
浏览全部资源
扫码关注微信
1.电子科技大学机械与电气工程学院,四川省成都市 611731
2.丹麦奥尔堡大学能源技术学院,奥尔堡 9100,丹麦
胡家祥(1999—),男,硕士研究生,主要研究方向:深度学习在电力系统中的应用。E-mail:jx.hu@std.uestc.edu.cn
曹迪(1993—),男,通信作者,博士,主要研究方向:配电网优化运行、机器学习理论及其在配电网的应用。E-mail:caodi_uestc@163.com
胡维昊(1982—),男,教授,博士生导师,主要研究方向:人工智能技术在电力系统中的应用、智慧能源网络等。E-mail:whu@uestc.edu.cn
收稿:2022-10-28,
修回:2022-11-18,
网络出版:2023-03-21,
纸质出版:2023-05-25
移动端阅览
胡家祥,曹迪,胡维昊等.融入拓扑知识的图神经网络配电网抗差状态估计方法[J].电力系统自动化,2023,47(10):84-97.
HU Jiaxiang,CAO Di,HU Weihao,et al.Robust State Estimation Method for Distribution Network Based on Graph Neural Network Incorporating Topology Knowledge[J].Automation of Electric Power Systems,2023,47(10):84-97.
胡家祥,曹迪,胡维昊等.融入拓扑知识的图神经网络配电网抗差状态估计方法[J].电力系统自动化,2023,47(10):84-97. DOI: 10.7500/AEPS20221028003.
HU Jiaxiang,CAO Di,HU Weihao,et al.Robust State Estimation Method for Distribution Network Based on Graph Neural Network Incorporating Topology Knowledge[J].Automation of Electric Power Systems,2023,47(10):84-97. DOI: 10.7500/AEPS20221028003.
量测信息少、量测精度不足是影响配电网状态估计精度的重要问题。基于确定物理拓扑的传统算法依赖于配电网精确参数,而普通的神经网络模型容易受到异常值与缺失值的干扰。为了解决上述问题,提出一种基于深度图学习的配电网抗差状态估计方法。所提模型依据物理拓扑知识进行信息传递,并使用全局图注意力机制聚合节点信息,以实现多种拓扑结构下的抗差状态估计任务。通过嵌入物理拓扑知识与注意力机制并融合深度学习技术,该模型可以学习数据深层次的结构信息,从而掌握多种拓扑的特征映射规律并获取抵抗异常值与干扰的能力。实验结果表明,该模型在多拓扑数据学习时具有更优的估计精度与抗差能力,并且可以使用多拓扑的历史运行数据提高估计精度。相较于普通的参数化模型,所提方法具有更强的鲁棒性与可解释性。
Less measurement information and insufficient measurement accuracy are the important problems that affect the accuracy of the state estimation for distribution networks. Traditional algorithms based on determining physical topology relay on the accurate parameters of the distribution network
while ordinary neural network models are easily disturbed by outliers and missing data. In order to solve the above problems
this paper proposes a robust state estimation method of the distribution network based on deep graph learning. The proposed model transfers the information based on physical topology knowledge
and implements the global graph attention mechanism to aggregate node information
which realizes the task of robust state estimation under various topologies. By embedding the physical topology knowledge and attention mechanism and integrating the deep learning technology
the model can learn the deep structural information of data
so as to master the feature mapping rules of various topologies and acquire the ability to resist outliers and interference. The experimental results show that the model has better estimation accuracy and robustness when learning with multi-topology data. In addition
the model can use the historical operation data of multi- topologies to improve the estimation accuracy. Compared with other algorithms
the proposed method has stronger robustness and interpretability.
陈艳波 , 谢瀚阳 , 王金丽 , 等 . 基于不确定测度的电力系统抗差状态估计(一)理论基础 [J]. 电力系统自动化 , 2018 , 42 ( 1 ): 8 - 15 .
CHEN Yanbo , XIE Hanyang , WANG Jinli , et al . Uncertain measure based robust state estimation of power system: Part one theoretical principle [J]. Automation of Electric Power Systems , 2018 , 42 ( 1 ): 8 - 15 .
胡鹏飞 , 朱乃璇 , 江道灼 , 等 . 柔性互联智能配电网关键技术研究进展与展望 [J]. 电力系统自动化 , 2021 , 45 ( 8 ): 2 - 12 .
HU Pengfei , ZHU Naixuan , JIANG Daozhuo , et al . Research progress and prospects of key technologies of flexible interconnected smart distribution network [J]. Automation of Electric Power Systems , 2021 , 45 ( 8 ): 2 - 12 .
FENG F , ZHANG P , ZHOU Y F . Authentic microgrid state estimation [J]. IEEE Transactions on Power Systems , 2022 , 37 ( 2 ): 1657 - 1660 .
ZHAO J B , GÓMEZ-EXPÓSITO A , NETTO M , et al . Power system dynamic state estimation: motivations, definitions, methodologies, and future work [J]. IEEE Transactions on Power Systems , 2019 , 34 ( 4 ): 3188 - 3198 .
杨挺 , 赵黎媛 , 刘亚闯 , 等 . 基于深度强化学习的综合能源系统动态经济调度 [J]. 电力系统自动化 , 2021 , 45 ( 5 ): 39 - 47 .
YANG Ting , ZHAO Liyuan , LIU Yachuang , et al . Dynamic economic dispatch for integrated energy system based on deep reinforcement learning [J]. Automation of Electric Power Systems , 2021 , 45 ( 5 ): 39 - 47 .
陈得治 , 郭志忠 . 基于表计配置的配电网可观测性分析 [J]. 中国电机工程学报 , 2005 , 25 ( 12 ): 35 - 41 .
CHEN Dezhi , GUO Zhizhong . Distribution system network observability analysis based on meter placement [J]. Proceedings of the CSEE , 2005 , 25 ( 12 ): 35 - 41 .
陈艳波 , 谢瀚阳 , 王鹏 , 等 . 基于不确定测度的电力系统抗差状态估计(二)模型方法 [J]. 电力系统自动化 , 2018 , 42 ( 2 ): 26 - 33 .
CHEN Yanbo , XIE Hanyang , WANG Peng , et al . Uncertain measure based robust state estimation of power system: Part two model and solution [J]. Automation of Electric Power Systems , 2018 , 42 ( 2 ): 26 - 33 .
何光宇 , 董树锋 . 基于测量不确定度的电力系统状态估计(三)算法比较 [J]. 电力系统自动化 , 2009 , 33 ( 21 ): 28 - 31 .
HE Guangyu , DONG Shufeng . Power system static-state estimation based on uncertainty of measurement: Part three algorithms compared [J]. Automation of Electric Power Systems , 2009 , 33 ( 21 ): 28 - 31 .
李碧君 , 薛禹胜 , 顾锦汶 , 等 . 基于权函数的电力系统状态估计算法 [J]. 电力系统自动化 , 1999 , 23 ( 8 ): 32 - 34 .
LI Bijun , XUE Yusheng , GU Jinwen , et al . Weight function based power system state estimation algorithm [J]. Automation of Electric Power Systems , 1999 , 23 ( 8 ): 32 - 34 .
SCHWEPPE F C . Power system static-state estimation, part Ⅲ: implementation [J]. IEEE Transactions on Power Apparatus and Systems , 1970 , 89 ( 1 ): 130 - 135 .
BIAN X M , LI X R , CHEN H M , et al . Joint estimation of state and parameter with synchrophasors—part Ⅰ: state tracking [J]. IEEE Transactions on Power Systems , 2011 , 26 ( 3 ): 1196 - 1208 .
陈艳波 , 马进 , 文一宇 . 一种电力系统量测噪声自适应抗差状态估计方法 [J]. 电力系统自动化 , 2015 , 39 ( 8 ): 66 - 73 .
CHEN Yanbo , MA Jin , WEN Yiyu . An adaptive robust state estimation approach for measurement noise [J]. Automation of Electric Power Systems , 2015 , 39 ( 8 ): 66 - 73 .
CHEN Y B , CHEN H , JIAO Y , et al . Data-driven robust state estimation through off-line learning and on-line matching [J]. Journal of Modern Power Systems and Clean Energy , 2021 , 9 ( 4 ): 897 - 909 .
陈艳波 , 马进 . 一种双线性抗差状态估计方法 [J]. 电力系统自动化 , 2015 , 39 ( 6 ): 41 - 47 .
CHEN Yanbo , MA Jin . A bilinear robust state estimation method for power systems [J]. Automation of Electric Power Systems , 2015 , 39 ( 6 ): 41 - 47 .
颜全椿 , 卫志农 , 徐泰山 , 等 . 基于主导与非主导参数的非线性加权最小绝对值参数估计 [J]. 电力系统自动化 , 2013 , 37 ( 20 ): 71 - 76 .
YAN Quanchun , WEI Zhinong , XU Taishan , et al . Nonlinear weighted absolute least value parameter estimation based on dominant and non-dominant parameter [J]. Automation of Electric Power Systems , 2013 , 37 ( 20 ): 71 - 76 .
GARANAYAK P , PANDA G . An adaptive linear neural network with least mean M-estimate weight updating rule employed for harmonics identification and power quality monitoring [J]. Transactions of the Institute of Measurement and Control , 2018 , 40 ( 6 ): 1936 - 1949 .
郭烨 , 张伯明 , 吴文传 . 指数型目标函数电力系统抗差状态估计的解法与性能分析 [J]. 中国电机工程学报 , 2011 , 31 ( 7 ): 89 - 95 .
GUO Ye , ZHANG Boming , WU Wenchuan . Solution and performance analysis to a robust state estimation method with exponential objective function [J]. Proceedings of the CSEE , 2011 , 31 ( 7 ): 89 - 95 .
赵巍岳 , 靳松 , 吕天成 . 基于神经网络的状态估计方法研究 [J]. 电力系统保护与控制 , 2018 , 46 ( 22 ): 109 - 115 .
ZHAO Weiyue , JIN Song , LÜ Tiancheng . Research on state estimation based on artificial neural networks [J]. Power System Protection and Control , 2018 , 46 ( 22 ): 109 - 115 .
TIAN G Y , GU Y Z , SHI D , et al . Neural-network-based power system state estimation with extended observability [J]. Journal of Modern Power Systems and Clean Energy , 2021 , 9 ( 5 ): 1043 - 1053 .
SHAO L , ZHU F , LI X L . Transfer learning for visual categorization: a survey [J]. IEEE Transactions on Neural Networks and Learning Systems , 2015 , 26 ( 5 ): 1019 - 1034 .
YANG L , JING L P , YU J , et al . Learning transferred weights from co-occurrence data for heterogeneous transfer learning [J]. IEEE Transactions on Neural Networks and Learning Systems , 2016 , 27 ( 11 ): 2187 - 2200 .
李鹏 , 王瑞 , 冀浩然 , 等 . 低碳化智能配电网规划研究与展望 [J]. 电力系统自动化 , 2021 , 45 ( 24 ): 10 - 21 .
LI Peng , WANG Rui , JI Haoran , et al . Research and prospect of planning for low-carbon smart distribution network [J]. Automation of Electric Power Systems , 2021 , 45 ( 24 ): 10 - 21 .
王守相 , 刘琪 , 赵倩宇 , 等 . 配电网弹性内涵分析与研究展望 [J]. 电力系统自动化 , 2021 , 45 ( 9 ): 1 - 9 .
WANG Shouxiang , LIU Qi , ZHAO Qianyu , et al . Connotation analysis and prospect of distribution network elasticity [J]. Automation of Electric Power Systems , 2021 , 45 ( 9 ): 1 - 9 .
陈源奕 , 王玉彬 , 杨强 . 基于时空特征图卷积网络的配电网实时状态估计方法 [J]. 高电压技术 , 2021 , 47 ( 7 ): 2386 - 2395 .
CHEN Yuanyi , WANG Yubin , YANG Qiang . Real-time state estimation method for distribution networks based on spatial-temporal feature graph convolution network [J]. High Voltage Engineering , 2021 , 47 ( 7 ): 2386 - 2395 .
孟雨田 , 严正 , 徐潇源 , 等 . 配电网状态估计的全局灵敏度分析及应用 [J]. 电力系统自动化 , 2020 , 44 ( 2 ): 114 - 122 .
MENG Yutian , YAN Zheng , XU Xiaoyuan , et al . Global sensitivity analysis on state estimation of distribution network and its application [J]. Automation of Electric Power Systems , 2020 , 44 ( 2 ): 114 - 122 .
SCARSELLI F , GORI M , TSOI A C , et al . The graph neural network model [J]. IEEE Transactions on Neural Networks , 2009 , 20 ( 1 ): 61 - 80 .
DONG C , LOY C C , HE K M , et al . Image super-resolution using deep convolutional networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2016 , 38 ( 2 ): 295 - 307 .
HONG D F , GAO L R , YAO J , et al . Graph convolutional networks for hyperspectral image classification [J]. IEEE Transactions on Geoscience and Remote Sensing , 2021 , 59 ( 7 ): 5966 - 5978 .
YAN Y Y , REN W Q , HU X B , et al . SRGAT: single image super-resolution with graph attention network [J]. IEEE Transactions on Image Processing , 2021 , 30 : 4905 - 4918 .
LONG H , WU Z , FANG C , et al . Cyber-attack detection strategy based on distribution system state estimation [J]. Journal of Modern Power Systems and Clean Energy , 2020 , 8 ( 4 ): 669 - 678 .
陈晓爽 , 林今 , 宋永华 . 配电网状态估计误差的电路表示方法 [J]. 中国电机工程学报 , 2014 , 34 ( 28 ): 4839 - 4846 .
CHEN Xiaoshuang , LIN Jin , SONG Yonghua . A circuit representing method for state estimation errors in distribution networks [J]. Proceedings of the CSEE , 2014 , 34 ( 28 ): 4839 - 4846 .
臧海祥 , 郭镜玮 , 黄蔓云 , 等 . 基于深度迁移学习的时变拓扑下电力系统状态估计 [J]. 电力系统自动化 , 2021 , 45 ( 24 ): 49 - 56 .
ZANG Haixiang , GUO Jingwei , HUANG Manyun , et al . State estimation for power systems with time-varying topology based on deep transfer learning [J]. Automation of Electric Power Systems , 2021 , 45 ( 24 ): 49 - 56 .
黄巍 , 吴俊勇 , 鲁思棋 , 等 . 电压偏差和谐波约束下配网光伏最大渗透率评估 [J]. 电力系统保护与控制 , 2016 , 44 ( 6 ): 49 - 55 .
HUANG Wei , WU Junyong , LU Siqi , et al . Maximum photovoltaic penetration considering voltage deviation and harmonic constraints [J]. Power System Protection and Control , 2016 , 44 ( 6 ): 49 - 55 .
曲正伟 , 张嘉曦 , 王云静 , 等 . 考虑分布式电源不确定性的配电网改进仿射状态估计 [J]. 电力系统自动化 , 2021 , 45 ( 23 ): 104 - 112 .
QU Zhengwei , ZHANG Jiaxi , WANG Yunjing , et al . Improved affine state estimation for distribution network considering uncertainty of distributed generator [J]. Automation of Electric Power Systems , 2021 , 45 ( 23 ): 104 - 112 .
曹迪 . 主动配电网电压控制策略研究 [D]. 成都 : 电子科技大学 , 2021 .
CAO Di . Research on voltage control strategy of active distribution network [D]. Chengdu : University of Electronic Science and Technology of China , 2021 .
王彩霞 , 时智勇 , 梁志峰 , 等 . 新能源为主体电力系统的需求侧资源利用关键技术及展望 [J]. 电力系统自动化 , 2021 , 45 ( 16 ): 37 - 48 .
WANG Caixia , SHI Zhiyong , LIANG Zhifeng , et al . Key technologies and prospects of demand-side resource utilization for power systems dominated by renewable energy [J]. Automation of Electric Power Systems , 2021 , 45 ( 16 ): 37 - 48 .
0
浏览量
0
下载量
3
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621