Energy self-sufficient households with photovoltaics and electric vehicles are feasible in temperate climate

particular. First, MFBs fail to achieve full self-sufficiency across a wider range of cases, simply because they have less area available for PV installations, relative to their electricity demand. However, MFBs are also substantially more cost-efficient on a per-person basis as investment into components is spread across more inhabitants. Thus, where MFB cases do achieve the potential for self-sufficiency, they ought to be economically more attractive. The wide range of possible parameters to vary meant that assumptions for the creation of the self-sufficient cases needed to be restricted, for example, the assumption that only one third of one facade side is available, or the assumptions we made about the oversizing of hydrogen storage systems to address longer-term variability of PV generation. Based on our sensitivity analyses the results appear robust nevertheless, and we consider our assumptions to generally be conservative. A major uncertainty that we cannot fully address is the economic attractiveness of energy self-sufficiency by 2050. Under our base assumptions, it seems that grid-connected and fossil-powered cases have a cost advantage over self-sufficient cases. How- ever, we cannot forecast 2050 prices, especially not for fuels. Furthermore, we know that con- tinued reliance on fossil fuels is incompatible with preserving a stable global climate. Thus, it is highly likely that government intervention either through bans of or taxes on fossil fuels will change the price balance. Our analyses suggests that with even a relatively modest increase of

fossil fuel costs, some of the self-sufficient scenarios become cost-competitive with grid-con- nected fossil-fired cases. Future work could also investigate the vulnerability of self-sufficient buildings to extreme weather conditions in more detail. By analyzing each weather year indi- vidually, we do not transfer stored hydrogen at end-of-year to the following year; this would result in additional supply security in reality. However, the effect of long dark winter periods could also be more substantial in particularly bad years, given our coverage of weather only from 2000–2015 occur that one year in the future has a worse solar yield than the worst year in the analyzed period. In such cases, hydrogen would have to be purchased on the market or hydrogen backup system size would have to be further increased. Additional work is also needed on performing full environmental impact assessments of self-sufficiency including all upstream energy and material use for different cases, as well as modeling cases located in other destinations worldwide with adjusting the data. Finally, we do not model optimal control of EV charging in combination with building-integrated storage components–implementing Fig 11. Cost sensitivity analysis of self-sufficient cases, showing the range of percent change in costs across all self-sufficient cases for each of the specific numbered adjustments made to the default cost assumptions, and differentiated between new/retrofitted buildings by color and between SFBs/MFBs by color intensity. https://doi.org/10.1371/journal.pone.0227368.g011

Energy self-sufficient households with photovoltaics and electric vehicles are feasible in temperate climate PLOS ONE | https://doi.org/10.1371/journal.pone.0227368 March 4, 2020 20 / 25