Topology Identification of Distribution Network Based on Fourier Kolmogorov-Arnold Networks

Large-scale distributed energy resource access has made the operation mode of the distribution network more complex, thereby increasing the difficulty of quickly, and accurately identifying its topology. To address this problem, this paper proposes a topology identification method for distribution networks based on Extreme Gradient Boosting (XGBoost) and Fourier Kolmogorov-Arnold Networks (FourierKAN). Firstly, the importance of voltage amplitudes of all nodes is calculated through the XGBoost algorithm. Then, feature selection is performed, and the feature subset is constructed. Kolmogorov-Arnold Networks (KAN) enhanced with Fourier series is utilized to establish the FourierKAN model, and the mapping relationship between the sample data and the distribution network topology can be derived. Finally, the proposed topology identification method is verified on the standard IEEE 33-node and IEEE 70-node distribution networks. The results show that the proposed method can use the voltage amplitudes of certain nodes to identify the network topology accurately, and the FourierKAN model has outstanding accuracy and computational efficiency. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.