Identification of aluminum alloy by laser‐induced breakdown spectroscopy combined with machine algorithm

Abstract

Discarded aluminum alloys are a form of recyclable metal materials, and their classification and identification are highly important. In this work, laser‐induced breakdown spectroscopy (LIBS) technique combined with principal component analysis (PCA) and least‐squares support‐vector machine (LSSVM) algorithm were used to classify and identify five types of aluminum alloys. Exploratory analysis of five types of aluminum alloys by PCA was performed to achieve better segregation. The identification accuracy of the support‐vector machine (SVM) and LSSVM for aluminum alloy were 98.33% and 100%, respectively. The higher identification success rate was obtained using the LSSVM algorithm. Therefore, the LIBS technique combined with the PCA and LSSVM algorithms represents an efficient approach to identifying aluminum alloys.

Quantitative analysis of Cu in Traditional Chinese medicinal materials using laser‐induced breakdown spectroscopy

“…Dai et al 16 identified aluminum alloy by LIBS combined with machine algorithm. Sirven et al 17 used LIBS combined with neural networks to perform qualitative and quantitative analysis of chromium-contaminated soil. Wang et al 18 used external and internal standard methods, and multiple linear regression to quantitatively analyze Pb in tea.…”

Determining the temperature of aluminum plasma produced by laser‐induced breakdown spectroscopy based on its ultraviolet emission spectra

“…It is widely used in qualitative elemental analysis. 7,8 However, its reliability is relevant to the laser properties, substrates, and ambient surroundings. 9 Furthermore, it is important to measure accurate temperature for improving the application of LIBS.…”

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

“…Machine learning applied to LIBS data from aluminium alloys has been reported in other papers. 46,47 The paper by Dai et al identied different aluminium alloys (060, 6061, 5052, 2024, and 7075) with the prospect of re-cycling them from scrap. 47 The authors rst used PCA to help with the identication and then followed this by least squares support vector machine to classify ve different aluminium alloys.…”

“…46,47 The paper by Dai et al identied different aluminium alloys (060, 6061, 5052, 2024, and 7075) with the prospect of re-cycling them from scrap. 47 The authors rst used PCA to help with the identication and then followed this by least squares support vector machine to classify ve different aluminium alloys. The results were compared with those obtained from data analysis using support vector machine alone.…”