Calculation Method of Available Transfer Capacity Based on Graph Convolutional Network

Studying and evaluating the available transfer capacity (ATC) of transmission lines is of great significance when determining the reasonable operation mode of a power grid and optimizing the allocation of power grid resources. With the rapid development of deep learning technology, it is possible to quickly find the most ATC by mining historical data and prior knowledge without relying on physical model. Therefore, an ATC calculation method based on graph convolutional network (GCN) is proposed. The topology information between the nodes of a power grid is represented by adjacency matrix. The proposed algorithm can effectively mine the correlation between the nodal injection power, and map the complex non-linear relationship between ATC and power data with deep graph convolutional architecture. The simulation results show that the calculation accuracy of ATC based on GCN can meet the needs of practical application of power grid.