Electric Power ›› 2026, Vol. 59 ›› Issue (2): 71-80.DOI: 10.11930/j.issn.1004-9649.202510069

• New-Type Power Grid • Previous Articles     Next Articles

Method for quantifying the average carbon footprint of coal-fired power based on non-random samples

WANG Zhixuan1(), ZHANG Jingjie1(), SHI Lina1, FENG Tianfeng2, WANG Chenlong1,3, DU Xinxin1,4, LEI Yuwei1, GU Erxue5   

  1. 1. China Electricity Council, Beijing 100761, China
    2. Electric Power Marketing Company of National Energy Group, Beijing 100011, China
    3. Electric Power Development Research Institute Co., Ltd. of China Electricity Council, Beijing 100162, China
    4. China Huadian Engineering Co., Ltd., Beijing 100160, China
    5. National Institute of Clean-and-Low-Carbon Energy, Beijing 102211, China
  • Received:2025-10-24 Revised:2025-12-31 Online:2026-03-04 Published:2026-02-28
  • Supported by:
    This work is supported by Smart Grid National Science and Technology Major Proiect (No.2025ZD0807900).

Abstract:

To address the challenge of selecting representative units for quantifying the average carbon footprint per unit of electricity generation of coal-fired units nationwide or in a specific region, and to provide scientific support for the quantification of the average carbon footprint factors of coal-fired power across the country, it is necessary for the research to focus on the representativeness analysis method of non-random samples for the overall carbon footprint of coal-fired power. As the dominant source of carbon emissions in China's power industry (accounting for approximately 88%), coal-fired power, due to its large installed capacity and complex influencing factors, makes analyzing overall characteristics through representative samples a feasible approach. Firstly, it is clarified that the core links of coal-fired power carbon footprint are coal combustion (accounting for 93.0%) and coal acquisition (accounting for 6.5%), jointly contributing to over 99% of carbon emissions. The coal consumption level and the carbon emission factor of power coal are the essential factors affecting the carbon footprint. Secondly, three types of methods for constructing new representative samples based on over a hundred existing quantitative samples are proposed, including indirectly proving representativeness through consistency verification of key parameters, supplementing missing samples by multi-dimensional stratification, and conducting weighted resampling to match the overall distribution. Finally, 171 sample datasets are generated by combining existing data. The deviations between their power generation coal consumption (286.9 g/(kW·h)) and carbon content per unit calorific value (26.39 t/TJ) with the corresponding indicators of the overall population composed of 1964 power plants (286.7 g/(kW·h) and 26.28 t/TJ) are only –0.07% and –0.415%, respectively, verifying the effectiveness of the proposed methods.

Key words: coal-fired power generation, carbon footprint, non-random samples, mathematical analysis