Learning Without Adversarial Training: A Physics-Informed Neural Network for Secure Power System State Estimation under False Data Injection Attacks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914505773023232 |
|---|---|
| author | Falas, Solon Asprou, Markos Konstantinou, Charalambos Michael, Maria K. |
| author_facet | Falas, Solon Asprou, Markos Konstantinou, Charalambos Michael, Maria K. |
| contents | State estimation is a cornerstone of power system control-center operations, and its robust operation is increasingly a cyber-physical security concern as modern grids become more digitalized and communication-intensive. Neural network-based approaches have gained attention as alternatives to conventional model-based state estimation methods. Physics-Informed Neural Networks (PINNs), which embed power-flow consistency into the learning objective, have shown improved accuracy over existing approaches. This work proposes a PINN-based model for Power System State Estimation (PSSE) that protects the estimation process against the stealth-constrained AC False Data Injection Attacks (FDIAs) considered in this study. The model is developed without adversarial training. Instead, a dynamic loss-weighting formulation based on homoscedastic uncertainty learns the relative scaling of supervised data-fit and physics-residual terms during training, reducing sensitivity to manual weight tuning. Robustness is evaluated on the IEEE 118-bus system using representative stealthy-FDIA families including state distortion, load redistribution, line overloading, and residual-constrained stealth corruption. Performance is measured using Mean Absolute Error (MAE) on voltage magnitudes and phase angles. Results demonstrate higher accuracy and stability than existing fixed-weight PINN variants. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_22784 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning Without Adversarial Training: A Physics-Informed Neural Network for Secure Power System State Estimation under False Data Injection Attacks Falas, Solon Asprou, Markos Konstantinou, Charalambos Michael, Maria K. Machine Learning State estimation is a cornerstone of power system control-center operations, and its robust operation is increasingly a cyber-physical security concern as modern grids become more digitalized and communication-intensive. Neural network-based approaches have gained attention as alternatives to conventional model-based state estimation methods. Physics-Informed Neural Networks (PINNs), which embed power-flow consistency into the learning objective, have shown improved accuracy over existing approaches. This work proposes a PINN-based model for Power System State Estimation (PSSE) that protects the estimation process against the stealth-constrained AC False Data Injection Attacks (FDIAs) considered in this study. The model is developed without adversarial training. Instead, a dynamic loss-weighting formulation based on homoscedastic uncertainty learns the relative scaling of supervised data-fit and physics-residual terms during training, reducing sensitivity to manual weight tuning. Robustness is evaluated on the IEEE 118-bus system using representative stealthy-FDIA families including state distortion, load redistribution, line overloading, and residual-constrained stealth corruption. Performance is measured using Mean Absolute Error (MAE) on voltage magnitudes and phase angles. Results demonstrate higher accuracy and stability than existing fixed-weight PINN variants. |
| title | Learning Without Adversarial Training: A Physics-Informed Neural Network for Secure Power System State Estimation under False Data Injection Attacks |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2604.22784 |