Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 61, Issue 17, 1730001(2024)

Rapid Identification of Homogeneous Alloys Based on Laser-Induced Breakdown Spectroscopy Combined with Machine-Learning Algorithms

Wanxue Li1、*, Yaxiong He2, Yang Li3, Feinan Cai4, and Yong Zhang2
Author Affiliations
  • 1College of Physics and Engineering Technology, Chengdu Normal University, Chengdu 611130, Sichuan, China
  • 2School of Electrical Engineering, Southwest Jiaotong University, Chengdu 610031, Sichuan, China
  • 3Shandong Institute of Geology and Mineral Resources Exploration, Yantai 264011, Shandong, China
  • 4State Grid Sichuan Power Company Chengdu Power Supply Company, Chengdu 610041, Sichuan, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (11)
    Equations (7)
    References (27)
    Tools
    Paper Information
    Topics
    Figures & Tables(11)
    Schematic diagram of LIBS experimental device

    Fig. 1. Schematic diagram of LIBS experimental device

    Download full size
    Schematic diagram of SNIP algorithm in peak area

    Fig. 2. Schematic diagram of SNIP algorithm in peak area

    Download full size
    Schematic diagram of the SVM principle

    Fig. 3. Schematic diagram of the SVM principle

    Download full size
    Typical LIBS spectra of alloy-steel samples

    Fig. 4. Typical LIBS spectra of alloy-steel samples

    Download full size
    Comparison of LIBS spectra before and after background deduction

    Fig. 5. Comparison of LIBS spectra before and after background deduction

    Download full size
    PCA dimensionality reduction results. (a) PCA score chart; (b) first two PC scatter plots

    Fig. 6. PCA dimensionality reduction results. (a) PCA score chart; (b) first two PC scatter plots

    Download full size
    Flow chart of identification model establishment and result prediction for homogeneous alloy-steel

    Fig. 7. Flow chart of identification model establishment and result prediction for homogeneous alloy-steel

    Download full size
    Confusion matrix of training set. (a) SVM; (b) decision tree; (c) KNN; (d) LDA

    Fig. 8. Confusion matrix of training set. (a) SVM; (b) decision tree; (c) KNN; (d) LDA

    Download full size
    Confusion matrix of testing set. (a) SVM; (b) decision tree; (c) KNN; (d) LDA

    Fig. 9. Confusion matrix of testing set. (a) SVM; (b) decision tree; (c) KNN; (d) LDA

    Download full size

    • Table 1. Certification form for mass fractions of various elements in alloy-steel samples

      View table

      Table 1. Certification form for mass fractions of various elements in alloy-steel samples

      No.Mass fraction /%
      CCrSiMnAlVCu
      BHG1606-10.0522.5100.5231.9150.0100.0510.052
      BHG1606-20.2961.4700.3390.9410.0740.2760.301
      BHG1606-30.1641.9420.0980.1600.0500.1330.179
      BHG1606-40.4290.9320.3120.6490.3230.4070.419
      BHG1606-50.5720.3830.2621.5800.2300.5250.580
      YSBS18201-940.0760.0131.6401.1000.0190.3910.034
      YSBS18203-940.3161.4600.5680.6130.1900.4510.239
      YSBS18205-940.4250.4650.1271.3900.0920.0290.135
      YSBS18206-940.3470.7530.7520.290‒0.1320.233

    • Table 2. Comparison of homogeneous metal identification results with different algorithms

      View table

      Table 2. Comparison of homogeneous metal identification results with different algorithms

      Algorithm type

      Brand

      Training set

      Test set

      Algorithm type

      Brand

      Training set

      Test set

      Accuracy

      Mean accuracy

      Accuracy

      Mean accuracy

      Accuracy

      Mean accuracy

      Accuracy

      Mean accuracy

      SVM

      S1

      92.9

      99.06

      100

      96.29

      KNN

      S1

      95.7

      90.47

      100

      67.04

      S2

      100

      100

      S2

      95.7

      70.0

      S3

      100

      83.3

      S3

      97.1

      56.7

      S4

      100

      100

      S4

      78.6

      70.0

      S5

      100

      100

      S5

      84.3

      63.3

      S6

      100

      100

      S6

      90.0

      96.7

      S7

      98.6

      100

      S7

      94.3

      23.3

      S8

      100

      100

      S8

      91.4

      56.7

      S9

      100

      83.3

      S9

      87.1

      66.7

      Decision tree

      S1

      98.6

      97.47

      100

      79.63

      LDA

      S1

      100

      100

      100

      100

      S2

      98.6

      70.0

      S2

      100

      100

      S3

      98.6

      56.7

      S3

      100

      100

      S4

      98.6

      100

      S4

      100

      100

      S5

      100

      100

      S5

      100

      100

      S6

      100

      100

      S6

      100

      100

      S7

      90.0

      100

      S7

      100

      100

      S8

      97.1

      20.0

      S8

      100

      100

      S9

      95.7

      70.0

      S9

      100

      100

    Tools

    Get Citation

    Copy Citation Text

    Wanxue Li, Yaxiong He, Yang Li, Feinan Cai, Yong Zhang. Rapid Identification of Homogeneous Alloys Based on Laser-Induced Breakdown Spectroscopy Combined with Machine-Learning Algorithms[J]. Laser & Optoelectronics Progress, 2024, 61(17): 1730001

    Download Citation

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

    Category: Spectroscopy

    Received: Nov. 1, 2023

    Accepted: Jan. 29, 2024

    Published Online: Sep. 14, 2024

    The Author Email: Wanxue Li (071065@cdnu.edu.cn)

    DOI:10.3788/LOP232417

    CSTR:32186.14.LOP232417

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology