Smart Energy Forecasting: Energy Consumption Prediction in Data Centers

The rapid expansion of data centers has been driven by the growing demand for digital services. As a consequence, the energy usage follows the real challenges in terms of cost and environmental effect. As a result, this study proposes a hybrid technique for predicting energy consumption in data centers that combines machine learning approaches, notably Long Short-Term Memory (LSTM) networks and Auto-Regressive Integrated Moving Average (ARIMA) models. This approach builds a model capable of accurately capturing both short-term variations and long-term trends in energy consumption. As a model for optimizing energy use in data centers, the suggested approach would save costs while potentially contributing to sustainability through improved operational efficiency. This research also shows the need for real-time prediction techniques in energy management while making decisions that support processes with a visual dashboard and alert system for mitigating risks of overconsumption of energy.