Class classification of glass products based on K-means clustering

Abstract

The main raw material of glass is quartz sand, due to the high melting point of pure quartz sand, to reduce the melting temperature, it is necessary to add fluxes during refining. Different fluxes are added, and their main chemical composition is also different. To accurately classify high potassium glass and lead-barium glass, this paper establishes a K-means clustering model to classify glass types. High potassium glasses were classified into three subclasses by the content of iron oxide and barium oxide and lead barium glasses were classified into three subclasses by the content of magnesium oxide and barium oxide. Sensitivity analysis was performed on the model and the results remained significantly correlated.