基于改进循环神经网络的配电网超短期功率预测方法
CSTR:
作者:
中图分类号:

TM71

基金项目:

国家重点研发计划(2020YFF0305800);国家电网有限公司科技项目(520201210025)


Ultra-short-term power prediction method of distribution network based on improved recurrent neural network
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [20]
  • | | | |
  • 文章评论
    摘要:

    针对传统单向循环神经网络在配电网超短期功率预测领域存在的预测曲线失形、模型过拟合现象以及预测精度不高和收敛速度慢等问题,提出基于小波变换和自注意力机制的双向循环神经网络改进模型。通过双向结构学习功率数据的前向和逆向规律提高模型预测精度;通过小波变换分摊整体功率预测难度以及改善过拟合和加快模型收敛速度;通过自注意力机制把握模型隐藏层维度关系进一步提高预测精度。算例证明改进模型可以有效改善上述问题,改进模型与传统单向模型相比,在有功预测场景中,MAE提升了50.1%,MAPE提升了43.3%,RMSE提升了51.1%;在无功预测场景中,MAE提升了60.5%,MAPE提升了63.8%,RMSE提升了60.1%。

    Abstract:

    The traditional one-directional neural network has some problems in the field of ultra-short-term power prediction in distribution networks, such as the out-of-shape curve prediction, the over-fitting phenomenon of the model, low prediction accuracy and slow convergence speed, etc. Thus, an improved bi-directional recurrent neural network model is proposed based on the wavelet transform and self-attention mechanism to overcome these problems. Firstly, the forward and reverse laws of the power data are studied by the bi-directional network to improve the prediction accuracy of the model. Afterward, the wavelet transform is employed to reduce the overall difficulty of power prediction. Consequently, the model overfitting is reduced, and the convergence speed is increased in the meantime. In the end, the self-attention mechanism is adopted to grasp the hidden layer dimensional relationship of the model to further improve the prediction accuracy. An example shows that the proposed improved model can eliminate the existing problems effectively. Compared with the traditional model, the MAE increased by 50.1%, MAPE increased by 43.3%, RMSE increased by 51.1%; in the reactive dataset, dataset MAE increased by 60.5%, MAPE increased by 63.8%, and RMSE increased by 60.1%.

    参考文献
    [1] 黄伟,刘琦,杨舒文,等.基于主动配电系统供电能力的安全态势感知方法[J].电力自动化设备,2017,37(8):74-80.HUANG Wei,LIU Qi,YANG Shuwen,et al.Security situation awareness based on power-supply ability model of active distribution system[J].Electric Power Automation Equipment,2017,37(8):74-80.
    [2] 杨肖虎,罗剑波,郁琛,等.适应大规模新能源并网的电力系统备用配置及优化综述[J].电力工程技术,2020,39(1):10-20+63.YANG Xiaohu,LUO Jianbo,YU Chen,et al.Review of power system reserve configuration and optimization for large-scale renewable energy integration[J].Electric Power Engineering Technology,2020,39(1):10-20+63.
    [3] 刘威,张东霞,王新迎,等.基于深度强化学习的电网紧急控制策略研究[J].中国电机工程学报,2018,38(1):109-119+347.LIU Wei,ZHANG Dongxia,WANG Xinying,et al.A decision making strategy for generating unit tripping under emergency circumstances based on deep reinforcement learning[J].Proceedings of the CSEE,2018,38(1):109-119+347.
    [4] 江秀臣,盛戈皞.电力设备状态大数据分析的研究和应用[J].高电压技术,2018,44(4):1041-1050.JIANG Xiuchen,SHENG Gehao.Research and application of big data analysis of power equipment condition[J].High Voltage Engineering,2018,44(4):1041-1050.
    [5] 冷华,陈鸿琳,李欣然,等.基于功率或电量预测的智能配电网统计线损同期化方法[J].电力系统保护与控制,2016,44(18):108-114.LENG Hua,CHEN Honglin,LI Xinran,et al.A method for synchronous line loss statistics of distribution network based on load or electricity consumption forecas-ting[J].Power System Protection and Control,2016,44(18):108-114.
    [6] 李正浩,李孟凡.基于深度学习的智能型负荷预测方法的研究[J].智慧电力,2020,48(10):78-85+112.LI Zhenghao,LI Mengfan.Smart load forecasting method based on deep learning[J].Smart Power,2020,48(10):78-85+112.
    [7] 牛哲文,余泽远,李波,等.基于深度门控循环单元神经网络的短期风功率预测模型[J].电力自动化设备,2018,38(5):36-42.NIU Zhewen,YU Zeyuan,LI Bo,et al.Short-term wind power forecasting model based on deep gated recurrent unit neural network[J].Electric Automation Equipment,2018,38(5):36-42.
    [8] 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.
    [9] 杨子民,彭小圣,郎建勋,等.基于集群动态划分与BLSTM深度学习的风电集群短期功率预测[J].高电压技术,2021,47(4):1195-1203.YANG Zimin,PENG Xiaosheng,LANG Jianxun,et al.Short-term wind power prediction based on dynamic cluster division and BLSTM deep learning method[J].High Voltage Engineering,2021,47(4):1195-1203.
    [10] 肖白,肖志峰,姜卓,等.基于降噪自编码器、奇异谱分析和长短期记忆神经网络的空间电力负荷态势感知[J].中国电机工程学报2021,41(14):4858-4867.XIAO Bai,XIAO Zhifeng,JIANG Zhuo,et al.Spatial load situation awareness based on denoising autoencoder,singular spectrum analysis and long short-term memory neural networks[J].Proceedings of the CSEE,2021,41(14):4858-4867.
    [11] 王婧骅,张娟,赵婉茹,等.基于时间序列的分布式光伏电站发电数据采集方法[J].电网与清洁能源,2022,38(6):137-142.WANG Jinghua,ZHANG Juan,ZHAO Wanru,et al.A method for collecting power generation data of distributed photovoltaic power station based on time series[J].Power System and Clean Energy,2022,38(6):137-142.
    [12] 谢小瑜,周俊煌,张勇军.深度学习在泛在电力物联网中的应用与挑战[J].电力自动化设备,2020,40(4):77-87.XIE Xiaoyu,ZHOU Junhuang,ZHANG Yongjun.Application and challenge of deep learning in ubiquitous power internet of things[J].Electric Power Automation Equipment,2020,40(4):77-87.
    [13] BENGIO Y,SIMARD P,FRASCONI P.Learning long-term dependencies with gradient descent is difficult[J].IEEE Transactions on Neural Networks,1994,5(2):157-66.
    [14] 陈铁,陈卫东,李咸善,等.基于EMD和GCT的变压器油中溶解气体预测[J].高压电器,2022,58(4):70-79.CHEN Tie,CHEN Wedong,LI Xianshan,et al.Dissolved gas prediction in transformer oil based on EMD and GCT[J].High Voltage Apparatus,2022,58(4):70-79.
    [15] CHO K,VAN MERRIENBOER B,GULCEHRE C,et al.Learning phrase representations using RNN encoder-decoder for statistical machine translation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing,Doha,Qatar,2014.
    [16] 胡翀,徐斌,甄超,等.基于电压暂降监测数据的敏感负荷非侵入式识别方法[J].中国电力,2021,54(8):35-42+51.HU Chong,XU Bin,ZHEN Chao,et al.A non-invasive identification method for sensitive load based on voltage sag monitoring data[J].Electric Power,2021,54(8):35-42+51.
    [17] 吕新华,武斌,攸阳,等.小波变换Mallat算法实现中的边界延拓研究[J].天津理工大学学报,2006,22(2):14-17.LU Xinhua,WU Bin,YOU Yang,et al.Border extension in the implementation of Mallat algorithm[J].Journal of Tianjin University of Technology,2006,22(2):14-17.
    [18] 谢小瑜,周俊煌,张勇军,等.基于W-BiLSTM的可再生能源超短期发电功率预测方法[J].电力系统自动化,2021,45(8):175-184.XIE Xiaoyu,ZHOU Junhuang,ZHANG Yongjun,et al.W-BiLSTM based ultra-short-term generation power prediction method of renewable energy[J].Automation of Electric Power Systems,2021,45(8):175-184.
    [19] 王伟,张彦龙,翟登辉,等.基于OpenCV+SSD深度学习模型的变电站压板状态智能识别[J].电测与仪表,2022,59(1):106-112.WANG Wei,ZHANG Yanlong,ZHAI Denghui,et al.Intelligent identification of substation pressure plate state based on OpenCV+SSD deep learning model[J].Electrical Measurement & Instrumentation,2022,59(1):106-112.
    [20] 彭曙蓉,黄士峻,李彬,等.基于深度学习分位数回归模型的充电桩负荷预测[J].电力系统保护与控制,2020,48(2):44-50.PENG Shurong,HUANG Shijun,LI Bin,et al.Charging pile load prediction based on deep learning quantile regression model[J] Power System Protection and Control,2020,48(2):44-50.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

赵振兵,强一凡,李信,等.基于改进循环神经网络的配电网超短期功率预测方法[J].电力科学与技术学报,2022,37(5):144-154.
Zhao Zhenbing, Qiang Yifan, Li Xin, et al. Ultra-short-term power prediction method of distribution network based on improved recurrent neural network[J]. Journal of Electric Power Science and Technology,2022,37(5):144-154.

复制
分享
文章指标
  • 点击次数:199
  • 下载次数: 627
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 在线发布日期: 2022-12-01
文章二维码