Model Predictive Control for Real-Time Residential Energy Scheduling under Uncertainties

This paper proposes a real-time strategy based on Model Predictive Control (MPC) for the energy scheduling of a grid-connected smart residential user equipped with deferrable and non-deferrable electrical appliances, a renewable energy source (RES), and an electrical energy storage system (EESS). The proposed control scheme relies on an iterative finite horizon on-line optimization, implementing a quadratic cost function to minimize the electricity bill of the user's load demand and to limit the peak-to-average ratio (PAR) of the energy consumption profile whilst considering operational constraints. At each time step, the optimization problem is solved providing the cost-optimal energy consumption profile for the user's deferrable loads and the optimal charging/discharging profile for the EESS, taking into account forecast uncertainties by using the most updated predicted values of local RES generation and non-deferrable loads consumption. The performance and effectiveness of the proposed framework are evaluated for a case study where the dynamics of the considered residential energy system is simulated under uncertainties both in the forecast of the RES generation and the non-deferrable loads energy consumption.