IoT-based Smart Energy Monitoring and Forecasting using Long Short-Term Memory and Cloud Integration

This paper describes a combined Internet of Things (IoT) and machine learning approach to enable energy efficiency in smart homes. It has been established that intelligent systems to reduce energy waste, lower operational costs, and lead to sustainable consumption habits are necessary. The use of electric vehicles, battery storage, and a detailed Home Energy Management Systems (HEMS) framework, provides high sampling frequency data and analysis. This approach allowed the HEMS framework to perform dynamic forecasting in conjunction with automatic fault detection, using the LSTM AI algorithms trained on sensor readings. Overall, the results indicate that the framework was able to improve energy performance of households and support sustainable operation and predictive maintenance for future homes.