Learning Without Adversarial Training: A Physics-Informed Neural Network for Secure Power System State Estimation under False Data Injection Attacks

TL;DR

This paper introduces a physics-informed neural network model for power system state estimation that is robust against false data injection attacks without using adversarial training, employing a dynamic loss-weighting scheme.

Contribution

It proposes a novel PINN-based PSSE approach with dynamic loss weighting to enhance robustness against cyber-attacks, avoiding adversarial training.

Findings

Higher accuracy in voltage and phase angle estimation compared to fixed-weight PINNs.

Demonstrated robustness against various stealthy false data injection attack types.

Achieved stable performance on IEEE 118-bus system under attack simulations.

Abstract

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.