In this paper, domestic homes with a photovoltaic system, a battery storage and variable electricity tariffs are considered. First, a control oriented model of the system is developed. Second, three controllers are designed with the aim to minimize costs for the user, and at the same time help the energy provider maintaining grid stability, by proper battery load management. A base-level controller serves as performance reference for the advanced optimization based controllers (a nonlinear controller with full system knowledge and a linear controller) which utilize forecasts of energy production and consumption. Performance evaluation was done in simulation using real-world measurement data over a full year. Results show that significant cost reductions can be achieved with the nonlinear controller while the linear one suffers from imprecise system knowledge.