基于系统聚类算法和Fisher判别模型的古代玻璃成分分析与类别鉴定

Composition Analysis and Category Identification of Ancient Glass Based on Sys-tematic Clustering Algorithm and Fisher Discriminant Model

古代玻璃体现了我国古代高度发达的工艺技术,新时代,玻璃成为了人们生产生活的必需品,研究古代玻璃有助于加深人们对玻璃的认识,唤醒我国人民对工艺研究的热情。本文对古代玻璃的实验数据进行分析,首先使用描述性统计分析和卡方检验分析研究玻璃文物样品表面有无风化化学成分含量的统计规律,得到玻璃表面风化与类型的关系;然后依据玻璃表面风化与类型的关系建立系统聚类算法模型,研究不同类别玻璃文物样品的化学成分之间的关联关系;最后依据化学成分之间关联关系建立Fisher判别模型,得到玻璃类别鉴定的Fisher判别公式,可以对未知类别的玻璃文物进行类别鉴定。本文对古代玻璃的成分分析与类别鉴定提供一定的借鉴意义。

The ancient glass reflected the highly developed technology of Chinese ancient times. In the new era, glass has become a necessity of people’s production and life. The study of ancient glass is helpful to deepen people’s understanding of glass and awaken the enthusiasm of Chinese people for the re-search of craft. In this paper, the experimental data of ancient glass are analyzed. Firstly, descrip-tive statistical analysis and Chi-square test were used to study the statistical rule of weathering chemical components on the surface of glass cultural relics samples, and the relationship between weathering and types of glass surface was obtained. Then, according to the relationship between glass surface weathering and glass types, a systematic clustering algorithm model was established to study the correlation between the chemical components of different types of glass cultural relics samples. Finally, the Fisher discriminant model was established according to the correlation be-tween chemical components, and the Fisher discriminant formula of glass category identification was obtained, which can be used for category identification of glass relics of unknown category. This paper provides some reference for the composition analysis and category identification of ancient glass.