Composition analysis and identification of ancient glass products
DOI:
https://doi.org/10.54097/hset.v40i.6792Keywords:
Ancient glass products, composition analysis, least squares method, BP neural network.Abstract
Aiming at the problem of composition analysis and identification of ancient glass products, a multiple linear regression model based on least square method was established, and the relationship between surface weathering and type, ornamentation and color was obtained.Firstly, the data were preprocessed, and the invalid data were eliminated by BP neural network, and the basic information of glass relics was quantified.Secondly, a multiple linear regression model was established, and the least square method was used to obtain that the surface weathering was negatively correlated with ornamentation, and positively correlated with color and type.Secondly, a component correlation model based on Spearman's coefficient was established to analyze the statistical rules of chemical components.Finally, JB test was used to verify whether the components obey the normal distribution, and the normal distribution test was completed.
Downloads
References
ZHAO Hualing. Study on the Development of Ancient Glass History in Zibo Area [D]. Soochow University, 2008, 1.
ZHU Rongjie. Study on spatial-temporal prediction and pollution warning model of river water quality based on improved BP neural network [D]. Harbin Institute of Technology, 2021, 9 - 12.
FU S. Research on security evaluation of cryptographic algorithms based on side channel analysis [D]. Beijing University of Posts and Telecommunications, 2020, 38.
Yu Qun, Huo Xiaodong, He Jian, Li Lin, Zhang Jianxin, Feng Yuyao. [4] YU Q, HE J, Li L, FENG Y Y. Trend prediction of power outage accidents in China power grid based on Spearman correlation coefficient and system inertia [J/OL]. Proceedings of the CSEE :1-12, 2022 - 09 - 16.
Zhang Yuanning, Huang Jingguang, Zhang Yanhua, Yu Nan, Chen Yong, Weng Hanli, Jia Yongbo, Li Zhenxing. [5] High Order Statistical Principle of Identifying transformer inrush Current with Dual Characteristic Quantity [J]. Protection and Control of Electric Power Systems, 2021, 49 (24): 21 - 30.
Zhou Xiuya. Descriptive analysis of factors influencing the high suicide rate in Nunavut, Canada, 2001-2011[D]. Beijing Foreign Studies University,2015. 2019, 07: 51 - 52.]
WANG S. Research and application of time series prediction algorithm based on deep learning [D]. University of Electronic Science and Technology of China, 2022, 10 - 12.
JIANG Na. Research on Performance Degradation Prediction and Maintenance Strategy of Turbofan Engine [D]. Harbin Institute of Technology, 2020, 90.
Che Yujian. Research on Health State Prediction of Power Battery of Pure Electric Vehicle [D]. Jilin University, 2022, 56 - 60.
YU Weiheng. Research on retrieving large earthquake rupture scale based on Fisher linear identification method [D]. Institute of Engineering Mechanics, China Earthquake Administration, 2016, 9 - 11.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
