Electricity-carbon modeling of flat glass industry based on correlation variable

Abstract

The flat glass industry is a typical industry with high energy consumption and extensive carbon emission. The carbon emission of flat glass industry in China ranks first in the same industry in the world. At present, there are few researches on carbon emission prediction for industrial enterprises, especially for the flat glass industry, due to lack of monitoring data. This paper selects electricity consumption as the influencing factor of carbon emission. This paper firstly preprocesses the electricity consumption data according to the China greenhouse gas emission standard. Next, this paper selects a correlation variable to fit the historical data of the flat glass industry based on Support Vector Regression (SVR). Finally, it obtains the parameters of the basic form of the electricity consumption to carbon emission model of the industry. The validity of the electricity-carbon modeling method and the accuracy of the model are verified by simulation experiments. The electricity-carbon model established by the method in this paper has a high accuracy rate, and the coefficient of determination (R2) reaches 0.98, which can play an auxiliary role in the verification of carbon emissions.

Keywords

• Flat glass industry
• Electricity-carbon model
• Carbon emission prediction
• Support Vector Regression (SVR)