The forecast of the electrical energy generated by photovoltaic systems using neural network method

The purpose of this paper is to forecast the electrical energy generated by photovoltaic systems using the method of neural network. A database, which includes the actual measured electrical energy and the parameters of weather conditions that can influence the electrical energy generated by the photovoltaic system (PV system), is established in advance in order to be used in electrical energy forecasts. The Matlab/Simulink software is used in this paper to set up a neural network model with the learning algorithm of back-propagation network in order to forecast the generated electrical energy of the PV system. After observing the results of electrical energy forecast and divergence evaluation, it can be found that the proposed neural network model can accurately forecast the generated electrical power and output current under different weather conditions. The feasibility and accuracy of the proposed forecast system is then validated.