Handling Load Uncertainty during On-Peak Time via Dual ESS and LSTM with Load Data Augmentation

This paper proposes a scheduling method of dual ESSs (Energy Storage Systems) for the purpose reducing peak load when there are sudden loads or generation changes during on-peak time. The first ESS is scheduled once day based on day-ahead prediction, and second every 15 min time short-term prediction by LSTM (Long Short-Term Memory). Special attention paid to training using augmented past data which generated adding possible uncertainties temperature data. Based forecast, optimization problems formulated. proposed validated from real building. In other words, applied building energy in case study, it not only shaves interval effectively but also results lower electricity price although interval.