Analysis and identification of ancient glass components based on statistical methods

Abstract

Ancient glass products are prone to weathering, resulting in the exchange of elements, which affects the identification of glass products. In order to solve the problem of determining the significance of surface weathering related factors and predicting the chemical composition before and after weathering, chi-square test was performed after data preprocessing, and the P-value was used to determine whether the relevant factors were significant according to the cross-thermal map.Then, with the help of the bar chart, the characteristics of the data before and after weathering of high potassium and lead and barium were visualized, and the relevant statistical rules were obtained by comparison.According to the mean values of weathering points before and after weathering, the prediction formula was constructed to calculate the content of chemical components before and after weathering, and the rationality was tested.