Driving factors analysis and scenario prediction of CO2 emissions in power industries of key provinces along the Yellow River based on LMDI and BP neural network

Abstract

IntroductionPower industry is one of the largest sources of CO2 emissions in China. The Yellow River Basin plays a supportive role in guaranteeing the effective supply of electricity nationwide, with numerous power generation bases. Understanding the drivers and peak of CO2 emissions of power industry in the Yellow River Basin is vital for China to fulfill its commitment to reach carbon emissions peak by 2030.MethodsThe Logarithmic Mean Divisia Index (LMDI) model was employed to explore the drivers to the change of CO2 emissions in power industries of three study areas, including Inner Mongolia Autonomous Regions, Shanxi Province, and Shandong Province in the Yellow River Basin. And Back Propagation (BP) neural network was combined with scenario analysis to empirically predict the trend of the amount of CO2 emitted by power industry (CEPI) from provincial perspective.ResultsCEPI in Inner Mongolia under the scenarios of a low degree of CO2 emissions promotion with a medium degree of CO2 emissions inhibition (LM) and a low degree of CO2 emissions promotion with a high degree of CO2 emissions inhibition (LH) scenario can reach a peak as early as 2030, with the peak value of 628.32 and 638.12 million tonnes, respectively. Moreover, in Shanxi, only CEPI under a low degree of CO2 emissions promotion scenarios (LL, LM, LH) can achieve the peak in 2025 ahead of schedule, with amounts of 319.32, 308.07, and 292.45 million tonnes. Regarding Shandong, CEPI under scenarios of a low degree of CO2 emissions promotion with a high degree of CO2 emissions inhibition (LH) and a medium degree of CO2 emissions promotion with a high degree of CO2 emissions inhibition (MH) could achieve the earliest peak time in 2025, with a peak of 434.6 and 439.36 million tonnes, respectively.DiscussionThe earliest peak time of CEPI in Shandong Province and Shanxi Province is 2025, but the peak of CEPI in Shanxi is smaller than that of Shandong. The peak time of CEPI in Inner Mongolia is relatively late, in 2030, and the peak is larger than that of the other two provinces. The per capita GDP is the most positive driving factor that contributes to the CEPI. Shandong has a strong economy, and its per capita GDP is much higher than Shanxi’s. Therefore, even under the same peak time, the CEPI in Shandong is much higher than that of Shanxi. Inner Mongolia is extensive and sparsely populated, which makes its per capita GDP rank among the top in China. In addition, Inner Mongolia’s coal-based power generation structure and high power generation also contribute to its late CO2 peak time and large CO2 peak.

Ultra-short-term prediction of microgrid source load power considering weather characteristics and multivariate correlation

“…FA serves as a multivariate statistical method that, by solving the correlation matrix of variables, identifies common factors describing relationships among numerous variables and simplifies data, thereby reducing the dimensionality of the dataset. The fundamental principles and computational procedures of FA are detailed in Wu et al (2024), wherein the basic model entails the linear relationship between observed variables and common factors as Eq. 5:…”

“…Addressing this limitation, this study investigates the simultaneous correlation between source and load power in a microgrid and weather features, conducting research on the joint ultra-short-term prediction of source and load power in a microgrid. Additionally, commonly used dimensionality reduction algorithms include Principal Component Analysis (PCA) (Wang et al, 2023), Independent Component Analysis (ICA) (Kobayashi and Iwai, 2018), Factor Analysis (FA) (Ramirez et al, 2019;Wu et al, 2024), etc. FA merges numerous features into several representative common factors to extract latent factors among features, accurately capturing the relevant information in the data (Zhou et al, 2020).…”