Quantitative determination of Al-Cu-Mg-Fe-Ni aluminum alloy using laser-induced breakdown spectroscopy combined with LASSO-LSSVM regression

As an important aerospace equipment material, the content of its constituent elements will directly affect the microstructure and properties of the Al-Cu-Mg-Fe-Ni aluminum alloy. Quantitative determination of the constituent elements in the aluminum alloy is an important part of the online detection of alloy composition. The noise of the emission source and self-absorption effect have a certain influence on the determination of minor elements in the aluminum alloy by laser-induced breakdown spectroscopy (LIBS). The univariate model has poor performance in LIBS spectral data processing and analysis. We have employed LIBS technology combined with least absolute shrinkage and selection operator (LASSO) for spectral feature selection and the least squares support vector machine (LSSVM) for regression to establish a multivariate quantitative analysis model for the four main non-aluminum elements (Mg, Cu, Fe, and Ni) in aluminum alloy, and then compare it with the traditional univariate linear calibration and partial least squares regression (PLSR) model to verify the accuracy of the multivariate calibration model. The results demonstrated that the capacity of LIBS combined with machine learning in determination for minor elements in aluminum alloys, which could be potentially used for metal composition detection in aerospace equipment.

Automated summary

Al-Cu-Mg-Fe-Ni aluminum alloy is a crucial aerospace equipment material, and the combination of LIBS technology with machine learning shows potential in accurately determining minor elements in the alloy. By establishing a multivariate quantitative analysis model, it can be applied for metal composition detection in aerospace equipment.