江苏省碳排放预测和控制研究.PDFVIP

2010 The Second China Energy Scientist Forum The Research on Prediction and Control of Carbon Emission for Jiangsu Province Xiaofang Wang, Yangwei Tao, Mei Sun Center for Energy Development and Environmental Protection, Jiangsu University, 212013,Zhenjiang, China E-mail: sunm@ujs.edu.cn Abstract: With the rapid economic development in recent years, increasing energy consumption, as well as the relatively low energy efficiency and the coal-focused energy consumption structure, causes the carbon emission of Jiangsu Province increase continued. Predicting the trend of carbon emission and analyzing the impact factors of carbon emission is extremely important for mitigation strategy and the development of low carbon economy. This paper analyzes the impact factors of carbon emission and establishes an improved BP neural network model according to the statistical data in 1995~2007. Through comparing this model with the principal component regression model, the result shows that the improved BP neural network has better pre- dictive effect. So carbon emission in next three years is predicted through the model, it’s predicted that the carbon emission is increasing. Finally, in 2010, for example, controlling for other factors constant, the influ- ence of energy efficiency, industrial structure and energy consumption to carbon emission is analyzed, it will provide the basic for the development of low-carbon economic for Jiangsu province. Key words: carbon emission; factors; principal component regression; improved BP neural network; predic- tion; factors analysis 江苏省碳排放预测与控制研究 王小芳，陶阳威，孙梅 江苏大学能源发展与环境保护战略研究中心，镇江，中国，212013