1.北京科东电力控制系统有限责任公司,北京市 100192
2.国网电力科学研究院有限公司(南瑞集团有限公司),江苏省南京市 211106
何蕾(1978—),女,高级工程师,主要研究方向:电力系统调度自动化、电力市场、综合能源。E-mail:helei6@sgepri.sgcc.com.cn
耿闯(1994—),男,通信作者,硕士,工程师,主要研究方向:人工智能、时序预测、电力市场。E-mail:gengchuang@bupt.cn
包铁(1984—),男,硕士,工程师,主要研究方向:电力系统调度自动化、电力市场。E-mail:bt0504@126.com
收稿:2026-02-25,
修回:2026-07-16,
网络首发:2026-09-22,
纸质出版:2026-11-10
移动端阅览
何蕾,耿闯,包铁等.基于时序基础模型知识迁移与多元特征融合的负荷预测方法[J].电力系统自动化,2026,50(21):185-193.
HE Lei,GENG Chuang,BAO Tie,et al.Load Forecasting Method Based on Knowledge Transfer of Time-series Foundation Models and Multivariate Feature Fusion[J].Automation of Electric Power Systems,2026,50(21):185-193.
何蕾,耿闯,包铁等.基于时序基础模型知识迁移与多元特征融合的负荷预测方法[J].电力系统自动化,2026,50(21):185-193. DOI: 10.7500/AEPS20260225002.
HE Lei,GENG Chuang,BAO Tie,et al.Load Forecasting Method Based on Knowledge Transfer of Time-series Foundation Models and Multivariate Feature Fusion[J].Automation of Electric Power Systems,2026,50(21):185-193. DOI: 10.7500/AEPS20260225002.
高比例新能源并网与气象条件变化加剧引发电力负荷剧烈波动,传统模型已难以捕捉负荷序列中的深层次特征。时序基础模型依托预训练获得的可迁移知识,能够有效捕捉负荷序列中的复杂模式,然而,基础模型在协变量信息的利用上存在局限。为了进一步提升负荷预测精度,文中提出一种将基础模型与协变量信息深度融合的范式。首先,采用负荷数据对时序基础模型微调,将其知识迁移至电力负荷预测领域;然后,设计一个通用的外生特征嵌入模块,该模块可与任意基础模型灵活集成,补充基础模型外部信息利用上的不足;最后,基于电网实测负荷数据和全球能源预测竞赛数据验证了所提方法的有效性。
The high penetration of renewable energy and the intensifying variability of meteorological conditions have led to severe fluctuations in power load
making it difficult for traditional models to capture the deep-level features within load sequences. Time-series foundation models
leveraging transferable knowledge acquired through pre-training
are capable of effectively capturing the complex patterns in load sequences. However
foundation models still have limitations in utilizing covariate information. To further improve load forecasting accuracy
this paper proposes a paradigm that deeply integrates foundation models with covariate information. First
a time-series foundation model is fine-tuned using load data
transferring its knowledge to the field of power load forecasting. Then
a general exogenous feature embedding module is designed
which can be flexibly integrated with any foundation model to compensate for the insufficiency of foundation models in utilizing external information. Finally
the effectiveness of the proposed method is verified based on actual load data from a power grid and data from the Global Energy Forecasting Competition.
孔祥玉 , 马玉莹 , 艾芊 , 等 . 新型电力系统多元用户的用电特征建模与用电负荷预测综述 [J]. 电力系统自动化 , 2023 , 47 ( 13 ): 2 - 17 .
Kong Xiangyu , Ma Yuying , Ai Qian , et al . Review on electricity consumption characteristic modeling and load forecasting of power consumption for diverse users in new power system [J]. Automation of Electric Power Systems , 2023 , 47 ( 13 ): 2 - 17 .
张兴华 , 鲁南 , 王阳 , 等 . 新型电力系统下的负荷预测挑战与展望 [J]. 电网技术 , 2026 , 50 ( 6 ): 2231 - 2243 .
Zhang Xinghua , Lu Nan , Wang Yang , et al . Challenges and opportunities of load forecasting under new power systems [J]. Power System Technology , 2026 , 50 ( 6 ): 2231 - 2243 .
Contreras J , Espinola R , Nogales F J , et al . ARIMA models to predict next-day electricity prices [J]. IEEE Transactions on Power Systems , 2003 , 18 ( 3 ): 1014 - 1020 .
Nepal B , Yamaha M , Yokoe A , et al . Electricity load forecasting using clustering and ARIMA model for energy management in buildings [J]. Japan Architectural Review , 2020 , 3 ( 1 ): 62 - 76 .
Al-Hamadi H M , Soliman S A . Short-term electric load forecasting based on Kalman filtering algorithm with moving window weather and load model [J]. Electric Power Systems Research , 2004 , 68 ( 1 ): 47 - 59 .
Zhao H , Guo S . An optimized grey model for annual power load forecasting [J]. Energy , 2016 , 107 : 272 - 286 .
Abd Jalil N A , Ahmad M H , Mohamed N . Electricity load demand forecasting using exponential smoothing methods [J]. World Applied Sciences Journal , 2013 , 22 ( 11 ): 1540 - 1543 .
陈纬楠 , 胡志坚 , 岳菁鹏 , 等 . 基于长短期记忆网络和LightGBM组合模型的短期负荷预测 [J]. 电力系统自动化 , 2021 , 45 ( 4 ): 91 - 97 .
Chen Weinan , Hu Zhijian , Yue Jingpeng , et al . Short-term load prediction based on combined model of long short-term memory network and light gradient boosting machine [J]. Automation of Electric Power Systems , 2021 , 45 ( 4 ): 91 - 97 .
于永进 , 鉴奕霖 , 刘琪 , 等 . 基于多尺度并行特征融合与Kolmogorov-Arnold网络重构的短期电力负荷预测 [J]. 电力系统保护与控制 , 2026 , 54 ( 12 ): 176 - 187 .
Yu Yongjin , Jian Yilin , Liu Qi , et al . Short-term power load forecasting based on multi-scale parallel feature fusion and Kolmogorov-Arnold network reconstruction [J]. Power System Protection and Control , 2026 , 54 ( 12 ): 176 - 187 .
Zhou H , Zhang S , Peng J , et al . Informer: beyond efficient transformer for long sequence time-series forecasting [J]. Proceedings of the AAAI Conference on Artificial Intelligence , 2021 , 35 ( 12 ): 11106 - 11115 .
肖白 , 王安睿 , 杜彬斌 , 等 . 基于可解释人工智能的空间电力负荷预测方法 [J ]. 电力系统自动化 , 2026 , 50 ( 12 ): 180 - 191 .
Xiao Bai , Wang Anrui , Du Binbin , et al . Spatial power load forecasting method based on explainable artificial intelligence [J]. Automation of Electric Power Systems , 2026 , 50 ( 12 ): 180 - 191 .
Liu Y , Hu T G , Zhang H R , et al . iTransformer: inverted transformers are effective for time series forecasting [C]// The Twelfth International Conference on Learning Representations , 2024 .
李楠 , 金淳熙 , 陶亮 , 等 . 基于多尺度二次特征提取的短期电力负荷预测模型 [J]. 电力系统保护与控制 , 2025 , 53 ( 19 ): 114 - 126 .
Li Nan , Jin Chunxi , Tao Liang , et al . A short-term electric load forecasting model based on multi-scale secondary feature extraction [J]. Power System Protection and Control , 2025 , 53 ( 19 ): 114 - 126 .
赵俊华 , 文福拴 , 黄建伟 , 等 . 基于大语言模型的电力系统通用人工智能展望:理论与应用 [J]. 电力系统自动化 , 2024 , 48 ( 6 ): 13 - 28 .
Zhao Junhua , Wen Fushuan , Huang Jianwei , et al . Prospect of artificial general intelligence for power systems based on large language model: theory and applications [J]. Automation of Electric Power Systems , 2024 , 48 ( 6 ): 13 - 28 .
孙秋野 , 张瑞霞 , 陈东岳 . 大模型技术赋能电力系统的应用及技术路线展望 [J]. 电力系统自动化 , 2026 , 50 ( 13 ): 3 - 24 .
Sun Qiuye , Zhang Ruixia , Chen Dongyue . Applications and technical route prospects of large model technology in empowering power systems [J]. Automation of Electric Power Systems , 2026 , 50 ( 13 ): 3 - 24 .
Vaswani A , Shazeer N , Parmar N , et al . Attention is all you need [C]// 31st Annual Conference on Neural Information Processing Systems (NeurIPS 2017 ), 2017 .
Jin M , Wang S , Ma L , et al . Time-LLM: time series forecasting by reprogramming large language models [C]// The Twelfth International Conference on Learning Representations , 2024 .
Ansari A F , Stella L , Turkmen A C , et al . Chronos: learning the language of time series [PP/OL]. [ 2026-01-14 ]. https://arxiv.org/abs/2403.07815 https://arxiv.org/abs/2403.07815 .
Das A , Kong W , Sen R , et al . A decoder-only foundation model for time-series forecasting [C]// The Forty-first International Conference on Machine Learning , 2024 .
SHI X M , WANG S Y , NIE Y Q , et al . Time-MoE: billion-scale time series foundation models with mixture of experts [C]// The Thirteenth International Conference on Learning Representations , 2025 .
Liu X , Liu J , Woo G , et al . Moirai-MOE: empowering time series foundation models with sparse mixture of experts [PP/OL]. [ 2026-01-14 ]. https://arxiv.org/abs/2410.10469 https://arxiv.org/abs/2410.10469 .
Dai D , Deng C , Zhao C , et al . DeepSeekMoE: towards ultimate expert specialization in mixture-of-experts language models [C]// Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics , 2024 : 1280 - 1297 .
Arango S P , Mercado P , Kapoor S , et al . ChronosX: adapting pretrained time series models with exogenous variables [C]// 28th International Conference on Artificial Intelligence and Statistics , 2025 : 2242 - 2250 .
Garza A , Challu C , Mergenthaler-Canseco M . TimeGPT-1 [PP/OL]. [ 2026-01-14 ]. https://arxiv.org/abs/2310.03589 https://arxiv.org/abs/2310.03589 .
Ansari A F , Shchur O , Küken J , et al . Chronos-2: from univariate to universal forecasting [PP/OL]. [ 2026-01-14 ]. https://arxiv.org/abs/2510.15821 https://arxiv.org/abs/2510.15821 .
Liu Y , Qin G , Huang X , et al . Timer-XL: long-context transformers for unified time series forecasting [C]// The Thirteenth International Conference on Learning Representations , 2025 .
Raffel C , Shazeer N , Roberts A , et al . Exploring the limits of transfer learning with a unified text-to-text transformer [J]. Journal of Machine Learning Research , 2020 , 21 ( 1 ): 5485 - 5551 .
Nie Y , Nguyen N H , Sinthong P , et al . A time series is worth 64 words: long-term forecasting with transformers [C]// The Eleventh International Conference on Learning Representations , 2023 .
Tolstikhin I O , Houlsby N , Kolesnikov A , et al . MLP-mixer: an all-MLP architecture for vision [J]. Advances in Neural Information Processing Systems , 2021 , 34 : 24261 - 24272 .
Chen S A , Li C L , Arik S O , et al . TSMixer: an all-MLP architecture for time series forecasting [PP/OL]. [ 2026-01-14 ]. https://arxiv.org/abs/2303.06053 https://arxiv.org/abs/2303.06053 .
张昆明 , 蔡珊珊 , 章天晗 , 等 . 考虑多维时域特征的行业中长期负荷预测方法 [J]. 电力系统自动化 , 2023 , 47 ( 20 ): 104 - 114 .
Zhang Kunming , Cai Shanshan , Zhang Tianhan , et al . Medium- and long-term industry load forecasting method considering multi-dimensional temporal features [J]. Automation of Electric Power Systems , 2023 , 47 ( 20 ): 104 - 114 .
Salinas D , Flunkert V , Gasthaus J , et al . DeepAR: probabilistic forecasting with autoregressive recurrent networks [J]. International Journal of Forecasting , 2020 , 36 ( 3 ): 1181 - 1191 .
Lim B , Arık S Ö , Loeff N , et al . Temporal fusion transformers for interpretable multi-horizon time series forecasting [J]. International Journal of Forecasting , 2021 , 37 ( 4 ): 1748 - 1764 .
Wang Y , Wu H , Dong J , et al . TimeXer: empowering transformers for time series forecasting with exogenous variables [J]. Advances in Neural Information Processing Systems , 2024 , 37 : 469 - 498 .
Herzen J , Lässig F , Piazzetta S G , et al . Darts: user-friendly modern machine learning for time series [J]. Journal of Machine Learning Research , 2022 , 23 ( 124 ): 1 - 6 .
Hong T , Pinson P , Fan S . Global energy forecasting competition 2012 [J]. International Journal of Forecasting , 2014 , 30 ( 2 ): 357 - 363 .
Hong T , Pinson P , Fan S , et al . Probabilistic energy forecasting: global energy forecasting competition 2014 and beyond [J]. International Journal of Forecasting , 2016 , 32 ( 3 ): 896 - 913 .
0
浏览量
1
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621