Abstract: In view of the voltage characteristic changes brought by the renewable energy integration and the fine-grained need for low-voltage topological relationship identification, a user link identification model based on graph signal processing（GSP）is proposed. The identification model uses the voltage similarity between adjacent nodes to represent the smoothness of the graph signal. The constraints of node current law on network structure are combined to overcome the influence of the in-phase user voltage similarity deterioration caused by renewable energy integration. The inherent link properties of the graph structure are used to realize the identification of the low-voltage topology down to the upstream and downstream connections between users. Finally,the simulation model based on real user data is used to verify the effectiveness of the proposed method. The performance of the proposed methods in different scenarios is discussed, and the proposed method is compared with existing identification methods.The example shows that, compared with the existing methods, the proposed method can effectively identify the user link relationship, and has better robustness to the penetration rate of renewable energy integration and data error rate.