Speaker
Description
Rising electricity prices necessitate more efficient management of household electricity consumption. Home Energy Management Systems (HEMS) provide a technological foundation for such management; however, their practical effectiveness depends critically on their analytical capabilities. This study presents a comparative analysis of four representative HEMS platforms for household electricity management — Google Nest, Sense, Smappee, and Home Assistant — evaluated according to five key criteria: the presence of an AI-based forecasting component, model adaptability, explainability of recommendations with quantitatively expressed financial impact, hardware independence, and compliance with local tariff structures. The results indicate that none of the examined systems simultaneously satisfies all criteria, with each covering only a subset of functionalities. Based on these findings, a conceptual model, EnergyForecast, is proposed, integrating four main components: a hybrid forecasting model (XGBoost+LSTM) with adaptive weighting across forecasting horizons; a mechanism for concept drift detection and automatic retraining; an explainable recommendation module with quantitatively evaluated financial impact; and an interactive simulator for scenario-based planning. The proposed model is designed to be hardware-independent and adaptable to local energy conditions.