Short-Term Electric Load Demand Forecasting on Greek Energy Market using Deep Learning: A comparative study

In this paper, we deal with the short-term Electric Load Demand Forecasting problem, considering the Greek Energy Market. Particularly, we focus on two short-term cases, namely one-hour-ahead forecasting and one-day-ahead forecasting. The objective of this paper is to provide a comparative study considering the aforementioned problems on Greek Energy Market. To this end, we implement a wide variety of deep learning models (e.g., MLPs, CNNs, LSTMs, GRUs) that have been utilized in the recent literature either considering the ELDF task or generic time-series forecasting tasks. The extensive experimental evaluation has led to useful remarks.