State Estimation for Power System Based on Graph Neural Network

Abstract

The structure of power grid is becoming more and more complex, and the proportion of clean energy in power grid is increasing, which puts forward higher requirements for power system state estimation. The traditional algorithm only uses the measurement data of supervisory control and data acquisition (SCADA) system and wide area measurement system (WAMS) at the same time section for state estimation, fails to make effective use of WAMS measurement data, and the time resolution is low. Therefore, based on graph neural network model, this paper proposes a fast state estimation method of nodes in the whole network. This paper simulates on 57 nodes in New England and generates three different data sets. The example results show that compared with the traditional algorithm, this method can effectively use WAMS measurement data for high-precision and high-time resolution state estimation of the whole network.