Summary
Modern electric power systems are undergoing a profound transformation, primarily driven by the increasing integration of renewable energy sources such as wind and solar power. These renewable resources are typically connected to the grid through power electronic interfaces known as inverter-based resources (IBRs). This fundamental shift introduces new complexities for maintaining grid stability, particularly in "weak" grids where traditional synchronous generation is scarce. Conventional metrics like Short Circuit Ratio (SCR) are static and often fail to account for the complex, non-linear interactions during transients. More advanced indicators, such as Dynamic Stiffness (DS), offer a detailed perspective but are computationally intensive to calculate using traditional electromagnetic transient (EMT) simulations. To address these challenges, this paper proposes a Physics-Informed Neural Network (PINN) framework to estimate dynamic system strength. Unlike previous scalar approaches, this study develops a
Read more Read lessMulti-Input Multi-Output (MIMO) model that accounts for the complex nature of power system dynamics, including voltage magnitude-angles and reactive-active power support. The methodology is validated on a 9 MW grid-connected Doubly-Fed Induction Generator (DFIG) wind farm using a dataset of 300 high-fidelity simulation scenarios. Results demonstrate that the optimized PINN model reduces computational time by orders of magnitude compared to
EMT simulations and provides superior assessment quality compared to traditional SCR by accurately capturing non-linear recovery trajectories that static metrics miss.
Additional informations
| Publication type | Session Materials |
|---|---|
| Reference | C4_10102_2026 |
| Publication year | |
| Publisher | CIGRE |
| Country | Ukraine |
| Study committees |
|
| File size | 1 MB |
| Price for non member | 30 € |
| Price for member | 30 € |
Authors
AGAMALOV Oleg - TPSPP/Independent Researcher
Keywords
Physics-Informed Neural Networks (PINN), Short Circuit Ratio (SCR), Dynamic stiffness (DS), Direct dynamic stiffness (DDS), quadrature dynamic stiffness (QDS)
