Enhancing Advanced Time-Series Forecasting of Electric Energy Consumption Based on RNN Augmented with LSTM Techniques

Forecasting energy consumption remains an important endeavor, given the indispensable role of energy in human existence and most economic activities. In recent years, the growing importance of Artificial Intelligence and Machine Learning approaches to forecasting energy consumption has become increasingly pronounced. These advanced techniques are adept at processing large datasets, enabling nuanced analyses of consumption patterns and precise rate determinations. However, energy overproduction leads to resource depletion and increased operational costs, underscoring the need for careful monitoring and accurate forecasting of energy consumption rates. Such forecasts are essential for devising efficient, safe, and sustainable energy management strategies, optimizing equipment maintenance protocols, and facilitating informed production decision-making. This paper strives to provide a predictive model for hourly energy consumption rates leveraging Recurrent Neural Network (RNN) enhanced with Long Short-Term Memory (LSTM) Deep Learning techniques. These techniques have demonstrated effective results and performance, receiving an outstanding R-squared of 99.00%.