%0 Journal Article
%T 古代玻璃制品的成分分析与鉴别<br>Analysis and Verification on Components of Ancient Glass Products
%A 陈月
%A 陈文俊
%A 刘海萍
%A 舒常江
%J Modeling and Simulation
%P 408-420
%@ 2324-870X
%D 2024
%I Hans Publishing
%R 10.12677/MOS.2024.131039
%X 本文基于所给数据，根据玻璃文物的基本信息和化学成分比例，利用主成分分析、聚类分析、灰色关联、BP神经网络等方法，通过建立模型的方法研究了古代玻璃制品的成分分析与鉴别问题，同时对玻璃文物的化学成分比例进行处理，分析各成分与古代玻璃类型的关系来预测不同化学成分的玻璃对应的类型和玻璃中各个化学成分之间的关系。完成了对高钾与铅钡玻璃的分类以及对未知类型的玻璃制品的鉴别，最后分析了不同类别之间的关联关系，算出关联度并比较两种类型的差异性。<br />
Based on the provided data, and according to the basic information and chemical component pro-portion of glass artifacts, this paper makes use of methods of principal component analysis, cluster analysis, grey correlation, BP neural network and others to study analysis and verification on an-cient glass products through modeling. At the same time, this paper processes the chemical com-ponent proportion to the glass artifacts to analyze the relationships between each component and the types of ancient glass, so as to predict the relationships between each chemical component and the type of glass corresponding to that chemical component. The classification of glasses with high potassium and lead barium and the verification on unknown glass products were completed. Finally, the association in different classifications was analyzed and the degree of association was calculated and the difference between both was compared.
%K 主成分分析，聚类分析，灰色关联，BP神经网络，敏感性分析<br>Principal Component Analysis
%K Cluster Analysis
%K Grey Correlation
%K BP Neural Network
%K Sensitivity Analysis
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=79417