Analysis of chemical composition of ancient glass products based on multiple linear regression

Abstract

During the Silk Road period, glass was introduced to China, and China began to produce ancient glass with Chinese characteristics by absorbing and developing its technology. However, ancient glass products is very vulnerable to weathering in the buried environment, and many components of glass change accordingly during the weathering process. Analyzing the changes of various factors and components of ancient glass before and after weathering is of great significance to the exploration and study of ancient glass. Therefore, this paper obtains the relationship between weathering on the surface of glass artifacts and glass type, decoration and color by establishing a multiple linear regression model, and obtains the statistical law of chemical composition content with and without weathering on the surface of glass samples through data analysis: weathering on the surface of glass artifacts has a strong correlation with glass type and decoration, but a weak correlation with glass color. The predicted chemical composition content before weathering was carried out by the weighted average method, and the prediction table as well as the scatter plot were obtained. From the data, it can be seen that most of the total chemical composition of each predicted sample fluctuated above and below 100%, and the average value of all samples was obtained as 98.05%, which meets the requirement of valid data.