Robust Federated Learning for Energy Storage Systems
Bibliographic record
Abstract
One of the Sustainable Development Goals of the United Nations is affordable and clean energy. True utilization of renewable energy is only possible via battery-based energy storage systems. Overseeing the operation of battery-based energy storage systems and diagnosing abnormal batteries are of the utmost importance for their durability and stability. Because of inadequate anomalous samples and privacy considerations, we jointly train a global autoencoder on various battery-based energy storage systems to detect anomalous batteries. Due to potentially unstable network connectivity in energy storage systems, a chunk of model parameters may be lost during model transmission, leading to dramatic performance deterioration. The trained model tends to classify all measurements as anomalies. To solve this problem, we propose a robust federated learning scheme to mitigate negative impact caused by packet loss during model transmission. By permuting and unpermuting model parameters before and after model transmission, we are able to distribute the lost parameters across the entire model. Such a loss can no longer have a significant negative impact on anomalous battery detection. Experimental results illustrate that the proposed algorithm is robust against packet loss during the model exchange between the cloud server and battery-based energy storage systems.
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 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".