A Stacked Ensemble of Attention-Augmented Deep Learning Models for Robust Anomaly Detection in Smart Grids
Bibliographic record
Abstract
The detection of anomalies in smart meter electricity data represents a critical task for ensuring power grid stability and security, an endeavor imperative for grid modernization. This undertaking is complicated by complex temporal patterns, seasonality, and subtle irregularities that challenge conventional detection methods. To address this, a novel attention-augmented deep ensemble framework is proposed. The methodology involves the independent training of four diverse neural architectures-a Bidirectional Long Short-Term Memory (BiLSTM), a Gated Recurrent Unit (GRU), a Temporal Convolutional Network (TCN), and a Transformer model-all enhanced with attention mechanisms to capture a wide spectrum of temporal dependencies. The individual predictions are subsequently integrated via a stacking ensemble utilizing an eXtreme Gradient Boosting (XGBoost) meta-learner. By leveraging the unique strengths of recurrence, convolution, and self-attention, the hybrid architecture achieves superior anomaly detection capabilities. Empirical evaluations and ablation studies on public energy consumption data demonstrate that the framework attains a state-of-the-art F1-score of 0.572 and an AUC of 0.927, significantly outperforming individual baseline models and other deep learning architectures. The result is a practical and reliable solution for intelligent energy monitoring that combines interpretability with robust performance.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| 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".