Optimización del Consumo Energético en el Hogar mediante Redes Neuronales Recurrentes en el contexto del Caribe Colombiano: Optimization of Energy Consumption in the Home through Recurrent Neural Networks in the context of the Colombian Caribbean

Abstract

In this study, the use of a Long Short-Term Memory (LSTM) recurrent neural network (RNN) model for home energy management (HEMS) has been investigated and evaluated. Using power consumption data from different devices at 15-minute intervals, an LSTM model was implemented to predict the connection of devices and estimate their power consumption. The obtained results showed that the LSTM model achieved satisfactory accuracy in device connection classification and accurate estimation of energy consumption. These findings highlight the potential of the proposed approach to improve energy efficiency in homes by enabling smarter management of device energy consumption. Additionally, the model provides valuable insights into energy consumption patterns, which can help users make informed decisions to reduce their consumption and optimize the performance of their devices.