AI to find optimal electric car recharge point locations

By 24 abril, 2018Technology

Researchers at the Universitat Politècnica de València have developed a tool based on Artificial Intelligence techniques which make it possible to study in which spots of cities electric car charge stations are necessary – or if they will be in the future. Three companies are also taking part in the research, two Spanish ones – Global Energy Trading and Gecival – and one from Peru – Green Energy.
An initial version has been developed for the city of Valencia and researchers are working on another for Lima (Peru), but it can be applied to any location in the world.
The tool, named Movindeci, makes it possible to analyse the general state of transport and mobility in the city to be able to make strategic decisions in these areas.
“Our objective is to provide an intelligent system that facilitates the planning of the locations for these recharging devices in the short, medium and long term. It is a tool of great utility in order to, depending on the evolution of transport, the number of electric vehicles and population and mobility in the city, place the recharging spots in optimal locations, add new stations or remove those which are inefficient,” explains Vicente Julián, researcher of the Artificial Computing-Intelligence Group of the UPV.
Forecast for 2030
There are currently 1,700 public recharging points in Spain for 35,000 electric vehicles, and it is believed that a network of around 80,000 will be needed by 2030 in order to recharge the four million expected vehicles.
On the other hand, the Valencia region’s Electric Vehicle and Recharge Infrastructure Deployment Plan forecasts that in 2030, as many as 25% of new vehicles will be electric, making the installation of around 2,500 public recharge points necessary.
The tool developed by the UPV will help plan the location of these points both in Spain as well as others that will be installed in any other city worldwide.
Furthermore, Movindeci will also help electricity companies with the network planning and development, as well as guaranteeing the energetic supply to electronic vehicle battery charge infrastructures in cities, as well as to optimise investment in infrastructures and recharging points as well as to contribute to the efficiency of the electric system, “which leads to greater sustainability,” adds Javier Palanca, fellow researcher of the UPV’s Artificial Computing-Intelligence Group.
How does it perform the localisation?
The tool includes an artificial intelligence algorithm which automatically evaluates the possible locations for recharging stations and determines which are the most recommendable based on a group of factors that can be specified by the user.
Among these parameters are the following: the density of the area, urban mobility, an estimate of the time that vehicles spend at any given place, the main financial activities in the area, or the cost from an electricity point of view that it would have to bring the required energy from a substation to the recharging point.
Alongside them, another parameter also comes into play: social media activity in each point in the city. “Said activity will give us an approximate idea of the ways in which people move around the city, making it possible to get the most out of the user’s digital footprint on the internet, using said information to better understand their behavioural patterns, from the viewpoint of transport and mobility,” explains Javier Palanca.
This project is co-financed by Global Energy Trading and the European Regional Development Fund (ERDF) as part of the Operative Program of Intelligent Growth 2014-2020, with the objective of boosting research, technological development and innovation. The results have been published in the Applied Sciences journal.