Conformal prediction improves wind forecasts
Original title: The Cost of Coverage: Conformalising Native Forecast Uncertainty for GB Wind Power
Regime-specific calibration fixes high-wind undercoverage while narrowing bands.
Conformal prediction applied to native forecast uncertainty bands. Six forecasters tested at two GB wind farms under operational conditions. Calibration repairs marginal coverage with little penalty but regime-level calibration cuts high-wind out-of-band volume by over 40 percent.
- Six forecasters compared from ARIMA to foundation model
- GB wind farms under operational conditions
- High-wind undercoverage fixed by regime calibration
- 40 percent reduction in out-of-band volume
- Bands narrowed overall
From the original article
1 min read at arxiv.orgUncertainty bands on wind forecasts help inform reserve sizing and commitments, but often fail to deliver their stated coverage. We apply conformal prediction to native bands to see what it adds and what it costs, comparing six forecasters, from ARIMA to a tabular foundation model, at two wind farms in Great Britain (GB) under operational conditions.
By Carl Somers · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.