Composition Analysis and Identification of Ancient Glass Products

Glass is a valuable proof of the initial transaction. Ancient glass buried underground is easily weathered by the environment, leading to changes in its composition ratio and affecting its classification. This article studies the composition analysis and identification of ancient Chinese glass products. Firstly, multiple correspondence analysis was used to study the correlation of surface (weathered or unweathered), decoration, type, and color. It was found that whether the glass surface is weathered is not related to type, but decoration and color, type and color are all related. Through non parametric testing, it was found that there are significant differences in the mean values of components such as silicon dioxide, potassium oxide, barium oxide, lead oxide, and calcium oxide among the four different categories of glass. Secondly, principal component analysis was performed separately on the composition for potassium and lead barium glass, and classify them into subcategories based on their main components, calcium oxide and lead oxide, and the rationality of the results was verified through cluster analysis. Then, Fisher discrimination, Bayesian discrimination, and principal component based Mahalanobis distance discrimination were used to predict the type of unknown glass samples, and the majority voting method was used to determine the type of unknown samples. Finally, for different categories of glass samples, Pearson correlation coefficient was used to analyze the correlation between chemical components. By comparison, it was found that there were significant differences in the correlation coefficients of certain components before and after weathering, indicating that surface weathering caused changes in the composition, which in turn led to changes in the correlation between components.