Composition analysis and identification of ancient glass products

Abstract

Based on the proportion of various components of ancient glass products and the classification method, the composition, change and relation of different types of cultural relics under different weathering conditions are analyzed in detail in this paper. The statistical rules and the basis for the classification of cultural relics are given, and on this basis, the unknown types of glass cultural relics are identified. Firstly, a statistical model is established, and Chi-square test is used to analyze the factors that affect the weathering of cultural relics. It is concluded that the weathering of cultural relics is related to the type of glass and the type of decoration, and the type of glass has a greater influence on the weathering. The chemical composition of the weathered cultural relics before weathering is predicted. Then, after the classification rule of glass is preliminarily explored, the grey correlation model is established to calculate the correlation degree between each chemical component and the two kinds of glass, and then the K-means clustering model is used to divide the glass into subclasses. Thirdly, BP neural network was used to analyze the chemical composition of glass relics of unknown category, identify their types, and analyze the sensitivity of classification results. Finally, according to the idea of discrete Frechet distance algorithm, the cultural relics are grouped according to the type of glass, and the discrete Frechet distance of the broken line of the change rate of the content of different chemical components between cultural relics is used as the basis to judge the correlation relationship, and the correlation relationship between the two substances is analyzed.