Building Flexibility using Deep Learning

In the power system, power data analysis, such as electricity consumption predictions, have always been used as a basis for adjusting electricity prices, substation regulation, total load prediction, and peak avoidance management. In order to create intelligent green cities, we have to increasingly explore the flexibility potential of commercial buildings with the increasing output of renewable energy sources (RES) and the increasing number of electric vehicles (EV).It is noteworthy, however, that when a variety of activities are analyzed such as cooling, heating, fans, lights, and equipment, the volume of consumption data is significantly higher than it was previously. We present here a way to identify the flexibility of commercial buildings from very large datasets using a big data processing framework.This paper uses deep learning algorithm LSTM (Long-Short Term Memory) to predict the future electricity usage.