ORIGINAL RESEARCH
Analysis of the Influence Mechanism of Energy-Related Carbon Emissions with a Novel Hybrid Support Vector Machine Algorithm in Hebei, China
Jingmin Wang 1, 2
,  
Fan Yang 1, 2  
,  
 
 
More details
Hide details
1
Department of Economics and Management, North China Electric Power University, Baoding, China
2
Beijing Key Laboratory of New Energy and Low-Carbon Development (North China Electric Power University), Changping, Beijing, China
CORRESPONDING AUTHOR
Fan Yang   

North China Electric Power University, Department of Economics and Management, North China Electric Power University,, 689 Huadian Road, Baoding 071000, China, 071000 Baoding, China
Online publish date: 2019-03-25
Submission date: 2018-07-04
Final revision date: 2018-08-06
Acceptance date: 2018-08-14
 
KEYWORDS
TOPICS
ABSTRACT
The Beijing-Tianjin-Hebei Region (BTH) as a national strategic highland attracts attention with its haze problem. In particular, Hebei is a major emitter of carbon emissions in BTH. The establishment of the Xiong’an New District in Hebei, known as the “Millennium plan,” faces complex and diverse development in the future, so the carbon emission prediction and influence mechanism are of great significance. This paper has made two improvements to the particle swarm optimization algorithm (PSO), then the improved algorithm is used to optimize parameters of the traditional support vector machine (SVM). Therefore, a new model, IPSO-SVM, is established. This paper uses the STIRPAT model to determine the impact factors, through 64 predict scenarios of 2017-2020 to reveal that economic growth is the most important factor of carbon emissions in Hebei, followed by resident population, industrial structure, urbanization level, energy structure, and technical level. In the case of positive economic development, the contribution of technology to carbon reduction will increase. Based on the “new normal,” Hebei ought to develop sustainable urbanization and emphasis on the role of technology in low-carbon development to control carbon emissions.
eISSN:2083-5906
ISSN:1230-1485