Video-Based Prediction of In-Flight Particle Characteristics in Atmospheric Plasma Spraying
Researchers have developed a method using high-speed video to predict particle characteristics in atmospheric plasma spraying (APS). This technique aims to non-invasively monitor particle temperature and velocity, which are crucial for coating quality. Various machine learning models, including TabPFN and CNNs, were evaluated, with pretrained CNNs achieving the highest accuracy in predicting both temperature and velocity directly from video frames. AI
IMPACT Enables real-time process monitoring and quality control in industrial coating applications.