Rapid Evaluation of Power System Transient Stability Based on Fusion of Data-driven and Time-domain Simulation

With the rapid development of the large-scale renewable energy and the AC/DC hybrid power systems, the proportion of components with power electronic characteristics in the power system has increased rapidly so that the power electronic characteristics in the power system becomes increasingly obvious. Aiming at the slow calculation speed of the time-domain simulation in the transient stability assessment of the power electronic power system, this paper proposes a fast transient stability assessment based on the fusion of data-driven and time-domain simulation. Starting with the calculation process of the time domain simulation, this method establishes a parallel evaluation framework of transient stability driven by the hybrid-driven model composed of a physics-driven model and a data-driven model, and uses the knowledge-enhanced data-driven model to replace the differential equations in the power electronic components to speed up the time domain simulation calculations. On this basis, the data is preprocessed according to the Stride Euler method proposed in this paper to realize the stride prediction of the data-driven model, changing the small step size of the power electronic components to a large step size consistent with the synchronous generator to further speed up the simulation. In this paper, the improved IEEE14 node model is used to verify the proposed algorithm and model. The results show that the proposed method is able to maintain a high accuracy while improving the simulation speed. © 2023 Power System Technology Press. All rights reserved.