Abstract
AbstractGlobal warming caused by greenhouse gas emissions has become a major challenge facing people all over the world. The study of regional human activities and their impacts on carbon emissions is of great significance to achieve the ambitious goal of carbon neutrality and sustainable economic development. Guizhou Province is a typical karst area in China, and its energy consumption is mainly based on fossil fuels.Therefore, it is necessary to predict and analyze its carbon emissions. In this paper, BP neural network and extreme learning machine (ELM) model, which have the advantage of nonlinear processing, will be used to predict the carbon emissions of Guizhou Province from 2020 to 2040. Based on the energy consumption data of Guizhou Province, the carbon emissions of Guizhou Province are calculated by using the conversion method and the inventory compilation method. The data show that the carbon emissions of Guizhou Province show an “S” growth trend; In this paper, 12 influencing factors of carbon emissions are selected, and five influencing factors with larger correlation are screened out by using grey correlation analysis method, and the prediction model of carbon emissions in Guizhou Province is established and simulated, and the prediction performance of BP neural network, ELM and WOA-ELM are compared respectively. Compared with ELM model and BP neural network model, the prediction accuracy of WOA-ELM model is higher; Finally, three development scenarios of carbon emissions are set by scenario analysis, which are baseline scenario, high-speed scenario and low-carbon scenario. On this basis, the size and time of peak carbon emissions in Liaoning Province from 2020 to 2040 are predicted based on WOA-ELM model. The results show that the peak value of carbon dioxide in the low carbon scenario is up to 0.98 million tons 31294 in 2033, the peak value of carbon emissions in the high speed scenario is up to 0.37 million tons 30251 in 2036, and the peak value of carbon emissions in the baseline scenario is up to 0.61 million tons 26243 in 2038. Based on the peak time and prediction results of carbon emissions under the three scenarios, the main factors contributing to the reduction of carbon emissions in Guizhou Province are analyzed, and important data basis is provided for energy conservation and emission reduction in Guizhou Province.
Publisher
Cold Spring Harbor Laboratory
Reference53 articles.
1. Carbon emission intensity in electricity production: A global analysis[J];Energy Policy,2016
2. Determinants of CO2 emissions in Brazil and Russia between 1992 and 2011 A decomposition analysis[J];Energy Policy,2016
3. Analysis of energy related carbon dioxide emission and reduction potential in Pakistan[J];Journal of Cleaner Production,2017
4. The decomposition of CO2 emissions from energy use in Greece before and during economic and their decoupling from economic growth[J];Renewable and Sustainable Energy Reviews,2017
5. LMDI Decomposition Analysis of Energy Consumption in the Korean Manufacturing Sector[J];Sustainability,2017