China’s Energy Demand Prediction for 2019-2035 based on the MPSO-BP Model
Download as PDF
DOI: 10.23977/EMCG2020.028
Author(s)
Haoze Jiang, Wanyang Zuo, Xin Yu
Corresponding Author
Haoze Jiang
ABSTRACT
Energy demand prediction is of great important to ensure national energy security and maintain steady economic growth. In this paper, a model based on BP neural network is set up to predict energy demand of China from 2019 to 2035 under three different scenarios. To improve the accuracy of the prediction, the modified PSO algorithm is used to optimize BP neural network. To measure the real situation of China's economic operation, we construct Keqiang-new Keqiang index as the descriptive index of economic development. The energy demand is analyzed for the period from 1985 to 2018 based on Keqiang-new Keqiang index, industrialization level, population and urbanization level. The results show that compared with BP neural network, the MPSO-BP model has better simulative and predictive precision and the energy demand of China under different scenarios will be 7.11×109, 4.83×109 and 6.18×109tce, respectively. By analyzing the prediction results, we can conclude that China can achieve the result of slow growth of energy demand or even making it maintain the current level by reducing the energy consumption of the industry while maintaining steady economic growth.
KEYWORDS
China’s energy demand prediction, BP neural network,