[1]贾燕峰,俞 飞,李 飚,等.基于PCC提取负荷周期特征与时变趋势特征的精准短期负荷预测[J].机械与电子,2026,44(08):117-126.
　JIA Yanfeng,YU Fei,LI Biao,et al.Accurate Short-term Load Forecasting Based on PCC Extracted Load Cyclic Characteristics and Time-varying Trend Characteristics[J].Machinery & Electronics,2026,44(08):117-126.
点击复制

基于PCC提取负荷周期特征与时变趋势特征的精准短期负荷预测()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
44
期数:
2026年08期
页码:
117-126
栏目:
电力控制
出版日期:
2026-08-25

文章信息/Info

Title:
Accurate Short-term Load Forecasting Based on PCC Extracted Load Cyclic Characteristics and Time-varying Trend Characteristics
文章编号:
1001-2257(2026)08-0117-10
作者:
贾燕峰1; 俞 飞1; 李 飚1; 严亚帮2; 乔澳雪2; 季玉琦2
1.国网河南省电力公司三门峡供电公司,河南 三门峡 472000;
2.郑州轻工业大学电气信息工程学院,河南 郑州 450002
Author(s):
JIA Yanfeng1; YU Fei1; LI Biao1; YAN Yabang2; QIAO Aoxue2; JI Yuqi2
(1.Sanmenxia Provincial Electric Power Company,State Grid Henan Province Power Supply Company,Sanmenxia 472000,China;
2.School of Electrical and Information Engineering,Zhengzhou University of Light Industry,Zhengzhou 450002,China)
关键词:
时变趋势特征提取; 皮尔逊相关系数; CNN-LSTM 混合神经网络; 短期负荷预测
Keywords:
time-varying trend feature extraction; Pearson correlation coefficient; CNN-LSTM hybrid neural network; short-term load forecasting
分类号:
TM73
文献标志码:
A
摘要:
为提升负荷预测精度和预测模型的泛化能力,提出融合负荷周期特性与负荷时变趋势特性的卷积神经网络长短期记忆网络混合预测模型(C-T-CNN-LSTM)。首先,使用皮尔逊相关系数量化各采样时刻负荷序列之间的相关性强弱,筛选出可以反映电力负荷固有周期演化规律的关键特征参量。然后,基于通过分析周期负荷数据时序变化特点,提出负荷时变趋势概念,并建立负荷时变趋势模型,提取负荷时变趋势特征。最后,将负荷周期特征变量与负荷时变趋势特征变量重构生成特征集,并输入到CNN-LSTM 混合模型中进行预测。以美国新英格兰地区2015年电力负荷数据集为例,采用9种预测模型进行仿真验证与对比分析,结果表明,提出的C-T-CNN-LSTM 模型拥有最高的预测精度。
Abstract:
To enhance the forecasting accuracy and generalization capability of prediction models,this paper proposes a hybrid forecasting model(C-T-CNN-LSTM) that integrates load cyclic characteristics and time varying trend characteristics into a convolutional neural network and long short term memory (CNN-LSTM) network.Firstly,the Pearson correlation coefficient (PCC) is employed to quantify the correlation strength among load sequences at different sampling instants,thereby selecting key feature variables that can reflect the inherent periodic evolution patterns of electric power load.Subsequently,based on an analysis of the temporal variation characteristics of periodic load data,the concept of load time varying trend is introduced,and a corresponding model is established to extract the time varying trend features.Finally,the load cyclic feature variables and the time-varying trend feature variables are reconstructed to form a feature set,which is then fed into the CNN-LSTM hybrid model for prediction.Taking the 2015 electric load dataset of the New England region of the United States as a case study,nine prediction models are employed for simulation validation and comparative analysis.The results demonstrate that the proposed C-T-CNN-LSTM model achieves the highest prediction accuracy among all compared models.

参考文献/References:

[1] Gan Dahua,Wang Yi,Yang Shuo,et al.Embedding based quantile regression neural network for probabilistic load forecasting[J].Journal of Modern Power Systems and Clean Energy,2018,6(2):244-254.

[2] 黄冬梅,陈欢,王宁,等.基于自适应图注意力网络的短期用户负荷预测[J].电力系统保护与控制,2023,51(20):140-149.
[3] 孙浩,万灿,曹照静,等.基于条件生成对抗网络曲线生成的短期负荷概率预测[J].电力系统自动化,2023,47(23):189-199.
[4] Lin Weixuan,Wu Di,Boulet B.Spatial-temporal residential short-term load forecasting via graph neural networks[J].IEEE Transactions on Smart Grid,2021,12(6):5373-5384.
[5] 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.
[6] Li Yiyan,Han Dong,Yan Zheng.Long-term system load forecasting based on data-driven linear clustering method[J].Journal of Modern Power Systems and Clean Energy,2018,6(2):306-316.
[7] Jiang Huaiguang,Zhang Yingchen,Muljadi E,et al.A short-term and high-resolution distribution system load forecasting approach using support vector regression with hybrid parameters optimization[J].IEEE Transactions on Smart Grid,2016,9(4):3341-3350.
[8] Goehry B,Goude Y,Massart P,et al.Aggregation of multi-scale experts for bottom-up load forecasting[J].IEEE Transactions on Smart Grid,2019,11(3):1895-1904.
[9] 周思思,李勇,郭钇秀,等.考虑时序特征提取与双重注意力融合的TCN超短期负荷预测[J].电力系统自动化,2023,47(18):193-205.
[10] Tan Mao,Yuan Siping,Li Shuaihu,et al.Ultra-short-term industrial power demand forecasting using LSTM based hybrid ensemble learning[J].IEEE Transactions on Power Systems,2019,35(4):2937-2948.
[11] Wang Shouxiang,Wang Xuan,Wang Shaomin,et al.Bi-directional long short-term memory method based on attention mechanism and rolling update for short-term load forecasting[J].International Journal of Electrical Power and Energy Systems,2019,109:470-479.
[12] 李丹,孙光帆,缪书唯,等.基于多维时序信息融合的短期电力负荷预测方法[J].中国电机工程学报,2023,43(增刊1):94-106.
[13] 杨维熙,刘勇,舒勤.基于补充集合经验模态分解的短期负荷预测模型[J].电网技术,2022,46(9):3615-3623.
[14] 孙辉,杨帆,高正男,等.考虑特征重要性值波动的 MI-BILSTM 短期负荷预测[J].电力系统自动化,2022,46(8):95-103.
[15] 王炜,冯斌,黄刚,等.基于自注意力编码器和深度神经网络的短期净负荷预测[J].中国电机工程学报,2023,43(23):9072-9084.
[16] 白冰青,刘江涛,王旭,等.基于MRMR 和双重注意力机制的城市能源多元负荷短期预测[J].电力系统自动化,2022,46(17):44-55.
[17] Abedinia O,Amjady N,Zareipour H.A new feature selection technique for load and price forecast of electrical power systems[J].IEEE Transactions on Power Systems,2016,32(1):62-74.
[18] 任建吉,位慧慧,邹卓霖,等.基于CNN-BiLSTM-Attention的超短期电力负荷预测[J].电力系统保护与控制,2022,50(8):109-116.
[19] 欧阳福莲,王俊,周杭霞.基于改进迁移学习和多尺度 CNN-BiLSTM-Attention 的短期电力负荷预测方法[J].电力系统保护与控制,2023,51(2):132-140.
[20] 刘杰,从兰美,夏远洋,等.基于 DBO-VMD 和 IWOA-BILSTM 神经网络组合模型的短期电力负荷预测[J].电力系统保护与控制,2024,52(8):123-133.
[21] He Shengtao,Li Canbing,Liu Xubin,et al.A per-unit curve rotated decoupling method for CNN-TCN based day-ahead load forecasting[J].IET Generation, Transmission and Distribution,2021,15(19):2773-2786.
[22] Sadaei H J,Silva P C L,Guimaraes F G,et al.Short-term load forecasting by using a combined method of convolutional neural networks and fuzzy time series[J].Energy,2019,175:365-377.
[23] 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.
[24] Yi Shiyan,Liu Haichun,Chen Tao,et al.A deep LSTM-CNN based on self-attention mechanism with input data reduction for short-term load forecasting[J].IET Generation,Transmission and Distribution,2023,17(7):1538-1552.
[25] 蒙飞,单连飞,卢峰,等.基于输入更新长短期记忆网络的调度自适应学习模型[J].电力系统自动化,2022,46(24):26-35.
[26] Shang Chuan,Gao Junwei,Liu Huabo,et al.Short-term load forecasting based on PSO-KFCM daily load curve clustering and CNN LSTM model[J]. IEEE Access,2021,9:50344-50357.
[27] Al Ja’afreh M A A,Mokryani G,Amjad B.An enhanced CNN-LSTM based multi stage framework for PV and load short term forecasting:DSO scenarios[ J].Energy Reports,2023,10:1387-1408.
[28] 季玉琦,严亚帮,和萍,等.基于K-Medoids聚类与栅格法提取负荷曲线特征的CNN-LSTM 短期负荷预测[J].电力系统保护与控制,2023,51(18):81-93.
[29] 李滨,高枫.基于虚拟相似日与DA-LSTPNet的地区电网短期负荷预测[J].电力系统自动化,2021,45(22):55-64.
[30] 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.
[31] 徐先峰,赵依,刘状壮,等.用于短期电力负荷预测的日负荷特性分类及特征集重构策略[J].电网技术,2022,46(4):1548 -1556.
[32] Liu Yang,Wang Wei,Ghadimi N.Electricity load forecasting by an improved forecast engine for building level consumers[J].Energy,2017,139:18-30.
[33] Choi J H.Investigation of the correlation of building energy use intensity estimated by six building performance simulation tools[J].Energy and Buildings, 2017,147:14-26.

备注/Memo

备注/Memo:
收稿日期:2026-05-07
基金项目:国网河南省电力公司科技项目资助(5217I0250004)
作者简介:贾燕峰 (1978-),男,河南太康人,高级工程师,研究方向为电网数字化技术与电网规划设计;季玉琦 (1989-),男,河南范县人,博士,副教授,硕士研究生导师,研究方向为新型配电系统规划与优化运行,通信作者,E-mail:jiyuqi1989@163.com。
更新日期/Last Update: 2026-08-28