Optimal Control of Domestic Storage via MPC: the Impact of the Prediction of User Habits, including Power Market and Battery Degradation

Domestic energy storage systems are making their way into the homes of e-vehicle users. They support dependable and cheap recharge during off-peak hours, relieving the power demand through implicit peak shaving. The minimization of charging costs at home relies heavily on the optimal management of the storage system, both in the design and in the control stages. In this work, the design of a battery energy storage system is optimized by means of quadratic programming. The resulting system is continuously governed by a model-predictive control, that accounts for power market and battery degradation online. The performance in terms of long-term system costs is assessed using simulation. An average daily energy consumption is compared to a profile tailored on the user.