China Energy Construction Group Corporation Limited
China Energy Construction Group Hunan Electric Power Design Institute Co., Ltd.

“碳达峰”和“碳中和”是事关全人类的一场社会性变革，如何准确的预测和降低碳排放是这场变革胜利的关键。为准确预测能源领域的碳排放量，采用多目标优化算法，以预测过程中的电力装机结构和能源消费结构成本最小化、结构更加清洁化以及碳排放量最小化为目标，基于电力平衡、电量平衡和调峰平衡等能源安全考虑，采用一种混合测算模型对碳排放量预测过程中电力装机结构和能源消费结构进行优化，从而根据优化后的能源消费结构得到基于电力装机成本和能源消费结构成本最优的最小碳排放值。整个预测过程主要分成3个部分，首先根据LMDI分解方法解析能源领域碳排放影响因素，可得GDP对碳排放的正向驱动效应明显，能源消费强度和产业结构对碳排放的负向驱动效应明显；其次，对碳系统结构进行初步优化、基于初步优化后的结果预测初始碳排放值；最后，基于过程的合理性和准确性，给出了一种模型评估方法，采用关键指标验证方法来对预测过程中的电能占终端能源消费占比、非化石能源消费占比和碳强度累计下降率等约束性指标进行反馈验证，经关键指标验证和多重反馈调整，最终得到满足多目标函数约束和各约束性指标的碳排放值。采用中部某省的实际数据对优化和预测过程的可行性和可靠性进行了验证，结果表明，仅在达峰场景下，该省能满足达峰目标以及通过3个关键指标验证。

“Carbon Peak” and “carbon neutrality” are a social revolution affecting all mankind. How to accurately predict and reduce carbon emissions is the key to the success of this revolution, and also gives the research on this topic a more difficult mission. In order to accurately predict the carbon emissions in the energy sector, based on the energy security considerations such as power balance, power amount balance and peak balancing, a hybrid calculation model is adopted to optimize the installed structure of electric power and energy consumption structure in the process of carbon emission prediction, so as to obtain the optimal minimum carbon emission value based on the installed cost of electric power and the cost of energy consumption structure according to the optimized energy consumption structure. The whole forecasting process is mainly divided into three parts. Firstly, the influencing factors of carbon emissions in the energy sector are analyzed according to the LMDI decomposition method. GDP has a positive driving effect on carbon emissions, while energy consumption intensity and industrial structure have a negative driving effect on carbon emissions. Secondly, the structure of the carbon system is preliminarily optimized and the initial carbon emission value is predicted based on the preliminary optimization results. Finally, based on the rationality and accuracy of the process, a model evaluation method is proposed. The key index verification method is adopted to perform feedback verification on the binding indicators such as the proportion of electric energy in terminal energy consumption, the proportion of non-fossil energy consumption and the cumulative decline rate of carbon intensity in the prediction process. After the key index verification and multiple feedback adjustment, the carbon emission value meeting the constraints of the multi-objective function and each constraint index is obtained. The feasibility and reliability of the optimization and prediction process are verified by using the actual data of a province in central China. The results show that only in the peak reaching scenario, the province can meet the target of carbon peak and pass the verification of three key indicators.

中国电力工程顾问集团有限公司重大科技专项资助项目(KW9-01-2022)

罗必雄，张毅，张力，等. 能源领域碳系统结构优化和碳排放预测方法与实例验证[J]. 煤炭学报，2023，48(7)：2657−2667.

LUO Bixiong，ZHANG Yi，ZHANG Li，et al. Carbon system structure optimization and carbon emission prediction method and case verification in energy field[J]. Journal of China Coal Society，2023，48(7)：2657−2667.