Exploring the regional characteristics of inter-provincial CO2 emissions in China: An improved fuzzy clustering analysis based on particle swarm optimization

The better to explore the regional characteristics of inter-provincial CO2 emissions and the rational distribution of the reduction of emission intensity reduction in China, this paper proposes an improved PSO-FCM clustering algorithm. This method can obtain the optimal cluster number and membership grade values by utilizing the global capacity of Particle Swarm Optimization (PSO) on Fuzzy C-means (FCM). The clustering results of CO2 emissions indicate that the 30 provinces of China are divided into five clusters and each has its own significant characteristics. Compared with other clustering methods, the results of PSO-FCM are more explanatory. The most important indicators affecting regional emission characteristics are CO2 emission intensity and per capita emissions, whereas CO2 emission per unit of energy is not obvious in clustering. Furthermore, some policy recommendations on setting emission reduction targets according to the emission characteristics of different clusters are made.

► An improved PSO-FCM clustering algorithm has been proposed for the characteristics of CO2 emissions.
► Thirty provinces of China are divided into five clusters by proposed PSO-FCM method.
► CO2 intensity and per capita emissions are the most important indicators affecting characteristics.
► To set CO2 emission reduction targets, the order of the five categories have been arranged.