Data Management System for Photovoltaic Solar Environment

Abstract

As communication environments and Internet-of-Things technologies evolve, smart grid is being studied for collecting and predicting facility data in real time. In this paper, we propose a management system that collects the data of a photovoltaic system and predicts the amount of power generation. For this purpose, we designed an environment that can collect solar data, adjacent environment data, and facility information in real time. In addition, to expand the limits of previous power generation prediction studies, a power generation prediction experiment was conducted using recurrent neural network-based methods. The experimental results are compared with those of previous studies.