Study on the correlation degree of chemical elements of glass based on grey theory

In order to classify glass samples with known chemical composition, this paper adopts K-Means clustering analysis algorithm. By distinguishing the main element content of the two kinds of glass, we have made a subclassification under the major category. The high potassium glass is divided into high silica low potassium oxide group and low silica high potassium oxide group. Lead barium glass can be divided into low barium oxide low phosphorus pentoxide group and high barium oxide high phosphorus pentoxide group. Therefore, after the content of various elements of various glass is obtained, the grey correlation analysis model is used to make the data of each chemical component as the parent sequence and the other chemical components as the sub-sequence to solve the grey correlation coefficient of pair chemical components successively, so as to further analyze the correlation between their chemical components. Therefore, after the completion of the subclassification, it is helpful for archaeologists to distinguish the glass cultural relics after weathering. With the analysis of grey relational degree, the composition relationship of ancient glass relics can be further explored.