Generating Wind Power Scenarios for Probabilistic Ramp Event Prediction Using Multivariate Statistical Post-Processing

Wind power forecasting is gaining international significance as more regions promote policies to increase the use of renewable energy. Wind ramps, large variations in wind power production during a period of minutes to hours, challenge utilities and electrical balancing authorities. A sudden decrease in wind energy production must be balanced by other power generators to meet energy demands, while a sharp increase in unexpected production results in excess power that may not be sold to the grid, leading to a loss of potential profits. In this study, we assess the performance of the High-Resolution Rapid Refresh (HRRR) numerical weather prediction model in predicting wind ramps with up to twelve hours of lead time at two tall-tower locations in the United States. We validate model performance using 18 months of 80-m wind speed observations from towers in Boulder, Colorado and near the Columbia River Gorge in eastern Oregon.

We employ three statistical post-processing methodologies, two of which are not currently used in the wind industry to correct biases in the model and to generate ensembles of short-term wind speed and power production scenarios. This probabilistic enhancement of HRRR point forecasts provides valuable prediction uncertainty of ramp events and significantly improves the skill of predicting up- and down-ramp events over the raw forecasts. We illustrate how we use one of the ensemble methods to generate forecasts and gain prediction uncertainty based on past historical wind speed scenarios in Fig. 1. We also compute skill scores for each ensemble method at predicting up- and down- ramps to determine which method provides the best prediction of ramp events. These statistical methods can be implemented by wind farm operators to generate a range of possible wind speed and power scenarios to aid and optimize decisions before ramp events occur.