毛俊晨,季亮,李胜文,常潇,李博通,李振坤
MAO Junchen,JI Liang,LI Shengwen,CHANG Xiao,LI Botong,LI Zhenkun

• 1:上海电力大学电气工程学院
• 2:国网山西省电力公司电力科学研究院
• 3:天津大学电气自动化与信息工程学院
• 4:教育部海上风电技术工程研究中心(上海电力大学)

摘要(Abstract)：

精准的负荷预测是保障综合能源系统经济可靠运行的关键，然而多元负荷的非平稳性、动态时变耦合性及高度随机性给负荷预测带来巨大挑战。为此，提出基于MVMD（多元变分模态分解）与动态混合图注意力网络的概率预测方法。首先，利用MVMD对原始负荷序列进行同步分解，使多元负荷间的跨序列相位同步并保持模态对齐，在平滑数据的同时保留了负荷间内在的时序耦合特性。其次，设计动态混合图注意力网络，通过门控机制自适应融合物理先验静态图与气象驱动动态图，并结合时空图卷积挖掘高阶时变耦合特征。最后，通过分位数回归构建概率输出层，实现从点预测到概率分布预测的拓展。基于美国亚利桑那州立大学数据集的实验结果表明，该方法能够有效量化负荷不确定性，其预测性能显著优于多种基准模型。
Accurate load forecasting is essential for ensuring the economic and reliable operation of integrated energy systems(IESs). However, the nonstationarity, dynamically time-varying coupling, and high stochasticity of multivariate loads pose significant challenges to load forecasting. To address these issues, this paper proposes a probabilistic forecasting method based on multivariate variational mode decomposition(MVMD) and a dynamic hybrid graph attention network. First, MVMD is employed to synchronously decompose the original load sequences, enabling cross-series phase synchronization and modal alignment among multivariate loads. This process smooths the data while preserving the intrinsic temporal coupling characteristics between load series. Second, a dynamic hybrid graph attention network is designed. Through a gating mechanism, physical prior static graphs and a meteorology-driven dynamic graphs are adaptively fused, and spatiotemporal graph convolution is incorporated to extract higher-order time-varying coupling features. Finally, a quantile regression-based probabilistic output layer is constructed, extending conventional point forecasting to full probabilistic distribution forecasting. Experimental results based on the dataset from Arizona State University demonstrate that the proposed method effectively quantifies load uncertainty and significantly outperforms several benchmark models in forecasting performance.

关键词(KeyWords)： 综合能源系统;多元负荷;概率预测;多元变分模态分解;动态图注意力网络
IES;multivariate loads;probabilistic forecasting;MVMD;dynamic graph attention network

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金（52177108）

作者(Author): 毛俊晨,季亮,李胜文,常潇,李博通,李振坤
MAO Junchen,JI Liang,LI Shengwen,CHANG Xiao,LI Botong,LI Zhenkun

DOI: 10.19585/j.zjdl.202606001

参考文献(References)：

• [1]林顺富，应子涵，谭津，等.考虑供能风险的电-气综合能源系统优化规划方法[J].浙江电力，2025,44(11):14-24.LIN Shunfu,YING Zihan,TAN Jin,et al. An optimal planning method for integrated electricity-gas systems considering energy supply risks[J].Zhejiang Electric Power,2025,44(11):14-24.
• [2]叶琪超，楼可炜，张宝，等.多能互补综合能源系统设计及优化[J].浙江电力，2018,37(7):5-12.YE Qichao,LOU Kewei,ZHANG Bao,et al.Design and optimization of multi-energy complementary integrated energy system[J]. Zhejiang Electric Power,2018,37(7):5-12.
• [3]蔡昌春，何瑶瑶，石庆伦，等.考虑绿证-碳交易与需求响应的含氢IES低碳优化调度[J].电力工程技术，2025,44(3):43-52.CAI Changchun,HE Yaoyao,SHI Qinglun,et al. Lowcarbon optimal dispatching of hydrogen-containing IES considering joint green certificate-carbon emission trading and demand response[J]. Electric Power Engineering Technology,2025,44(3):43-52.
• [4]刘涛，王瑞琪，刘松，等.考虑碳-绿证交易与需求响应的综合能源系统合作博弈调度[J].电力需求侧管理，2024,26(6):8-15.LIU Tao,WANG Ruiqi,LIU Song,et al. Collaborative game scheduling of integrated energy systems considering carbon-green certificate trading and demand response[J].Power Demand Side Management,2024,26(6):8-15.
• [5]吴永孝，肖辉，曾林俊，等.考虑供需双边响应的综合能源系统优化调度[J].电力建设，2024,45(10):13-23.WU Yongxiao,XIAO Hui,ZENG Linjun,et al. Optimal scheduling of integrated energy systems considering bilateral response of supply and demand[J]. Electric Power Construction,2024,45(10):13-23.
• [6]陈霁，王冰，陈玉全，等.基于绿证-阶梯碳交易联合机制的综合能源系统多主体协同优化方法[J].智慧电力，2025,53(9):10-18.CHEN Ji,WANG Bing,CHEN Yuquan,et al. A multiagent collaborative optimization method for integrated energy systems based on green certificate-stepped carbon trading joint mechanism[J].Smart Power,2025,53(9):10-18.
• [7]王婷，张晶，高冲，等.绿证-碳市场互认抵消下综合能源系统优化调度[J].智慧电力，2025,53(2):104-110.WANG Ting,ZHANG Jing,GAO Chong,et al.Optimal scheduling of integrated energy system under mutual recognition and offset of green certificate and carbon market[J].Smart Power,2025,53(2):104-110.
• [8]张玉敏，孙鹏凯，孟祥剑，等.基于碳势-能源价格双响应的综合能源系统低碳经济调度[J].电力系统自动化，2024,48(9):21-33.ZHANG Yumin,SUN Pengkai,MENG Xiangjian,et al.Low-carbon economic dispatching of integrated energy system based on dual response of carbon intensity and energy price[J]. Automation of Electric Power Systems,2024,48(9):21-33.
• [9]WANG X,WANG S X,ZHAO Q Y,et al. A multienergy load prediction model based on deep multi-task learning and ensemble approach for regional integrated energy systems[J].International Journal of Electrical Power&Energy Systems,2021,126:106583.
• [10]SHARMA C.Modeling of an island grid[J].IEEE Transactions on Power Systems,1998,13(3):971-978.
• [11]程浩忠，胡枭，王莉，等.区域综合能源系统规划研究综述[J].电力系统自动化，2019,43(7):2-13.CHENG Haozhong,HU Xiao,WANG Li,et al.Review on research of regional integrated energy system planning[J].Automation of Electric Power Systems,2019,43(7):2-13.
• [12]KONG W C,DONG Z Y,JIA Y W,et al.Short-term residential load forecasting based on LSTM recurrent neural network[J].IEEE Transactions on Smart Grid,2019,10(1):841-851.
• [13]ABEDINIA O,LOTFI M,BAGHERI M,et al.Improved EMD-based complex prediction model for wind power forecasting[J].IEEE Transactions on Sustainable Energy,2020,11(4):2790-2802.
• [14]李杰，靳孟宇，马士豪.改进粒子群算法优化支持向量机的短期负荷预测[J].测控技术，2021,40(4):76-79.LI Jie,JIN Mengyu,MA Shihao. Power short-term load prediction based on support vector machine optimized by improved particle swarm optimization algorithm[J].Measurement&Control Technology,2021,40(4):76-79.
• [15]CHEN B J,CHANG M W,LIN C J.Load forecasting using support vector Machines:a study on EUNITE competition 2001[J]. IEEE Transactions on Power Systems,2004,19(4):1821-1830.
• [16]CEPERIC E,CEPERIC V,BARIC A. A strategy for short-term load forecasting by support vector regression machines[J]. IEEE Transactions on Power Systems,2013,28(4):4356-4364.
• [17]LI F Q,ZHANG S Y,LI W X,et al.Forecasting hourly power load considering time division:a hybrid model based on K-means clustering and probability density forecasting techniques[J].Sustainability,2019,11(24):6954.
• [18]BELOEV H I,SAITOV S R,FILIMONOVA A A,et al.Short-term electrical load forecasting based on XGBoost model[J].Energies,2025,18(19):5144.
• [19]WU K H,WU J,FENG L,et al.An attention-basedCNNLSTM-BiLSTMmodel for short-term electric load forecasting in integrated energy system[J]. International Transactions on Electrical Energy Systems,2021,31(1):e12637.
• [20]田浩含，张智晟，于道林.基于改进LSTM的区域综合能源系统多元负荷短期预测研究[J].电力系统及其自动化学报，2021,33(9):130-137.TIAN Haohan,ZHANG Zhisheng,YU Daolin. Research on multi-load short-term forecasting of regional integrated energy system based on improved LSTM[J].Proceedings of the CSU-EPSA,2021,33(9):130-137.
• [21]邹智，吴铁洲，张晓星，等.基于贝叶斯优化CNN-BiGRU混合神经网络的短期负荷预测[J].高电压技术，2022,48(10):3935-3945.ZOU Zhi,WU Tiezhou,ZHANG Xiaoxing,et al. Shortterm load forecast based on Bayesian optimized CNNBiGRU hybrid neural networks[J]. High Voltage Engineering,2022,48(10):3935-3945.
• [22]李卓，王胤喆，叶林，等.从感知-预测-优化综述图神经网络在电力系统中的应用[J].中国电力，2024,57(12):2-16.LI Zhuo,WANG Yinzhe,YE Lin,et al.The application of graph neural networks in power systems from perspective of perception-prediction-optimization[J]. Electric Power,2024,57(12):2-16.
• [23]HUANG B,WANG J H. Applications of physicsinformed neural networks in power systems a review[J].IEEE Transactions on Power Systems,2023,38(1):572-588.
• [24]ZHOU H,AI Q,LI R T. Short-term multi-energy load forecasting method based on transformer spatio-temporal graph neural network[J].Energies,2025,18(17):4466.
• [25]靳双龙，成丹，任永建.极端气温事件对多类型电力负荷的影响研究进展[J/OL].南方能源建设，2025:1-12.(2025-11-11)[2025-11-28]. https：//doi. org/10.16516/j.ceec.2025-270.JIN Shuanglong,CHENG Dan,REN Yongjian.Research progress on the impact of extreme temperature events on multiple types of electrical loads[J].Southern Energy Construction,2025:1-12.(2025-11-11)[2025-11-28].https：//doi.org/10.16516/j.ceec.2025-270.
• [26]RICHMAN J S,MOORMAN J R. Physiological timeseries analysis using approximate entropy and sample entropy[J].American Journal of Physiology Heart and Circulatory Physiology,2000,278(6):H2039-H2049.
• [27]刘昕明，吉建光，李玮，等.基于双模态分解的发电站母线短期负荷预测[J].电气工程学报，2024,19(1):124-132.LIU Xinming,JI Jianguang,LI Wei,et al.Short-term load forecasting for power substations based on bimodal decomposition[J]. Journal of Electrical Engineering,2024,19(1):124-132.
• [28]ROUSSEEUW P J.Silhouettes:a graphical aid to the interpretation and validation of cluster analysis[J].Journal of Computational and Applied Mathematics,1987,20:53-65.
• [29]DICKEY D A,FULLER W A.Distribution of the estimators for autoregressive time series with a unit root[J].Journal of the American Statistical Association,1979,74(366a):427-431.
• [30]AKAIKE H.A new look at the statistical model identification[M]//Selected Papers of Hirotugu Akaike. New York,NY:Springer New York,1998:215-222.
• [31]SCHWARZ G.Estimating the dimension of a model[J].The Annals of Statistics,1978,6(2):461-464.
• [32]YU B,YIN H T,ZHU Z X.Spatio-temporal graph convolutional networks:a deep learning framework for traffic forecasting[C]//Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence.July13-19,2018.Stockholm,Sweden.International Joint Conferences on Artificial Intelligence Organization,2018:3634-3640.
• [33]BAI S,KOLTER J Z,KOLTUN V.An empirical evaluation of generic convolutional and recurrent networks for sequence modeling[EB/OL].(2018-04-19)[2025-11-28].https：//arxiv.org/pdf/1803.01271.
• [34]ORESHKIN B N,CARPOV V,CHAPADOS N.Temporal convolutional networks for multivariate time series forecasting[C]//Proceedings of the 36th International Conference on Machine Learning(ICML). Long Beach,CA,USA:PMLR,2019:4007-4016.