Smart residential district. A study of the integration of distributed generation and electric Vehicles

This work investigates the integration of Distributed Energy Generation and Electric Vehicles (EVs) in a residential district of a metropolitan area. First, an overview about the current energy system and its lack of sustainability is provided. Concentrating on the district level, the transition to a new paradigm, referred to as Smart District, is studied considering three involved sectors: building, energy and transports. As for the energy sector, particular attention is paid to the to the impact of Distributed Generation, mainly from renewables, on the management of the electrical network. Concerning transports, EVs are regarded as the most promising among the presented alternative mobility solutions. Their basic characteristics are described and some possible charging strategies are reported in order to analyse different potential interactions with the grid. Secondly, the model of a Smart residential district is proposed; it includes households, private and shared EVs, Photovoltaic (PV) panels and natural gas fuelled microturbines employed in a co-generative mode. Three potential Charging Scenarios (CS) for private cars are considered, leading to different power demand distributions during the day. Then, the amounts of installed power from PV panels and co-generative microturbines minimizing the annual energy cost in the district are found thanks to a global optimization procedure based on a Genetic Algorithm. Calculations are performed for the Base Case, in which no EVs are present in the district, and for the cases corresponding to the three different Charging Scenarios. The presented work has reached three main outcomes: (i) the use of EVs in the district introduces considerable savings with respect to the Base Case, (ii) the impact of the chosen Charging Scenario is nearly negligible under a purely economic perspective even if it is relevant for grid management, (iii) the optimum amounts of installed power vary in a limited range if the distance travelled by EVs and users' departure and arrival time change broadly.