Resumen
The large scale deployment of electric vehicles in cities will play a key role over the next decades to reduce air-pollutants in densely populated areas. This will imply a shift of the energy demand from the oil sector to the electric energy utilities, and hence the development of an adequate recharge infrastructure to meet the electricity demand from electric vehicles. Moreover, the full integration of electric vehicles within future smart cities will call for the development of new Vehicle-to-Grid applications, enabling the grid operators and end-users to dynamically interact with the vehicles. The purpose of this paper is to demonstrate how driving patterns and data mining can be used for the design of a smart recharge infrastructure and for identifying Vehicle-to-Grid hot-spots over vast geographical areas. The analysis relies on a database of approximately 16,000 vehicles and 2.6 million of parking events, interfaced with an electric vehicle model and two recharge behavioural models. The results help to understand to which extent battery electric vehicles can replace conventional fuel cars under real-world constraints, quantifying their electric energy demand and their recursive parking locations, to design a customised recharge infrastructure and identify Vehicle-to-Grid hot-spots. This paper provides a description of the developed model, analysing its potential and limitations and underlining its possible future applications for smart cities, providing new insights towards the definition, assessment and development of future energy and transport policy.