Real-Time Monitoring of Laser-Layered Paint Removal from CFRP Based on the Synergy of Laser-Induced Breakdown Spectroscopy and PLS-DA Models

Abstract

To achieve precise removal of different coatings from carbon fiber-reinforced polymer (CFRP), we propose real-time monitoring for laser-layered paint removal. Current methods for laser paint removal on CFRP surfaces primarily focus on temperature control to safeguard the CFRP against potential damage, yet encounter challenges in providing real-time monitoring capabilities. In this study, we present laser-induced breakdown spectroscopy (LIBS) combined with partial least-squares discriminant analysis (PLS-DA) models as a promising approach. Initially, in this study, we analyze the elemental composition of carbon fiber substrates, primer, and topcoat to identify key characteristic elements for evaluating the laser-layered paint removal effectiveness. Subsequently, we explore changes in the intensities of characteristic spectral lines associated with the characteristic elements in different layers. Lastly, we develop PLS-DA models to effectively identify and classify the carbon fiber substrates, primer, and topcoat, enabling real-time monitoring of laser-layered paint removal. Based on the measured LIBS characteristic intensities and PLS-DA models, we accurately identified materials using Al I (396.164 nm) and Cr I (428.984 nm), or exclusively Cr I (428.984 nm), with 100% accuracy. The results demonstrate the feasibility of integrating LIBS with PLS-DA for monitoring laser-layered paint removal and show its potential in high-quality surface cleaning and automation.