Hybrid Artificial Neural Networks for Electricity Consumption Prediction

We present a comparative study of electricity consumption predictions using the SARIMAX method (Seasonal Auto Regressive Moving Average eXogenous variables), HyFis2 model (Hybrid Neural Fuzzy Inference System) and LSTNetA (Long Short Time series Network Adapted), hybrid neural network containing GRU (Gated Recurrent Unit), CNN (Convolutional Network) dense layers, specially adapted for this case study. The experimental developed showed superior result with much closer to real consumption. in had rmse (root mean squared error) 198.44, 602.71 604.58.