Community-based resilient electricity sharing: Optimal spatial clustering

This paper extends our proposing (Yamagata and Seya 2012) concept of a community-based disaster resilient electricity sharing system (DRESS) as a complement or an alternative to the feed-in-tariff (FiT) to achieve CO2 neutral in cities. In this system, electricity generated from widely introduced solar photovoltaic panels (PVs) is stored to the “cars not in use” in a city. In the central part of the Tokyo metropolitan area, almost half of the cars is used only on weekends and are kept parking during the weekdays. Hence, there exists a huge new potential if those cars are replaced by electric vehicles (EVs) in the future, namely they may be used as new battery storages using vehicle to grid (V2G) at a community level. This study extends our previous paper. Firstly, by using actual ground areas of buildings, we estimate PVs supply potential more accurately. The result shows that the hourly electricity surplus (PV supply minus demand) can be fully stored without waste if 27% of the parking EVs are used as battery storage at the whole city level, although there exist significant spatial differences at local district level. Secondly, based on the geographical demand-supply estimates, we check the possibility of local electricity sharing by combing high and low storage potential districts to form electricity self-sufficient resilient communities. Finally, we analyze the optimal community clustering using Moran's I index. We show that the 40%, instead of 27%, is an optimal EV electricity sharing rate, if we consider the resilience against black-out risk.