Fixed-variable weight combination prediction of China's energy consumption structure from the perspective of compositional data
Energy is an important material foundation for China's economic and social development,so effectively forecasting the trend of energy consumption structure is of great practical significance for achieving China's carbon peak and carbon neutrality goals.However,the energy consumption structure belongs to a class of complex statistics with a special spatial structure,whose main feature is that the proportion of each component is non negative and the total sum is constant at 1.Therefore,the theory of compositional data was applied to the study on energy consumption structure,aiming to fully satisfy the special internal characteristics of the energy consumption structure and fully explore its relative information.The single forecast models,such as BPNN,ARIMA,and triple exponential smoothing,were constructed respectively for each component under spherical coordinate translation;based on the minimization of the sum of squared Aitchison distance in the compositional data,the combined forecast models with fixed and variable weights were constructed respectively;the changing trend of Chinas energy consumption structure from 2024 to 2030 was forecasted by the variable-weighted combination model with the minimum average Aitchison distance value MSD.The forecast results showed that China's future energy consumption structure would develop toward a more diversified pattern,with an increasing trend of the proportion of clean energy consumption,and a still dominant position of fossil resources such as coal.
万方数据
维普
