Federated Learning Detection of Cyberattacks on Virtual Synchronous Machines Under Grid-Forming Control Using Physics-Informed LSTM
Bibliographic record
Abstract
The global shift toward clean production, like using renewable energy, has significantly decreased the use of synchronous machines (SMs), which help maintain stability and control, causing serious frequency stability issues in power systems with low inertia. Fractional order controller-based virtual synchronous machines (FOC-VSMs) have become a promising option, but they rely on communication networks to work together in real time, causing them to be at risk of cyberattacks, especially from false data injection attacks (FDIAs). This paper suggests a new way to detect FDI attacks using a federated physics-informed long short-term memory (PI-LSTM) network. Each FOC-VSM uses its data to train a PI-LSTM, which keeps the information private but still helps it learn from a common model that understands various operating conditions. The PI-LSTM incorporates physical constraints derived from the FOC-VSM swing equation, facilitating residual-based anomaly detection that is sensitive to minor deviations in control dynamics, such as altered inertia or falsified frequency signals. Unlike traditional LSTMs, the physics-informed architecture minimizes false positives arising from benign disturbances. We assessed the proposed method on an IEEE 9-bus test system featuring two FOC-VSMs. The results show that our method can successfully detect FDI attacks while handling regular changes, proving it could be a strong solution.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".