[1]章 俊,梅 飞,刘宇航,等.基于 BiLSTM-Attention-ERM 层级修正模型的配电网母线净负荷预测[J].机械与电子,2026,44(05):111-119.
　ZHANG Jun,MEI Fei,LIU Yuhang,et al.Net Load Forecasting of Distribution Network Bus Based on BiLSTM-Attention-ERM Hierarchical Correction Model[J].Machinery & Electronics,2026,44(05):111-119.
点击复制

基于 BiLSTM-Attention-ERM 层级修正模型的配电网母线净负荷预测()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
44
期数:
2026年05期
页码:
111-119
栏目:
电力控制
出版日期:
2026-05-27

文章信息/Info

Title:
Net Load Forecasting of Distribution Network Bus Based on BiLSTM-Attention-ERM Hierarchical Correction Model
文章编号:
1001-2257 ( 2026 ) 05-0111-09
作者:
章 俊; 梅 飞; 刘宇航; 李 玟; 戴 翔
河海大学电气与动力工程学院,江苏 南京 211100
Author(s):
ZHANG Jun ;  MEI Fei ;  LIU Yuhang ;  LI Wen ;  DAI Xiang
( School of Electrical and Power Engineering , Hohai University , Nanjing 211100 , China )
关键词:
净负荷预测; 层级修正; BiLSTM-Attention 模型;  ERM 算法
Keywords:
net load forecasting ;  hierarchical correction ;  BiLSTM-Attention model ;  ERM algorithm
分类号:
TM615
文献标志码:
A
摘要:
针对分布式光伏因缺少实测气象数据导致出力预测偏差,进而影响母线净负荷预测精度的问题,提出一种结合光伏出力换算与分层预测修正的配电网母线净负荷预测方法。首先,筛选与分布式光伏出力相关性高的基准电站,结合气象数据和聚类算法确定天气类型,并根据历史出力比值计算最优换算系数。接着,构建双向长短期记忆网络( BiLSTM )结合 Attention 机制直接预测基准电站的光伏出力、母线负荷及净负荷,并通过换算系数推导分布式光伏出力预测结果。最后,结合分层预测和经验风险最小化( ERM )构建综合修正模型,将直接预测结果输入模型进行净负荷预测结果修正。以我国南方某地区真实数据作为实际算例,结果表明,在缺乏精确气象数据的配电网 10 kV 母线净负荷预测中,所提方法的预测准确率相比未换算修正的方法有明显提升。
Abstract:
To address the issue of distributed photovoltaic output prediction deviation caused by the lack of measured meteorological data , which in turn affects the accuracy of bus net load forecasting , a method that combines PV output conversion and hierarchical forecasting correction is proposed for distribution network bus net load forecasting.First , a reference power station with high correlation to the distributed PV output is selected.Weather types are determined by combining meteorological data with a clustering algorithm , and the optimal conversion coefficient is calculated based on historical output ratios.Next , a bidirectional long short term memory ( BiLSTM ) Attention model is constructed to directly forecast the PV output , bus load , and net load of the reference power station.The distributed PV output prediction result is then derived using conversion coefficients.Finally , a comprehensive correction model is built by integrating hierarchical prediction and empirical risk minimization ( ERM ) .The direct prediction results are fed into this model to correct the net load forecast.Case studies using real data from southern China show that the proposed method significantly improves the accuracy of 10 kV bus net load forecasting in distribution networks lacking precise meteorological data , compared to methods without conversion and correction.

参考文献/References:

[ 1 ] 车凯,朱进,金沫含,等 . 考虑负荷耦合影响的农场配电网光伏功率短期修正预测[ J ] . 供用电, 2025 , 42 ( 3 ): 76-85.

[ 2 ] 陈宇星,梁芙蓉,尤炜,等 . 含分布式光伏的配电网模型预测控制优化方法[ J ] . 电力工程技术,2023 , 42 ( 6 ):100-109.
[ 3 ] Faustine A , Nunes N J , Pereira L.Efficiency through simplicity : MLP-based approach for net-load forecasting with uncertainty estimates in low-voltage distribution networks [ J ] .IEEE Transactions on Power Systems , 2025 , 40 ( 1 ): 46-56.
[ 4 ] 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.
[ 5 ] 杨小龙,姚陶,孙辰军,等 . 计及分布式能源时序不确 性的短期负荷预测技术[J ] . 可再生能源, 2024 , 42 ( 1 ):96-103.
[ 6 ] 厉瑜,益西措姆,杜宁刚,等 . 基于多因素组合分析的电力系统长期负荷预测研究[ J ] . 电网与清洁能源, 2024 , 40 ( 7 ): 81-87.
[ 7 ] Shu Xingsheng , Ding Wei , Peng Yong , et al.Value of long-term inflow forecast for hydropower operation : a case study in a low forecast precision region [ J ] .Energy , 2024 , 298 : 131218.
[ 8 ] Mamun A A , Sohel M , Mohammad N , et al.A comprehensive review of the load forecasting techniques using single and hybrid predictive models [ J ] .IEEE Access , 2020 , 8 : 134911-134939.
[ 9 ] 肖白,于龙泽,刘洪波,等 . 基于生成虚拟净负荷的多能源电力系统日前优化调度[ J ] . 中 国 电机 工程 学报,2021 , 41 ( 21 ): 7237-7249.
[ 10 ] 李冉,王明强,杨明,等 . 考虑故障概率和净负荷不确定性的 鲁 棒 随 机 备 用 优 化 [ J ] . 电 力 系 统 自 动 化,2022 , 46 ( 6 ): 20-29.
[ 11 ] 苏常胜,王森,孙谊媊,等 . 面向网络节点的新能源高占比系统消纳阻力精细化评估[ J ] . 电力工程技术,2022 , 41 ( 5 ): 67-75.
[ 12 ] Alipour M , Aghaei J , Norouzi M , et al.A novel electrical net-load forecasting model based on deep neural networks and wavelet transform integration [ J ] . Energy , 2020 , 205 : 118106.
[ 13 ] Wu Qinghua , Bose A , Singh C , et al.Control and stability of large-scale power system with highly distributed renewable energy generation : viewpoints from six aspects [ J ] .CSEE Journal of Power and Energy Systems , 2023 , 9 ( 1 ): 8-14.
[ 14 ] Tziolis G , Livera A , Montes-Romero J , et al.Direct short term net load forecasting based on machine learning principles for solar-integrated microgrids [ J ] .IEEE Access , 2023 , 11 : 102038-102049.
[ 15 ] Poecke A V , Tabari H , Hellinckx P.Unveiling the backbone of the renewable energy forecasting process : exploring direct and indirect methods and their applications [ J ] .Energy Reports , 2024 , 11 : 544-557.
[ 16 ] Sreekumar S , Khan N U , Rana A S , et al.Aggregatednet-load forecasting using Markov-Chain Monte Carlo regression and C-vine copula [ J ] .Applied Energy , 2022 , 328 : 120171.
[ 17 ] Kaur A , Nonnenmacher L , Coimbra C F M.Net load forecasting for high renewable energy penetration grids [ J ] .Energy , 2016 , 114 : 1073-1084.
[ 18 ] Zheng Chaoran , Eskandari M , Li Ming , et al.GA-reinforced deep neural network for net electric load forecasting in microgrids with renewable energy resources for scheduling battery energy storage systems [ J ] .Algorithms , 2022 , 15 ( 10 ): 338.
[ 19 ] Kobylinski P , Wierzbowski M , Piotrowski K.High resolution net load forecasting for micro-neighbourhoods with high penetration of nenewable energy sources [ J ] .International Journal of Electrical Power and Energy Systems , 2020 , 117 : 105635.
[ 20 ] 刘友波,吴浩,刘挺坚,等 . 集成经验模态分解与深度学习的用户侧净负荷预测算法[ J ] . 电力系统自动化,2021 , 45 ( 24 ): 57-64.
[ 21 ] 张静,熊国江 . 考虑季节特性与数据窗口的短期光伏功率预测组合模型[ J ] . 电力工程技术, 2025 , 44 ( 1 ):183-192.
[ 22 ] Jebli I , Belouadha F Z , Kabbaj M I , et al.Prediction of solar energy guided by pearson correlation using machine learning [ J ] .Energy , 2021 , 224 : 120109.
[ 23 ] Yang Zonglin , Jiang Shurong , Yu Fusheng , et al.Linear fuzzy information-granule-based fuzzy c-means algorithm for clustering time series [ J ] .IEEE Transactions on Cybernetics , 2023 , 53 ( 12 ): 7622-7634.
[ 24 ] Liao Chengwu , Chen Chao , Xiang Chaocan , et al.Taxi passenger ’ s destination prediction via GPS embedding and attention-based BiLSTM model [ J ] .IEEE Transactions on Intelligent Transportation Systems , 2022 , 23 ( 5 ): 4460-4473.
[ 25 ] Taieb S B , Koo B.Regularized regression for hierarchical forecasting without unbiasedness conditions [ C ] ∥Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , 2019 : 1337-1347.
[ 26 ] Chung W , Chen Yongtong.A nonparametric least squares regression method for forecasting building energy performance [ J ] .Applied Energy , 2024 , 376 : 124219.
[ 27 ] 朱继忠,苗雨旺,董朝阳,等 . 基于 Attention-LSTM与多模型集成的短期负荷预测方法[ J ] . 电力工程技术,2023 , 42 ( 5 ): 138-147.
[ 28 ] Hyndman R J , Ahmed R A , Athanasopoulos G , et al. Optimal combination forecasts for hierarchical time series [ J ] .Computational Statistics and Data Analysis , 2011 , 55 ( 9 ): 2579-2589.
[ 29 ] Wickramasuriya S L , Athanasopoulos G , Hyndman R J.Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization [ J ] . Journal of the American Statistical Association , 2018 , 114 ( 526 ): 804-819.
[ 30 ] Liu Lin , Cao Xiaojing , Wang Hengsheng , et al.Optimization of model parameters and hyperparameters in deep learning models for spatial interaction prediction [ J ] .Expert Systems with Applications , 2025 , 266 : 126160.
[ 31 ] Xie Ling.The heat load prediction model based on BP neural network markov model [ J ] .Procedia Computer Science , 2017 , 107 : 296-300.

备注/Memo

备注/Memo:
收稿日期: 2026-01-22
基金项目:国家重点研发计划项目( 2022YFE0140600 );国家自然科学基金资助项目( U24A20149 )
作者简介:章 俊 ( 2000- ),男,江苏南通人,硕士研究生,研究方向为电力负荷预测;梅 飞 ( 1982- ),男,安徽滁州人,博士,高级工程师,研究方向为电气设备在线监测与故障诊断。
更新日期/Last Update: 2026-08-25