Optimal Energy Management of Residential Solar PV with Battery Storage: Effects of Fast Load and Generation Transients

Abstract

An energy management algorithm to optimally control a rooftop solar photovoltaic device in combination with a battery storage and considering variable pricing is analyzed for single family houses. A discrete time dynamic programming controller is developed to minimize the electricity costs. It is shown in simulation, using real world measurement data, that considerable savings can be achieved with the optimal controller. However, we also show the limitations of such discrete time approaches in realistic scenarios when load and generation changes in-between the samples where the controller is active. This can lead to a performance deterioration making the optimal controller worse than a standard approach.