Optimizing energy consumption structure in Chongqing of China to achieve low-carbon and sustainable development based on compositional data

Abstract

The optimal allocation of energy consumption structure has a vital impact on the sustainable development of economy and environment. However, the joint prediction and systematic analysis of the overall energy consumption structure are faced with great challenges due to its inherent structural characteristics. In this paper, an energy consumption structure prediction method based on dimension reduction through hyperspherical transformation and composite quantile regression neural network (DRHT-CQRNN) for compositional data is innovatively developed, which can reveal the relative information behind the absolute value of energy consumption. On this basis, the energy consumption structure of Chongqing of China during the 14th Five-Year Plan period (2021–2025) is predicted, and relevant policy recommendations are proposed to promote the transformation of the energy consumption structure of Chongqing. The research results show that by 2025, the consumption structure of coal, natural gas, petroleum and others will be 45.53%, 21.79%, 21.44% and 11.23%, respectively. The proportion of coal consumption in Chongqing will continue to decline, while the share of low-carbon, high-efficiency, high-quality energy will gradually increase. This paper demonstrates the excellent performance of the compositional data combination prediction model in terms of energy structure, which is also important for the formulation of energy structure policy in Chongqing.