Wu Liangzheng, Li Kaiman, Huang Yan, Wan Zhengdong, Tan Jieren
Energy Development Research Institute, China Southern Power Grid, Guangzhou, China.
South China University of Technology, Guangzhou, China.
PLoS One. 2025 Jan 3;20(1):e0316537. doi: 10.1371/journal.pone.0316537. eCollection 2025.
This study intends to optimize the carbon footprint management model of power enterprises through artificial intelligence (AI) technology to help the scientific formulation of carbon emission reduction strategies. Firstly, a carbon footprint calculation model based on big data and AI is established, and then machine learning algorithm is used to deeply mine the carbon emission data of power enterprises to identify the main influencing factors and emission reduction opportunities. Finally, the driver-state-response (DSR) model is used to evaluate the carbon audit of the power industry and comprehensively analyze the effect of carbon emission reduction. Taking China Electric Power Resources and Datang International Electric Power Company as examples, this study uses the comprehensive evaluation method of entropy weight- technique for order preference by similarity to ideal solution (TOPSIS). China Electric Power Resources Company has outstanding performance in promoting renewable energy, with its comprehensive evaluation index rising from 0.5458 in 2020 to 0.627 in 2022, while the evaluation index of Datang International Electric Power Company fluctuated and dropped to 0.421 in 2021. The research conclusion reveals the actual achievements and existing problems of power enterprises in energy saving and emission reduction, and provides reliable carbon information for the government, enterprises, and the public. The main innovation of this study lies in: using artificial intelligence technology to build a carbon footprint calculation model, combining with the data of International Energy Agency Carbon Dioxide (IEA CO2) emission database, and using machine learning algorithm to deeply mine the important factors in carbon emission data, thus putting forward a carbon audit evaluation system of power enterprises based on DSR model. This study not only fills the blank of carbon emission management methods in the power industry, but also provides a new perspective and basis for the government and enterprises to formulate carbon emission reduction strategies.
本研究旨在通过人工智能（AI）技术优化电力企业的碳足迹管理模型，以助力科学制定碳排放减排策略。首先，建立基于大数据和人工智能的碳足迹计算模型，然后运用机器学习算法深入挖掘电力企业的碳排放数据，以识别主要影响因素和减排机会。最后，采用驱动-状态-响应（DSR）模型对电力行业的碳审计进行评估，并综合分析碳排放减排效果。以中国电力资源公司和大唐国际电力公司为例，本研究采用熵权-逼近理想解排序法（TOPSIS）的综合评价方法。中国电力资源公司在促进可再生能源方面表现突出，其综合评价指标从2020年的0.5458升至2022年的0.627，而大唐国际电力公司的评价指标则出现波动，在2021年降至0.421。研究结论揭示了电力企业节能减排的实际成效和存在的问题，并为政府、企业和公众提供了可靠的碳排放信息。本研究的主要创新点在于：利用人工智能技术构建碳足迹计算模型，结合国际能源署二氧化碳（IEA CO2）排放数据库的数据，并运用机器学习算法深入挖掘碳排放数据中的重要因素，从而提出基于DSR模型的电力企业碳审计评价体系。本研究不仅填补了电力行业碳排放管理方法的空白，也为政府和企业制定碳排放减排策略提供了新的视角和依据。
