SDWPF: Spatial Dynamic Wind Power Forecasting for 134 Turbines

SDWPF is a dataset collected over two years from a wind farm with 134 turbines, detailing spatial layout and dynamic context factors. It was used to launch the ACM KDD Cup 2022, attracting over 2,400 teams worldwide. The associated paper was published in Scientific Data in 2024.

Use Cases

• Forecasting wind power output based on spatial turbine layout and dynamic context factors mentioned in the description.
• Benchmarking spatio-temporal machine learning models against KDD Cup 2022 competition results.
• Analyzing the impact of weather conditions and turbine elevation on power generation using the sdwpf_full dataset.

Strengths

• Data spans two years of collection from a real-world wind farm.
• Includes spatial layout data for 134 individual turbines.
• Was the basis for a major competition with over 2,400 participating teams.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Associated with a paper published in Scientific Data (2024) and an earlier arXiv preprint (2022).

Collection Method

Collected from a wind farm.

Time Range

Data collected over two years (specific years not stated).