A study on photovoltaic output prediction uncertainty and intermittency compensation method

Abstract

Variable Renewable Energy(VRE) is impossible to maintain stable output due to uncertainty and intermittency. In this paper, Photovoltaic(PV) output is predicted through deep learning prediction technology and compared with the actual PV output. The data used as the input of the deep learning prediction model was partially selected through correlation analysis to reduce overfitting and to have high prediction performance. A Hyperparameter optimization model was used in several deep learning prediction models and the Recurrent Neural Network(RNN) prediction model was selected after comparing the performances. The difference between Predicted PV output and actual PV is compensated using Battery Energy Storage System(BESS). Moreover, the BESS capacity is calculated and the BESS State of Charge (SoC) range profile is observed. Copyright © The Korean Institute of Electrical Engineers This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.