Energy Storage in Smart Homes: Grid-Convenience Versus Self-Use and Survivability

The number of local power generation units, such as photovoltaic panels (PV), increased enormously in recent years. Their production patterns are highly variable and depend on the current weather. The resulting fluctuation in production poses a major challenge to the stability of the power grid. The use of local energy storages may help to ensure that the locally produced power is fed into the grid in a grid-convenient way. It may also help clients to increase the self-use of locally generated power and to increase the so-called survivability of their homes in the presence of a power outage. This paper compares the interest of the power operator, i.e. grid-convenience with the interests of the user, i.e. self-use and survivability for different battery management strategies: i) direct loading, ii) delayed loading and iii) peak shaving. We use a Hybrid Petri Net model with one stochastic variable (HPnG) to model smart homes with local power generation, local storage and different battery management strategies in the presence of power outages. Recent algorithms for analyzing and model checking HPnGs enable the computation of the above mentioned measures of interest. We are able to show that whenever good predictions of production and demand exist, grid-convenience does not decrease the survivability and the self-use of a smart home.