Pedro M.E. Pinto, Gerardo J. Osório, Mohammad S. Javadi, João P.S. Catalão
• Innovative energy management modelling system considering multiple energy sources. • Developing a self-scheduling model for smart homes considering users’ comfort index. • Considering a comprehensive analysis of seasonal demand in real-world studies. • Formalising hysteresis-based thermoelectric devices for energy management models. Rising electricity prices and increasing energy demand have heightened the need for efficient, resilient residential energy use, underscoring the role of end-users in the consumer engagement roadmap. Home energy management systems, when combined with renewable generation, energy storage, and electric vehicles, offer a promising pathway to reduce bills while maintaining end-user comfort. This study develops a fast, mixed-integer linear programming model for smart home energy scheduling that considers diverse load types under seasonal operating conditions and different demand response tariffs. The model includes indoor temperature control via a thermostat-based air conditioning system, home electric vehicle charging, an energy storage unit, and an electric water heater for hot water supply, ensuring a comprehensive representation of typical residential energy usage. The proposed model converges in less than 1 s, making it suitable for real-time and model-predictive control applications. The model’s performance has been investigated over two typical weeks in the summer and winter seasons. Simulation results demonstrate that in the Summer, the maximum weekly savings ranged from –2.12 % to 36.25 %, while in the Winter, the maximum savings ranged from –5.19 % to 35.97 % across different tariff structures and demand response schemes. These findings highlight the strong potential of integrated self-scheduling approaches to reduce household energy costs while maintaining user comfort across multiple flexible energy assets.