Reliable time series of wind power capacity are critical for analysing past production and for normalising historical data when training forecasting models. Databases of wind power installations can be used to create time series of installed capacity, but are often incomplete and temporally coarse. Any inaccuracies in magnitude or timing of installed capacity propagate into downstream analyses.

The Combined Power-curve and Quadratic optimization (CPQ) method iteratively fits a generic power‑curve model to derive a capacity factor time series, and solves a constrained quadratic program that identifies the capacity time series most consistent with observed production.

We validate the CPQ method on the high-quality data of the Danish wind power system, showing that it effectively reconstructs historical capacity time series with a MAPE around 1.9\% and a reduction in MAE of 22\% compared to the cumulative maximum reference method. Further, the CPQ-derived capacity time series can be used to normalise training data for a forecasting model to reduce forecast MAE by 1.2\% compared to using capacity time series based on database information. Requiring only aggregated production and publicly available wind data, the CPQ method can potentially fill information gaps in regions with poor data coverage.