutron and gamma rays
    Intercepted pulse waveform (a) and baseline adjustment (b), smoothing filtering (c), and amplitude normalization (d)
    Diagram of the KPCA-ELM model structure
    Specific algorithm flowchart of the KPCA-MPA-ELM model
    Contribution rate and cumulative contribution rate of the first 10 principal component features
    Fitness change curve of the MPA-ELM
    Confusion matrix of the 3 000 test samples of the KPCA-ELM model
    Confusion matrix of the 3 000 groups of the test samples of the KPCA-MPA-ELM model
    Comparison of the classification accuracy of four models
    CCM discrimination result chart (color online)
    PGA discrimination result chart
    • Table 1. Comparison of the classification efficiency of four models

      View table
      View in Article

      Table 1. Comparison of the classification efficiency of four models

      模型

      Models

      平均准确率

      Average accuracy

      / %

      平均运行时间

      Average running

      time / s

      ELM86.882.53
      MPA-ELM96.5517.09
      KPCA-ELM97.510.83
      KPCA-MPA-ELM99.0716.87
    • Table 2. Comparison of the discrimination efficiency between the machine-learning model and traditional PSD method

      View table
      View in Article

      Table 2. Comparison of the discrimination efficiency between the machine-learning model and traditional PSD method

      方法

      Methods

      准确率

      Accuracy / %

      运行时间

      Running time / s

      CCM97.275.39
      PGA93.165.27
      KPCA-ELM97.510.83
      KPCA-MPA-ELM99.0716.87
    Tools

    Get Citation

    Copy Citation Text

    Wanping HU, Guiyu ZHANG, Yunlong ZHANG, Xianguo TUO, Hulin LI. Neutron/gamma (n/γ) discrimination method based on KPCA-MPA-ELM[J]. NUCLEAR TECHNIQUES, 2024, 47(4): 040403

    Download Citation

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

    Category: Research Articles

    Received: Sep. 25, 2023

    Accepted: --

    Published Online: May. 28, 2024

    The Author Email: ZHANG Guiyu (张贵宇)

    DOI:10.11889/j.0253-3219.2024.hjs.47.040403

    Topics