Big Data and Deep Learning Platform for Terabyte-Scale Renewable Datasets

Many renewable resources cover diverse geographical areas and it is increasingly important to analyze large sets of data to understand their spatial and temporal behaviors. In this paper, we propose and demonstrate a data platform to efficiently manipulate and visualize data on the scale of terabytes. As the application of interest, we focus on visualization and forecasting of wind power over large geographic areas at various different spatial and temporal resolutions. In particular, we show how to balance the amount of data used and the need for computational efficiency in real-time applications. The main data set we use is the recently released terabyte wind dataset by NREL.