A Research Method for Estimating Coal Consumption and Supply Guarantee Considering Changes in Cross-Regional Energy Interconnection Structure
In the face of increasingly severe climate change, China has set the “Carbon Peaking and Carbon Neutrality” goals. Under the current energy structure, coal consumption still occupies a significant position. Accurate prediction of coal consumption not only helps in the rational planning of coal production, avoiding excessive mining and waste, but also provides scientific support for achieving the “Dual Carbon Goals”. Considering the forecasting needs for coal inventory to ensure energy and power security, this paper uses data from the China Electricity Council's monthly report from 2021 to 2024. This dataset aggregates provincial monthly electricity consumption, installed capacities, and output across energy types (thermal, hydro, wind, solar, nuclear), alongside utilization hours and interprovincial power trade records. By analyzing the changes in the national energy structure, this paper establishes an LSTM-based prediction model for the national thermal power generation, which is then used to predict national coal consumption. The model was tested and validated with a large national energy group, and the experimental findings demonstrate that the proposed model achieves a notable enhancement in the precision of coal consumption forecasting.
Authors
- Yi-Zhong Fan (ORCID: https://orcid.org/0000-0002-8966-6911)
- Yan Liang (ORCID: https://orcid.org/0009-0006-6837-1280)
- Zheng Shi (ORCID: https://orcid.org/0000-0001-7531-4299)
- Zhao Liu (ORCID: https://orcid.org/0000-0001-9545-3552)
- Hongli Liu (ORCID: https://orcid.org/0000-0003-1908-6644)
- Wei Ma (ORCID: https://orcid.org/0000-0001-8945-5877)
Institutions
- State Grid Corporation of China (China) (CN)
- Beijing Jiaotong University (CN)
Publication Details
- Published
- 2025-04-15
- DOI
- https://doi.org/10.1109/acpee64358.2025.11041223
- Primary Topic
- Coal and Coke Industries Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00