Model Predictive Control of Residential Demand in Low Voltage Network using Ice Storage

Smart buildings and communities with residential storage systems can shift their electricity demand from high-priced peak periods to low-priced off-peak periods. Most of the literature in this field focuses on chemical batteries or hot water tanks as storage devices. Although residential ice storage systems are becoming increasingly popular, there is little research on modeling and evaluating their performance on low voltage networks. In this paper, a model of an ice storage system is presented taking into account its thermodynamic and heat transfer constraints. The model is linearized to take advantage of fast linear program solvers. The linearized model is then used in the framework of model predictive control (MPC) to determine the optimum demand of the buildings to achieve minimum electricity cost. The results of MPC shows that ice storage systems can be effective in shifting the cooling demand to the off-peak periods and improve the voltage profile of the low voltage electricity grid.