Composition Analysis and Identification of Glass Products Based on Hierarchical Clustering

Abstract

Ancient glass is an important carrier of the development of Chinese civilization. Affected by the buried environment of cultural relics, ancient glass is very easy to weather. A large number of internal elements of glass exchange with environmental elements, resulting in changes in its composition and appearance, thus affecting the judgment of its correct classification. In this paper, a comprehensive evaluation model was established to analyze the chemical composition of glass. First of all, the chi-square test is used to determine the factors related to surface weathering. Then the chemical composition of different glass was analyzed by mathematical statistics, and the number of clustering categories was determined by the method of system clustering, combined with the determination of square Euclidean distance and elbow rule. Finally, the final classification result is obtained by judging the size of the grey correlation degree of the type classification law and the sub-classification division method obtained in the model. The results show that the comprehensive model can be used to analyze and identify the composition of glass products.