Physics-guided lightweight detection of stealthy false data injection attacks under communication impairments and model shift
Abstract
Abstract Stealthy false data injection attacks can alter power-system state estimates while preserving the residual used by conventional bad-data detection. This study develops and critically evaluates a physics-guided lightweight detector (PG-LFD) for Internet-of-Things-enabled smart grids. The method combines DC weighted least-squares state estimation, temporal and topology-derived features, Fisher-ranked LogitBoost classification, and a shrinkage-Mahalanobis normality score with validation-calibrated fusion. A reproducible MATLAB/MATPOWER protocol is established on the IEEE 14-bus and 39-bus systems under measurement noise, packet loss, delay, attack-strength variation, and an AC-operating-point/DC-detector model shift. Across ten independent trials, PG-LFD achieved F1 scores of 0.953 ± 0.038 and 0.953 ± 0.028, with area-under-curve values of 0.977 ± 0.022 and 0.973 ± 0.020. It outperformed residual, linear, and several support-vector baselines, but did not exceed random forest or raw-feature LogitBoost under nominal conditions. At 3% noise on the 39-bus system, PG-LFD attained the best F1 among the nonlinear methods (0.896) while limiting the false-alarm rate to 6.8%, compared with 22.8% for raw LogitBoost. Inference latency was 0.040–0.053 ms per sample. Ablation showed that the one-class branch did not improve nominal decisions, and AC/DC shift caused 100% false alarms despite moderate ranking performance. The results support physics guidance as an interpretable,…
Read the paper
Related studies
- Preliminary Toxicity Analysis of 3DCRT versus IMRT on the High Dose Arm of the RTOG 0126 Prostate Cancer Trial
- Adaptive Radiotherapy for Head and Neck Cancer: Initial Clinical Outcomes from a Prospective Trial
- American Association of Physicists in Medicine Task Group 263: Standardizing Nomenclatures in Radiation Oncology
- Motion Management Strategies and Technical Issues Associated with Stereotactic Body Radiotherapy of Thoracic and Upper Abdominal Tumors: A Review from NRG Oncology
- High-dose-rate interstitial brachytherapy as a monotherapy for localized prostate cancer: treatment description and preliminary results of a phase I/II clinical trial
- Highly efficient training, refinement, and validation of a knowledge-based plan quality control system for radiotherapy clinical trials
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.