An overview of wind power probabilistic forecasts

Abstract

Over the past one to two decades, probabilistic forecasts have become more popular. Probabilistic forecasts take uncertainty into account and predict a probability distribution function (pdf). This study provides an overview of state-of-the-art technologies on wind-power probabilistic forecasts, and describes the fundamental concepts on those forecasting methods. Additionally, this study also summarizes various methods to evaluate the performance of a probabilistic forecast model. Finally, this study presents a case study using the ensemble prediction model, in which the wind-speed data were provided by the numerical weather prediction system (NWP) of the Central Weather Bureau (CWB) of Taiwan.