Solar power production forecasting based on recurrent neural network

Abstract

The Active energy unit is a complex system that is independent of an external source of electric energy. It uses mainly renewable sources namely a wind turbine and a photovoltaic power plant. Because of a stochastic character of renewable sources it is important to implement the Active Demand Side Management to control energy flows in the system and manage plans of connected appliances to preserve safe working. Forecasting of solar power production from photovoltaic power plant is one of the most important parts in the system. This paper presents a solar power production forecasting model based on the Recurrent Neural Network.