.png){kind=icon}
.png){kind=icon}
Return
Forecasting China’s energy-related carbon emissions under digital transformation: a scenario-based modeling framework
.png){kind=icon}
PRE
AI
M
M
H
DOI：10.1016/j.seta.2026.105183.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Accurately forecasting energy-related carbon emissions is crucial for China’s decarbonization. This study develops a comprehensive framework to project China’s energy-related emissions, integrating key drivers such as industrial structure, economic development, energy mix, and digitalization into optimized predictive models. Five policy pathways are simulated: industrial empowerment, economic development, energy transition, digital empowerment, and accelerated digital empowerment. In the short term (2023–2030), the accelerated digital and energy transition scenarios yield the lowest emissions. After peaking around 2030–2031, emissions begin to decline across all scenarios in the medium term (2031–2040), with the fastest reductions occurring under the accelerated digital and energy transition paths. In the long term (post-2040), the accelerated digital empowerment scenario achieves the deepest mitigation, with emissions projected to fall to approximately 2.0 Gt CO2 by 2060 compared to 9.83 Gt in the baseline, underscoring the critical role of digital economy development for long-term carbon reduction. The results demonstrate that combining industrial upgrading, economic adjustments, energy transition, and digital acceleration enable sustained decarbonization, with accelerated digitalization offering the most pronounced long-term benefits for achieving carbon neutrality.
Keywords:
Low-carbon energy transition
Machine learning
Decarbonization
China
GDPSO
,
Golden Ratio–based Sine-Cosine Particle Swarm Optimizer
GPR
,
Gaussian Process Regression
GRIME
,
Generalized RIME with Mapping and Inverse Learning Enhancement
GRNN
,
Generalized Regression Neural Network
GRU
,
Gated Recurrent Unit
GSSA
,
Generalized Sine–Cosine Enhanced Sparrow Search Algorithm
GWO
,
Grey Wolf Optimizer
IOA
,
Ivy Optimization Algorithm
KOA
,
Kepler Optimization Algorithm
LSTM
,
Long Short-Term Memory
MAE
,
Mean Absolute Error
MAPE
,
Mean Absolute Percentage Error
RMSE
,
Root Mean Squared Error
SVM
,
Support Vector Machine
WSO
,
White Shark Optimization
Journal
IF:
7
Papers:
4.4K
Citations:
2.2W
Organization
Cited Papers