Photovoltaic array power forecasting model based on energy storage

Abstract

With the rapid increase of the capacity in photovoltaic (PV) generated systems, how to deal with the problem caused by the random output in the system becomes more significant. One possible solution could be the use of energy storage. The forecasting output can be obtained by the support vector regression model (SVR) introduced in this article, then the capacity of energy storage can be optimized by the difference between actual and predicting outputs. That is to say, energy storage devices are taken to compensate the difference, so that the deviation between predictions and actual values can be decreased. The results show that the proposed algorithm ELSSVR is effective and the installed capacity of energy storage is reduced significantly.