基于Attention机制的CNN‑GRU配网线路重过载短期预测方法
CSTR:
作者:
作者单位:

(1.上海电力大学电气工程学院,上海 200090;2.国网上海市电力公司电力科学研究院,上海 200080;3.国网新疆电力有限公司电力科学研究院,新疆 乌鲁木齐 830002)

通讯作者:

胡钟毓(1995—),女,硕士研究生,主要从事电力大数据和人工智能技术研究;E?mail:huzhongyu@mail.shiep.edu.cn

中图分类号:

TM726

基金项目:

国家自然科学基金(51907114);上海电力人工智能工程技术研究中心研究项目(19DZ2252800)


Short‑term heavy overload forecasting method of distribution net line based on CNN‑GRU with Attention mechanism
Author:
Affiliation:

(1.Electric Power Engineering of Shanghai University of Electric Power, Shanghai 200090, China; 2.State Grid Shanghai Electrical Power Research Institute, Shanghai 200080, China; 3.State Grid Xinjiang Electrical Power Research Institute, Urumqi 830002, China)

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [22]
  • |
  • 相似文献
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    随着用户用电需求增加,迎峰度夏期间配网线路重过载较严重,为电网运行增加安全隐患。实现配网线路重过载短期预测,对合理安排负荷高峰时期运行方式和调度管理以及线路的安全运行具有重要意义。文中提出一种基于注意力(Attention)机制的卷积神经网络(CNN)—门限循环单元神经网络(GRU)组合预测模型。结合高相关性时间段的历史线路负载率数据和气象因素作为输入特征,利用CNN处理多源数据并提取有效特征作为GRU的输入,再通过GRU对时序特征集进行分析预测,利用Attention机制对重要数据分配更多的注意力权重,实现配网线路负载率的回归预测,最后根据负载等级划分标准将负载率预测结果转化为负载等级。使用所提方法对上海市某区某10 kV线路数据进行实验。实验结果表明,该预测方法比相同模型结构但以负载等级为输入的重过载分类预测,更适用于配网线路重过载预测。

    Abstract:

    With the increase of electricity demand, the heavy overload of distribution network lines during the peak period of electricity consumption becomes more serious, which increases the potential threats on the safety of grid operation. The short?term forecast of the heavy overload state of distribution lines is of great significance for rationally arranging the operation mode, for dispatch management, and for the safety operation of the line during peak load periods. This paper proposes a short?term forecast method for the heavy overload state of lines and a prediction model that CNN?GRU hybrid neural network with Attention mechanism. The historical load rate of lines with high auto?correlation and meteorological factors are combined as the input features, which is further used to extract the valid features by the CNN. The GRU neural network is utilized to analyze and predict time series data. By using the Attention mechanism to reassign corresponding weights, the load rate regression prediction result can be outputed,which can be finally converted into the load level according to the load level division standard. The method in this paper is performed on a 10kV line in a certain district of Shanghai. The experimental results show that this prediction method is more suitable for line heavy overload prediction than the method using the classification prediction model with the same model structure but with load level as input.

    参考文献
    [1] 卢艺,陶凯,林声宏.架空导线载流量动态计算与应用[J].电网技术,2009,33(20):76?81. YU Yi,TAO Kai,LIN Shenghong.Dynamic ampacity calculation of overhead transmission lines and its application[J].Power System Technology,2009,33(20):76?81.
    [2] 付善强,王孟夏,杨明,等.架空导线载流量的多时段联合概率密度预测[J].电力系统自动化,2019,43(17):102?113. FU Shanqiang,WANG Mengxia,YANG Ming,et al.Multi?period joint probability density forecasting for thermal rating of overhead line[J].Automation of Electric Power Systems,2019,43(17):102?113.
    [3] SONG P,XU Z,DONG H.UPFC?based line overload control for power system security enhancement[J].IET Generation,Transmission and Distribution,2017,11(13):3310?3317.
    [4] KO Y.An inference methodology based on the local minimum tree search to solve the distribution system events under heavy load condition[J].IEEE Transactions on Power Delivery,2012,27(2):725?734.
    [5] JIN X,MU Y,JIA H,et al.Alleviation of overloads in transmission network:A multi?level framework using the capability from active distribution network[J].International Journal Of Electrical Power & Energy Systems,2019,112:232?251.
    [6] BALARAMAN S,KAMARAJ N.Cascade BPN based transmission line overload prediction and preventive action by generation rescheduling[J].Neurocomputing,2012,94:1?12.
    [7] 陆继翔,张琪培,杨志宏,等.基于CNN?LSTM混合神经网络模型的短期负荷预测方法[J].电力系统自动化,2019,43(8):131?137. LU Jixiang,ZHANG Qipei,YANG Zhihong,et al.Short?term load forecasting method based on CNN?LSTM hybrid neural network model[J].Automation of Electric Power Systems,2019,43(8):131?137.
    [8] 吴云,雷建文,鲍丽山,等.基于改进灰色关联分析与蝙蝠优化神经网络的短期负荷预测[J].电力系统自动化,2018,42(20):67?72. WU Yun,LEI Jianwen,BAO Lishan,et al.Short?term load forecasting based on improved grey relational analysis and neural network optimized by bat algorithm[J].Automation of Electric Power Systems,2018,42(20):67?72.
    [9] 黄南天,齐斌,刘座铭,等.采用面积灰关联决策的高斯过程回归概率短期负荷预测[J].电力系统自动化,2018,42(23):64?71. HUANG Nantian,QI Bin,LIU Zuoming,et al.Probabilistic short?term load forecasting using Gaussian process regression with area grey incidence decision making[J].Automation of Electric Power Systems,2018,42(23):64?71.
    [10] 史常凯,闫文棋,张筱慧,等.基于BP网络和灰色模型的春节配变重过载预测[J].电力科学与技术学报,2016,31(3):140?145. SHI Changkai,YAN Wenqi,ZHANG Xiaohui,et al.Heavy overload forecasting of distribution transformer during the spring festival based on BP network and grey model[J]. Journal of Electric Power Science and Technology,2016,31(3):140?145.
    [11] 王凇瑶,张智晟.基于量子加权多层级GRU神经网络的综合能源系统多元负荷短期预测[J].电力系统保护与控制,2022,50(23):85?93. WANG Songyao,ZHANG Zhisheng.Short?term multivariate load forecasting of an integrated energy system based on a quantum weighted multi?hierarchy gated recurrent unit neural network[J].Power System Protection and Control,2022,50(23):85?93.
    [12] 徐鹏超,李琰,赵艳雷.基于自适应变异蝙蝠优化BP神经网络的短期风电功率预测[J].电测与仪表,2021,58(4):121?127. XU Pengchao,LI Yan,ZHAO Yanlei.Short?term wind power prediction based on BP neural network with adaptive mutation bat optimization algorithm[J].Electrical Measurement & Instrumentation,2021,58(4):121?127.
    [13] KONG W,DONG Z Y,JIA Y,et al.Short?term residential load forecasting based on LSTM recurrent neural network[J].IEEE Transactions on Smart Grid,2019,10(1):841?851.
    [14] 杨胡萍,余阳,汪超,等.基于VMD?CNN?BIGRU的电力系统短期负荷预测[J].中国电力,2022,55(10):71?76. YANG Huping,YU Yang,WANG Chao,et al.Short?term load forecasting of power system based on VMD?CNN? BIGRU[J].Electric Power,2022,55(10):71?76.
    [15] 马瑞,龚人杰,杨海晶.基于信息粒化和支持向量机的母线等效负荷波动预测方法[J].电力系统及其自动化学报,2019,31(4):25?31. MA Rui,GONG Renjie,YANG Haijing.Fluctuation prediction method for equivalent load of bus based on information granulation and support vector machine[J].Proceedings of the CSU?EPSA,2019,31(4):25?31.
    [16] 魏震波,余雷.基于FFT,DC?HC及LSTM的短期负荷预测方法[J].智慧电力,2022,50(3):37?43. WEI Zhenbo,YU Lei.Short?term load forecasting method based on FFT,DC?HC and LSTM[J].Smart Power,2022,50(3):37?43.
    [17] 王刚,杨晓静,张志军,等.基于深度神经网络与权值共享的工业园区负荷预测[J].电测与仪表,2021,58(1): 137?141. WANG Gang,YANG Xiaojing,ZHANG Zhijun,et al.Load forecasting in industrial park based on deep neural network and weightsharing[J].Electrical Measurement & Instrumentation,2021,58(1):137?141.
    [18] 赵兵,王增平,纪维佳,等.基于注意力机制的CNN?GRU短期电力负荷预测方法[J].电网技术,2019,43(12):4370?4376. ZHAO Bing,WANG Zengping,JI Weijia,et al.A short?term power load forecasting method based on attention mechanism of CNN?GRU[J].Power System Technology,2019,43(12):4370?4376.
    [19] 李正浩.基于NACEMD?GRU的组合型日前负荷预测方法[J].电网与清洁能源,2021,37(6):43?50. LI Zhenghao.A combined method of day?ahead load forecasting based on NACEMD?GRU[J].Power System and Clean Energy,2021,37(6):43?50.
    [20] 徐菲菲,芦霄鹏.结合卷积神经网络和最小门控单元注意力的文本情感分析[J].计算机应用与软件,2020,37(9):75?80+125. XU Feifei,LU Xiaopeng.Text sentiment analysis based on Attention C_MGU[J].Computer Applications and Software,2020,37(9):75?80+125.
    [21] 谢施君,雷汉坤,王乃会,等.基于BP神经网络的CVT暂态电压传递特性补偿技术[J]. 高压电器,2022,58(2):149?157. XIE Shijun,LEI Hankun,WANG Naihui,et al.Compensation technology on transient voltage transfer characteristics of CVT based on BP neural network[J]. High Voltage Apparatus,2022,58(2):149?157
    [22] 林盛振,谢敏,黄彬彬,等.基于贝叶斯正则优化NARX神经网络的电力负荷预测[J].供用电,2022,39(9):51?60. LIN Shengzhen,XIE Min,HUANG Binbin,et al.Power load forecasting based on Bayesian regularization optimized NARX neural network[J].Distribution & Utilization,2022,39(9):51?60.
    相似文献
    引证文献
引用本文

杨 秀,胡钟毓,田英杰,等.基于Attention机制的CNN‑GRU配网线路重过载短期预测方法[J].电力科学与技术学报,2023,38(1):201-209.
YANG Xiu, HU Zhongyu, TIAN Yingjie, et al. Short‑term heavy overload forecasting method of distribution net line based on CNN‑GRU with Attention mechanism[J]. Journal of Electric Power Science and Technology,2023,38(1):201-209.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 在线发布日期: 2023-04-10
文章二维码