Research on detection of different metallographic structures of high speed wheel steel based on laser-induced breakdown spectroscopy

Abstract

The laser-induced breakdown spectroscopy (LIBS) experimental platform was applied to obtain LIBS spectral the data of 10 CL60 wheel steel samples. The principle component analysis (PCA) was used to preliminarily analyze the macroscopic characteristics of LIBS spectral data. With the spectral intensity and spectral intensity combined with spectral intensity ratio as variables, three spectral correction methods including median filtering, baseline correction and multiple scattering correction (MSC) were used for pretreatment. And the support vector machine (SVM) qualitative model was established to determine the metallographic structure. It was found that the SVM model established by using the pre-processed data of MSC as the input variable has the best effect. The accuracy rate of calibration set is 100%, and the accuracy rate of prediction set is 98.4%. The research has shown that LIBS combined with SVM model can be used for discriminant analysis of different metallographic structures of train wheel steel.

Atomic spectrometry update: review of advances in the analysis of metals, chemicals and materials

“…Jun et al used LIBS to obtain spectra from high-speed train wheel steel and then subjected the data to PCA for preliminary analysis. 17 Using the three spectral correction methods of median filtering, baseline correction and multiple scattering correction as further pretreatment, a support vector machine (SVM) was used to provide a quantitative model. Overall, the model established using the pre-processed data of the multiple scattering correction proved to be the best.…”

Collaborative estimation of heavy metal stress in wheat seedlings based on LIBS-Raman spectroscopy coupled with machine learning

“…MSC is used to correct baselines to correct offset spectral data, eliminate scattering-induced spectral differences, and effectively promote correlation between spectra. 44,45 WT removes the random noise by retaining the low-frequency signal to remove the high-frequency noise. 46,47 SG and air-PLS could also be achieved to some extent to eliminate the effect of baseline dri and improve the accuracy of spectral analysis.…”

Microstructure classification of steel samples with different heat-treatment processes based on laser-induced breakdown spectroscopy (LIBS)

“…The calibration set exhibited impeccable accuracy at 100%, while the prediction set achieved an accuracy of 98.4%. 25 Zeng et al used restricted Boltzmann machines (RBM) and principal component analysis (PCA) for dimensionality reduction of the data set. The accuracy of the RBM-PCA model can reach 100%, and the maximum dimensionality reduction time is 33.18 s. 26 Alsayed et al employed typical correlation analysis (CCA) to reduce dimensionality, subsequently utilizing an optimized adaptive augmented random forest classifier to obtain representative information for quantitative microhardness estimation of the samples.…”