In the rapidly evolving technology landscape, smart homes are becoming increasingly common, driven by the demand for integrated management of information and services. However, despite advancements, managing electricity consumption efficiently remains a significant challenge, primarily due to the lack of detailed usage data and the complexity of predicting device behavior. This study addresses these challenges by utilizing the DinRail Cerberus meter for granular data collection on household electricity use and applying the DBSCAN clustering algorithm for unsupervised learning. Our research aims to develop a forecasting system that accurately discerns the operational status of household devices—active or inactive—based on energy consumption patterns. This innovative approach promises to revolutionize energy management in smart homes, offering detailed insights into device usage that facilitate more informed decisions for efficient electricity consumption. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.