Online Rolling Optimization for Energy Efficiency in Smart Homes

Abstract

To cope with the variability of electricity consumption patterns, this study proposes an online rolling optimization-based energy efficiency management strategy for smart homes, which considers user preferences on energy saving and electricity comfort. A weight parameter is used to balance these two objectives and users can set them according to their own electricity preferences. The strategy employs predictive models based on historical data and future inputs to forecast system outputs, and applies feedback correction to compensate prediction errors. In order to better express the power consumption satisfaction of users, two measures of user satisfaction are introduced: utility comfort and temperature comfort. Finally, the simulation result shows that the proposed method achieves an average energy reduction rate of 13.97%, which demonstrates that our strategy can achieve significant reductions in power consumption while enhancing user satisfaction.