MPC-based robust optimization of smart apartment building considering uncertainty for conservative reduction

Undergoing a significant transition is demand-side management (DSM), brought about by changes in the energy landscape, including the extensive deployment of renewable energy sources (RES). One of the most important challenges in DSM is dealing with uncertainty. Robust optimization methods are widely known as an approach to deal with uncertainty. However, this method tends to be conservative in its solution because it considers worst-case scenarios. In this study, we develop a robust approach of model predictive control (MPC) to reduce conservativeness and maintain robustness in DSM of smart apartment building (SAB). Consisting of two stages is the robust optimization strategy employed in this study. The first stage involves day-ahead scheduling, where four different optimization approaches are implemented. In the second stage, intra-day scheduling is implemented on the basis of day-ahead operational plan. The application of this technique achieves reduced conservativeness and maintains robustness. The simulation results show that by using the proposed approach, compared to conventional robust optimization, while maintaining robustness, the operation cost of day-ahead scheduling can be reduced by 10.79% and the operation cost of intra-day scheduling can be reduced by 15%.