Forecasting Solar Power Ramp Events Using Machine Learning Classification Techniques

The growing integration level of wind and solar energy resources introduces new challenges for the reliable operation of the electric grid. Tasks such as managing high ramp-rates of renewable generation, optimal energy management of energy storage systems, and voltage regulator settings on feeders with distributed generation, may be improved with the availability of solar power forecasts, especially those with accurate ramp event prediction. This paper presents classification techniques to classify and forecast the solar power ramp events. A case study over an entire year is conducted and several evaluation metrics are considered to assess the performance of the classification models of solar power ramp event forecasts.