Prediction of Spanish energy pricing and consumption based on hybrid deep learning model

The ongoing COVID-19 pandemic and the current crisis in Ukraine are two significant challenges facing the energy supply sector that affect the stability and efficacy of an energy system. The importance of energy pricing and the need for accurate models to compute prices and use are highlighted in this study, which focusses on Spain. This study uses CNN, Bi-LSTM, and GRU, among others, to anticipate power usage and pricing using hourly data. To forecast energy prices and consumption in Spain, the authors suggest a model they term a hybrid CNN-Bi-LSTM-GRU model. To assess the suggested models' performance, they have used the RMSE, MAE, and MAPE performance measures. For energy prices, it achieved a test RMSE of 5.032 and MAPE of 8.414, while for consumption, it attained a test RMSE of 18.92 and MAPE of 1.065, respectively. According to our findings, the hybrid model is the most effective model for simulating price and consumption with a small degree of significance.