Research on net carbon emissions, influencing factor analysis, and model construction based on a neural network model in the BTH region

• Published in: Journal of renewable and sustainable energy Vol. 14; no. 6
• Main Authors: Ren, Feng, Guo, Mengyuan
• Format: Journal Article
• Language: English
• Published: Melville American Institute of Physics 01.11.2022
• Subjects: Back propagation networks; Carbon; Emissions control; Energy utilization; Factor analysis; Mathematical models; Neural networks; Search algorithms; Urbanization; Beijing China; China
• ISSN: 1941-7012; 1941-7012
• DOI: 10.1063/5.0099474

Scientific prediction of net carbon emissions in the Beijing–Tianjin–Hebei (BTH) region is of significance to examine carbon emission reduction in the context of the “double carbon” target. In this study, the carbon peak and carbon neutrality states in the BTH region are determined through a regional double carbon target analysis framework, and a logistic chaotic sparrow search algorithm backpropagation neural network hybrid model (LCSSA-BP) optimized by a logistic chaotic sparrow search algorithm (LCSSA) is used to forecast the net carbon emissions. The findings reveal that the net carbon emissions in the BTH region generally increased during the study period; Beijing, Tianjin, and Hebei are in different stages toward the realization of the double carbon target; population size, affluence, and urbanization rate are positively correlated, whereas the proportion of foreign direct investment and energy intensity is negatively associated with net carbon emissions; the prediction accuracy of the LCSSA-BP is superior to that of the SSA-BP and BP neural network, and it can be used to forecast the net carbon emissions in the BTH region.