Estimating aggregate wind plant capacity from historical time series data

In investigating the large-scale integration of wind plants into the electric system, researchers are often in need of a realistic time series of aggregate wind plant power output corresponding to various capacity levels. One appealing approach to producing the time series is to use normalized historical data, which can then be scaled to the appropriate capacity levels for the analyses. A fundamental challenge in this approach is that the total wind plant capacity level in a system often changes throughout the historical data set, making accurate normalization difficult. This work presents an algorithm for estimating the capacity level for each datum in a data set of aggregate wind plant power output. The algorithm is tested on actual data from the Bonneville Power Administration (BPA) and Electric Reliability Council of Texas (ERCOT) systems. It is shown that capacity levels estimated by the algorithm follow the trend of reported values and account for incremental increases in wind plant capacity.