Topology recognition of power distribution networks is the foundation for the planning, operation, and scheduling of power grids. The introduction of neural network can improve the accuracy of topology recognition in power distribution networks, but how to effectively design the neural network model that can balance robustness and convergence is still a difficult problem that needs to be solved urgently. A method for topology recognition of power distribution network based on data fusion with multi-channel and multi-scale convolutional neural networks was proposed, where data such as voltage amplitude, voltage phase angle, active power and reactive power of each node in a power distribution network were preprocessed and input into independent channels; in each channel the input data was subjected to convolution and normalization calculation at multiple scales to extract multi-scale convolution features, these normalized features were fused through a 1 × 1 convolution kernel and then mixed with the original input data of the channel, the mixed data was then processed through nonlinear transformation and output; the above data processing processes of each channel were carried out in parallel, the data output from all channels was flattened and input into a topology recognizer which was composed of fully connected layer network and a Sigmoid layer to complete the task of topology recognition of power distribution network. The proposed method was characterized by its scalable model and less affected by the quality of the original power data. Through case study on the power distribution networks of IEEE 33, it had been verified that compared to the 1D-CNN, Resnet18 and DNN(deep neural network) methods, the proposed method has the advantages of fast convergence speed, strong anti-interference performance, high robustness. The result demonstrates that the proposed method, based on reasonable feature extraction and fusion utilization of various power data, effectively addresses the challenge of balancing robustness and convergence in topology identification models. It reduces the model training’s reliance on large-scale datasets and is applicable to distribution network scenarios with distributed power sources, providing a critical foundation for distribution network system analysis and scheduling management.
YE Xue-shun, JIA Dong-li, ZHOU Jun, TANG Ying, JIANG Cheng-yan.
Topology Recognition of Power Distribution Network Based on Data Fusion with Multi-channel and Multi-scale Convolutional Neural Networks[J].
Science Technology and Engineering, 2026, 26(27): 11812-11821 DOI:10.12404/j.issn.1671-1815.2504961
WangGuo, GuoWenkai, WangChangchun. A review and prospect of topology identification research in distribution networks[J]. Journal of Jilin University (Engineering Edition), 2023, 53(2): 312-327.
LiangJingchao, WeiBin, MengRunquan, et al. A topology identification method of low-voltage distribution transformer area based on DPK-means and membership factors[J]. Power System and Clean Energy, 2025, 41(6): 73-82.
HuangXingde, FangChen, WeiXinchi, et al. Power system network topology analysis based on synchrophasor measurements[J]. Science Technology and Engineering, 2023, 23(29): 12567-12576.
XuHao, FanXiujuan, JiRui, et al. Resilience optimization of power grids considering integration of multi-timescale restoration strategies[J]. Science Technology and Engineering, 2023, 23(2): 619-625.
JinYangxi, XuYongjin, HuShuhong. Research on topology identification of low-voltage distribution network by fusing multi-source data in time-frequency domain[J]. Acta Energiae Solaris Sinica, 2025, 46(10): 487-500.
LuoQun, LiuChunyu, GuQiang, et al. Topology identification method for distribution network based on optimal matching loop power[J]. Electrical Measurement and Instrumentation, 2019, 56(19): 1-6.
[13]
CunhaV C, FreitasW, TrindadeF C L, et al. Automated determination of topology and line parameters in low voltage systems using smart meters measurements[J]. IEEE Transactions on Smart Grid, 2020, 11(6): 5028-5038.
LuoJunteng, ZhangJianmin, YaoLi, et al. Optimization modeling and application of phase identification for low-voltage users based on voltage and power data[J]. Automation of Electric Power Systems, 2021, 45(7): 123-131.
LiuXiangbo, WangSen, YanKai, et al. Research on topology identification method of distribution network based on multi-source data fusion[J]. Process Automation Instrumentation, 2024, 45(3): 59-63, 69.
YangXiu, JiangJiafu, LiuFang, et al. Topology identification of distribution network based on attention mechanism and convolutional neural network[J]. Power Grid Technology, 2022, 46(5): 1672-1682.
LongHuan, ShiZiqing, ZhaoJingtao, et al. Topology identification of distribution network based on multi-label classification and CNN[J]. High Voltage Engineering, 2024, 50(10): 4520-4529.
DuWenkai, ChenXinyi, XueDong. Topology identification for smart distribution networks based on data-driven and deep neural networks[J]. Journal of East China University of Science and Technology, 2025, 51(3): 353-362.
SunWei, ZhuShirui, YangJianping, et al. Topology identification of microgrids based on graph convolutional networks[J]. Power System Automation, 2022, 46(5): 71-81.
[26]
GottiD, AmarisH, LedesmaP. A deep neural network approach for online topology identification in state estimation[J]. IEEE Transactions on Power Systems, 2021, 36(6): 5824-5833.
[27]
LiY F, WangS Y, LiL, et al. Artificial intelligence for real-time topology identification in power distribution systems[C]// 2020 52nd North American Power Symposium. Tempe: IEEE, 2021: 1-6.
LiangHaiwei. Research on data driven topology and line parameter identification method for distribution network[D]. Wuhan: Huazhong University of Science and Technology, 2022.
[30]
NiJ L, TangZ, LiuJ, et al. A topology identification method based on one-dimensional convolutional neural network for distribution network[J]. Energy Reports, 2023, 9(S1): 355-362.
LaXianshou, DuanJipeng, WuXixi. Measures to improve the power supply reliability of the double-ring network in the urban core area[J]. Electric Engineering, 2024(S2): 489-491.