Photovoltaics (PV) and other renewables sources have widely spread in recent years. Although those sources provide an environmentally-friendly solution, their integration is a real challenge in terms of power management as it depends on meteorological conditions. The ability to predict those variable sources considering meteorological uncertainty plays a key role in the management of power dispatching plans and grid charge control.
The aim of the project is to provide a forecasting model to set the day-ahead grid electricity need for ESR, a local energy distributor. Therefore, our research team has developed an easy-to-use methodology to predict PV production using time series analyses and sampling algorithms.
While most of PV production forecast studies use optimized neural networks, the main novelties of our approach is to provide a solution easy to implement and flexible. The method combines classification algorithms to predict the PV plant efficiency considering weather conditions and nonlinear regression to predict weather forecasted errors in order to improve prediction results. The results are based on the data collected from the PV plants of ESR.