A residential virtual power plant of electric vehicles and batteries for a resilient grid during normal and extreme weather

This paper proposes a virtual power plant (VPP) framework for aggregating and coordinating dispersed home electric vehicles (EVs) and batteries to support distribution system resilience under normal and extreme weather conditions. The framework includes a day-ahead optimization designed to address the worst scenario within the uncertainty set. The uncertainties of forecasted weather, as well as household demand and EV availability, are considered. For regular weather, VPP aims for optimal profit, adequate daily driving energy, and operating voltage within established limits. For extreme weather, VPP coordinates EVs and batteries to meet total demand with zero grid power. However, during expected driving hours, the VPP optimizer will propose that a minimum number of EVs remain connected to home chargers. The experiments are carried out on a modified IEEE 33-bus system, which includes a wind turbine and 992 houses outfitted with rooftop photovoltaics, batteries, EVs, and electric appliances. The system is modeled using mixed-integer linear programming, and the voltage and losses are computed using data-driven linear regression. The results show that aggregated EVs and batteries can assist grid resilience during storms and periods of low wind speed, as well as support distribution voltage during normal weather, besides economic benefits. • A residential VPP is proposed for aggregating EVs and batteries at homes to maintain the grid resilience besides profit. • A new multi-objective optimization function is proposed to adapt the objectives of the normal and extreme events. • A modified IEEE 33-bus system is proposed including 992 houses with rooftop PV systems, stationary batteries, and EVs. • New resilience metrics are offered to assess the resilience of distribution system under extreme weather conditions.