The continuous growth of the number of electric vehicles (EVs) increases the proportion of charging station load in the power grid, which has accelerated the distribution network upgrade including the topology structure change and the line renovation. As a result, the load prediction accuracy without considering the distribution network upgrade will be reduced. For the EV charging station load prediction considering the distribution network upgrade, this article proposes an EGAT-LSTM prediction method integrating edge aggregation graph attention network (EGAT) model and long short-term memory network (LSTM) model. The EGAT model can extract the key system information to a new node characteristic set integrating adjacency relationship, line impedance (R/X) and node data. The EGAT-LSTM prediction method through training learnable parameters captures the spatio-temporal correlations to reduce the non-regression learning of the prediction model. This method is tested on IEEE 33 and IEEE 69 bus distribution network systems. Simulation results show that for the EV charging station load prediction considering the distribution network upgrade, the proposed method improves training efficiency and prediction accuracy compared with the other methods without considering line characteristic.

False data injection attack (FDIA) is posing a threat to the security of power grids. Detection technology is an effective means to defend against FDIA, but the existing mainstream methods have insufficient detection capabilities for large-scale power grids. This study proposes a novel method that combines subgraph partitioning strategy and hypergraph model to detect FDIA. According to the principle the attack principle, the power grid is partitioned into subgraphs. Each subgraph is constructed as the hypergraph and then input into the hypergraph convolutional neural network (HGCNN). The hypergraph attention mechanism (HGAT) is adopted to pay attention to the hyperedge, where the attention score is calculated through the similarity between the node and the hyperedge. Simulations were conducted on IEEE 14-, 118-, and 300-bus systems. At the 10% attack intensity, the proposed method achieved 1.62%, 2.05%, and 2.18% higher accuracy than the optimal results of the comparison methods on three test systems, respectively.

The continuous growth of electric vehicle (EV) ownership has increased the proportion of EV charging station load (EVCSL) in the distribution network (DN). The prediction of EVCSL is important for the safe and stable operation of the DN. However, simply predicting the EVCSL based on the characteristics of the DN, ignoring the impact of coupled transportation network (TN) characteristics, will reduce prediction performance. Few studies focus on combining DN and TN data for EVCSL prediction. On the premise of protecting the privacy of TN data, this paper proposes a vertical adaptive attention-based federated prediction method of EVCSL based on an edge aggregation graph attention network combined with a long- and short-term memory network (V2AFedEGAT combined with LSTM) to fully utilize the characteristics of DN and TN. This method introduces a spatio-temporal hybrid attention module to alleviate the characteristic distribution skew of DN and TN. Furthermore, to balance the privacy protection and training efficiency after multiple modules are integrated into the secure federated linear regression framework, the training strategy of the federated framework and the update strategy of the model are optimized. The simulation results show that the proposed federated method improves the prediction performance by about 4% and has a sub-second response speed.