
浏览全部资源
扫码关注微信
1.浙江大学电气工程学院,浙江省杭州市 310027
2.南方电网科学研究院有限责任公司,广东省广州市 510663
3.国网浙江省电力有限公司衢州供电公司,浙江省衢州市 324000
4.国网江苏省电力有限公司电力科学研究院,江苏省南京市 211103
谭伟涛(1999—),男,硕士研究生,主要研究方向:配电网拓扑辨识及态势感知。E-mail:twt@zju.edu.cn
彭里卓(2001—),男,硕士研究生,主要研究方向:电力系统人工智能应用。E-mail:penglizhuo@zju.edu.cn
章天晗(1997—),男,博士,工程师,主要研究方向:电力系统态势感知。E-mail:eezhangth@zju.edu.cn
林振智(1979—),男,通信作者,博士,教授,博士生导师,主要研究方向:电力系统态势感知、电力大数据与AI应用、灵活性资源规划与运行。E-mail:linzhenzhi@zju.edu.cn
收稿日期:2025-01-27,
修回日期:2025-06-11,
网络出版日期:2025-07-18,
纸质出版日期:2025-09-25
移动端阅览
谭伟涛,彭里卓,章天晗等.基于半监督对抗学习的配电网动态拓扑开关状态鲁棒辨识[J].电力系统自动化,2025,49(18):43-54.
TAN Weitao,PENG Lizhuo,ZHANG Tianhan,et al.Robust Identification of Switch States in Dynamic Topology of Distribution Network Based on Semi-supervised Adversarial Learning[J].Automation of Electric Power Systems,2025,49(18):43-54.
谭伟涛,彭里卓,章天晗等.基于半监督对抗学习的配电网动态拓扑开关状态鲁棒辨识[J].电力系统自动化,2025,49(18):43-54. DOI: 10.7500/AEPS20250127003.
TAN Weitao,PENG Lizhuo,ZHANG Tianhan,et al.Robust Identification of Switch States in Dynamic Topology of Distribution Network Based on Semi-supervised Adversarial Learning[J].Automation of Electric Power Systems,2025,49(18):43-54. DOI: 10.7500/AEPS20250127003.
当处理标记拓扑与未标记拓扑之间和各类标记拓扑之间数量不均衡的拓扑类不匹配问题时,现有基于有监督学习和无监督学习的拓扑开关状态辨识方法均难以准确辨识配电网拓扑结构。为此,提出基于半监督对抗学习的配电网动态拓扑开关状态鲁棒辨识方法。首先,提出基于动态节点嵌入的配电网拓扑变化特征提取方法,其通过将拓扑动态变化过程中节点电压量测值表示为潜在低维向量以反映配电网拓扑重构过程中节点关联程度的变化,并降低模型的计算复杂度。然后,针对配电网拓扑开关状态辨识中存在的拓扑类不匹配问题,提出了基于半监督生成对抗网络的配电网拓扑开关状态鲁棒辨识模型,通过内嵌的半监督学习机制和对抗学习机制缓解拓扑类不匹配问题,实时准确地辨识配电网的拓扑结构。最后,以IEEE 33节点测试系统为例进行算例分析,仿真结果表明所提方法可以快速辨识配电网拓扑开关状态的动态变化,且在拓扑类不匹配的情况下依然能实现拓扑的准确辨识。
When dealing with the topology class mismatch problem of uneven quantity between labeled topologies and unlabeled topologies
as well as among various labeled topologies
the existing methods for identification of the switch state in topology based on supervised learning and unsupervised learning are unable to accurately identify the topology of the distribution network. Therefore
a robust identification of switch states in dynamic topology of distribution network based on semi-supervised adversarial learning is proposed. Firstly
a method for topology change feature extraction of distribution network based on dynamic node embedding (TCFEDN-DNE) is proposed. This method represents the node voltage measurement values during the dynamic topology change process as potential low-dimensional vectors to reflect the changes in the degree of node correlation during the topology reconstruction process and reduce the computational complexity of the model. Then
for the problem of topology class mismatch in the identification of the topology switch state of the distribution network
a robust identification model of switch states in dynamic topology of distribution network based on semi-supervised generative adversarial network (SGAN) is proposed. Through the embedded semi-supervised learning mechanism and adversarial learning mechanism
the problem of topology class mismatch is alleviated
and the topology of the distribution network can be accurately identified in real time. Finally
a case analysis using the IEEE 33-bus test system is conducted. The simulation results show that the proposed method can quickly identify the dynamic changes in the topology switch state of the distribution network
and can still accurately identify the topology even in the case of topology class mismatch.
国家发展改革委 , 国家能源局 . 关于新形势下配电网高质量发展的指导意见 [EB/OL].[ 2025-01-01 ]. https://www.gov.cn/zhengce/zheng-ceku/202403/content_6935790.htm https://www.gov.cn/zhengce/zheng-ceku/202403/content_6935790.htm .
National Development and Reform Commission , National Energy Administration . Guiding opinions on the high-quality development of distribution networks under the new situation [EB/OL]. [ 2025-01-01 ]. https://www.gov.cn/zhengce/zhenge ku/202403/content_6935790.htm https://www.gov.cn/zhengce/zhengeku/202403/content_6935790.htm .
PARK S , LEE E , YU W , et al . State estimation for supervisory monitoring of substations [J]. IEEE Transactions on Smart Grid , 2013 , 4 ( 1 ): 406 - 410 .
余盛达 , 王玉中 , 张祖伟 , 等 . 基于导线阻抗特征改进聚类鲁棒性的台区线-相-户拓扑识别方法 [J]. 南方电网技术 , 2023 , 17 ( 7 ): 55 - 64 .
YU Shengda , WANG Yuzhong , ZHANG Zuwei , et al . Line-phase-household topology identification method in substation area based on wire impedance characteristics to improve clustering robustness [J]. Southern Power System Technology , 2023 , 17 ( 7 ): 55 - 64 .
WANG X X , ZHAO Y K , ZHOU Y . A data-driven topology and parameter joint estimation method in non-PMU distribution networks [J]. IEEE Transactions on Power Systems , 2024 , 39 ( 1 ): 1681 - 1692 .
蒋玮 , 汤海波 , 祁晖 , 等 . 基于集成深度神经网络的配电网联络关系辨识技术 [J]. 电力系统自动化 , 2020 , 44 ( 1 ): 101 - 108 .
JIANG Wei , TANG Haibo , QI Hui , et al . Distribution network connectivity recognition based on ensemble deep neural network [J]. Automation of Electric Power Systems , 2020 , 44 ( 1 ): 101 - 108 .
杨秀 , 蒋家富 , 刘方 , 等 . 基于注意力机制和卷积神经网络的配电网拓扑辨识 [J]. 电网技术 , 2022 , 46 ( 5 ): 1672 - 1682 .
YANG Xiu , JIANG Jiafu , LIU Fang , et al . Distribution network topology identification based on attention mechanism and convolutional neural network [J]. Power System Technology , 2022 , 46 ( 5 ): 1672 - 1682 .
陈碧云 , 吕怡博 , 梁志坚 , 等 . 考虑数据失衡的新型配电网两阶段拓扑辨识 [J]. 电力系统保护与控制 , 2023 , 51 ( 21 ): 57 - 65 .
CHEN Biyun , LÜ Yibo , LIANG Zhijian , et al . Two-stage topology identification of a new-type distribution network considering data imbalance [J]. Power System Protection and Control , 2023 , 51 ( 21 ): 57 - 65 .
WU H Y , XU Z , ZHAO J , et al . Gridtopo-GAN for distribution system topology identification [J]. IEEE Transactions on Industrial Informatics , 2022 , 19 ( 4 ): 5356 - 5366 .
俞小勇 , 秦丽文 , 桂海涛 , 等 . 新一代人工智能在配电网智能感知与故障诊断中的应用 [J]. 南方电网技术 , 2022 , 16 ( 5 ): 34 - 43 .
YU Xiaoyong , QIN Liwen , GUI Haitao , et al . Application of new generation artificial intelligence technology in intelligent sensing and fault detection of power distribution network [J]. Southern Power System Technology , 2022 , 16 ( 5 ): 34 - 43 .
党建 , 闫运江 , 贾嵘 , 等 . 基于图计算的配电网最大供电能力评估研究 [J]. 电网技术 , 2022 , 46 ( 3 ): 1039 - 1049 .
DANG Jian , YAN Yunjiang , JIA Rong , et al . Total supply capability evaluation of distribution network based on graph computation [J]. Power System Technology , 2022 , 46 ( 3 ): 1039 - 1049 .
马智刚 , 卫志农 , 陈胜 , 等 . 基于图计算的交直流混合配电网优化调度 [J]. 电力系统自动化 , 2023 , 47 ( 18 ): 161 - 170 .
MA Zhigang , WEI Zhinong , CHEN Sheng , et al . Optimal dispatch of AC/DC hybrid distribution network based on graph computing [J]. Automation of Electric Power Systems , 2023 , 47 ( 18 ): 161 - 170 .
孙黎霞 , 彭嘉杰 , 白景涛 , 等 . 结合图嵌入算法的电力系统多任务暂态稳定评估 [J]. 电力系统自动化 , 2022 , 46 ( 2 ): 83 - 91 .
SUN Lixia , PENG Jiajie , BAI Jingtao , et al . Multi-task transient stability assessment of power system incorporating graph embedding algorithm [J]. Automation of Electric Power Systems , 2022 , 46 ( 2 ): 83 - 91 .
KIPF T N , WELLING M . Variational graph auto-encoders [EB/OL]. [ 2025-01-01 ]. https://arxiv.org/abs/1611.07308 https://arxiv.org/abs/1611.07308 .
HAMILTON W L , YING R , LESKOVEC J . Inductive representation learning on large graphs [EB/OL]. [ 2025-01-01 ]. https://arxiv.org/abs/1706.02216 https://arxiv.org/abs/1706.02216 .
杨艳红 , 卢鑫 , 张雷杰 , 等 . 考虑智能体前置状态及环境特征自适应机制的强化学习电网调度方法 [J]. 高电压技术 , 2024 , 50 ( 8 ): 3497 - 3509 .
YANG Yanhong , LU Xin , ZHANG Leijie , et al . Reinforcement learning power grid dispatching method considering the pre-state of intelligent agent and adaptive mechanism of environmental features [J]. High Voltage Engineering , 2024 , 50 ( 8 ): 3497 - 3509 .
郭小龙 , 李子康 , 刘灏 , 等 . 基于增强生成对抗网络的PMU丢失数据恢复方法 [J]. 电网技术 , 2022 , 46 ( 6 ): 2114 - 2121 .
GUO Xiaolong , LI Zikang , LIU Hao , et al . PMU missing data recovery algorithm based on enhanced generative adversarial network [J]. Power System Technology , 2022 , 46 ( 6 ): 2114 - 2121 .
杨玉莲 , 齐林海 , 王红 , 等 . 基于生成对抗和双重语义感知的配电网量测数据缺失重构 [J]. 电力系统自动化 , 2020 , 44 ( 18 ): 46 - 54 .
YANG Yulian , QI Linhai , WANG Hong , et al . Reconstruction of missing measurement data in distribution network based on generative adversarial network and double semantic perception [J]. Automation of Electric Power Systems , 2020 , 44 ( 18 ): 46 - 54 .
李元诚 , 杨珊珊 . 基于改进自注意力机制生成对抗网络的智能电网GPS欺骗攻击防御方法 [J]. 电力自动化设备 , 2021 , 41 ( 11 ): 100 - 106 .
LI Yuancheng , YANG Shanshan . Defense method of smart grid GPS spoofing attack based on improved self-attention generative adversarial network [J]. Electric Power Automation Equipment , 2021 , 41 ( 11 ): 100 - 106 .
邵晨颖 , 刘友波 , 邵安海 , 等 . 基于生成对抗网络与局部电流相量的配电网拓扑鲁棒辨识 [J]. 电力系统自动化 , 2023 , 47 ( 1 ): 55 - 62 .
SHAO Chenying , LIU Youbo , SHAO Anhai , et al . Robust identification for distribution network topology based on generative adversarial network and partial current phasor [J]. Automation of Electric Power Systems , 2023 , 47 ( 1 ): 55 - 62 .
WU S , DENG G C , LI J C , et al . Enhancing TripleGAN for semi-supervised conditional instance synthesis and classification [C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition , June 15-20 , 2019 , Long Beach, USA : 10083 - 10092 .
BARAN M E , WU F F . Network reconfiguration in distribution systems for loss reduction and load balancing [J]. IEEE Transactions on Power Delivery , 1989 , 4 ( 2 ): 1401 - 1407 .
STONE M . Cross-validatory choice and assessment of statistical predictions [J]. Journal of the Royal Statistical Society Series B: Statistical Methodology , 1974 , 36 ( 2 ): 111 - 133 .
FERNANDES T R , VENKATESH B , ALMEIDA M C , et al . Distribution system topology identification via efficient MILP-based WLAV state estimation [J]. IEEE Transactions on Power Systems , 2023 , 38 ( 1 ): 75 - 84 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621