Power System Contingency Analysis with Graph and Topology Measurements

Featuring results of power-system contingency analyses along with graph and topology measurements for modeling system responses to disruptions. It was developed for machine-learning predictive modeling and is a companion to a 2020 NREL presentation.

Use Cases

• Train machine learning models to predict power-system responses from graph and topology measurements under contingency scenarios.
• Analyze the relationship between topological data analysis features and system resilience metrics from contingency analyses.
• Build predictive models for power-system disruptions using simulation results and graph theory measurements.

Strengths

• Dataset is associated with a formal NREL presentation (PR-6A20-77485) from the Department of Energy.
• Data is specifically designed for machine-learning predictive modeling of power-system responses.
• Covers multiple analytical aspects including contingency analysis, graph measurements, and topology measurements.

Limitations

• Specific data volume (rows, columns) and file size are unknown, limiting assessment of scale.
• Data format is BIN, which may require specialized tools or libraries for access and processing.
• Sample data is unavailable, making initial inspection of data structure and quality difficult.

Provenance

Source

Department of Energy, National Renewable Energy Laboratory (NREL).

Collection Method

Results from power-system contingency analyses and simulations.

Freshness

Last updated on 2026-03-21.