PowerBench is a comprehensive benchmark suite for evaluating machine learning methods on resilience oriented grid-monitoring and security tasks in power distribution networks. It includes structured datasets generated synthetically using distribution network models in OpenDSS and standardized evaluation scenarios for:
- Line Failure Detection
- Cyberattack Detection
- State Estimation
Each dataset is organized by task type and IEEE test feeders (34-bus, 123-bus, 8500-node).
PowerBench/
│
├── Datasets/ # All datasets organized by task
│ ├── Cyber Attack Detection/
│ ├── Line Failure Detection/
│ └── State Estimation/
│
├── Experiments/ # Scripts or model files for evaluating datasets
├── README.md # You are here!
| Task | Description |
|---|---|
| Cyber Attack Detection | Detect attacks on EVCS, PV, and sensors and locate compromised devices |
| Line Failure Detection | Identify if lines have failed using partial obervability in unbalanced distribution networks |
| State Estimation | Estimate voltage magnitudes using partial measurements |
To evaluate or benchmark models, refer to the Experiments/ directory.
Each dataset folder includes a dedicated README.md file containing details about the dataset generation.