This work investigates the extent to which domestic energy storage, in the form of batteries, can increase the self-consumption of electricity generated by a photovoltaic (PV) installation. The work uses real world household energy consumption data (measurements) as the input to a household energy consumption model. The model maximizes household self-sufficiency, by minimizing the amount of electricity purchased from the grid, and thereby also maximizing the level of self-consumption of PV electricity, i.e., the amount of PV-generated electricity that is consumed in-house. This is done for different combinations of PV installation sizes (measured in array-to-load ratio; ALR: ratio of the PV capacity to the average annual electric load of a household) and battery capacities for different categories of single-family dwellings in Sweden (i.e., northern latitudes). The modeling includes approximately 2000 households (buildings). The results show that the use of batteries with capacities within the investigated range, i.e., 0.15-100 kW h, can increase the level of self-consumption by a practical maximum of 20-50 percentage points (depending on the load profile of the household) compared to not using a battery. As an example, for a household with an annual electricity consumption of 20 MW h and a PV installation of 7 kW,,, this range in increased self-consumption of PV-generated electricity requires battery capacities in the range of 1524 kW h (actual usable capacity), depending on the load profile of the specific household. The practical maximum range is determined by the seasonality of PV generation at Swedish latitudes, i.e., higher levels of increased self-consumption are possible, however, it would require substantially larger batteries than the up to 100 kW h investigated in this work. Thus, any additional marginal increment in battery capacity beyond the range investigated results in a low level of utilization and poor additional value. Furthermore, our results reveal that when a battery is used to store PV-generated electricity in-house, self-sufficiency increases (as compared to not using a battery) by 12.5-30 percentage points for the upper range of the investigated PV capacities (ALR. of 6). (C) 2016 Elsevier Ltd. All rights reserved.