PINN-based reconstruction of frequency spatiotemporal characteristics in power systems

Integrating renewable energy sources into power grids has led to prominent electromagnetic and frequency oscillation challenges. Analysing these oscillation mechanisms is inherently complex due to the lack of analytical solutions and the difficulties in accurate modelling. This paper focuses on mechanical oscillations as a representative case study to explore the application of Physics-Informed Neural Networks (PINNs) for analysing and solving oscillation problems in power systems. We demonstrate the effectiveness of PINNs through numerical results, establishing a foundation for future research on modelling and solving grid oscillations under more complex scenarios.