< Back to previous page
Project
A data driven approach for the planning of electric vehicle charging infrastructure in urban area (FWOTM1091)
Electric Vehicles (EVs) have a great potential to improve air quality
and reduce noise pollution generated by traffic in cities. An important
obstacle for the large-scale adoption of EVs is the limited driving
range perceived by potential EV drivers. The development of public
electric vehicle charging infrastructure with sufficient coverage is
crucial to mitigate this anxiety. This is especially the case in urban
areas, where a large share of residents is dependent on public
charging infrastructure for home charging and the available space is
limited. To prevent wasting scarce resources from the general public,
it is necessary that each charging station is being placed on its best
suited location at the right time.
This PhD research therefore aims to improve the localisation of
public charging infrastructure in both space and time. This is done by
proposing a data-driven two-step methodology. The first step
consists of predicting charging demand based on user-, destinationand route- oriented data. This novel approach combines supervised
and unsupervised techniques to differentiate between opportunityand residential-oriented charging. The second step consists of
building a location model that strategically locates both fast and slow
charging stations according to the predicted charging demand. This
in turn allows to develop a charging infrastructure that meets the EV
users’ preferences while relying on data-driven decision making.
and reduce noise pollution generated by traffic in cities. An important
obstacle for the large-scale adoption of EVs is the limited driving
range perceived by potential EV drivers. The development of public
electric vehicle charging infrastructure with sufficient coverage is
crucial to mitigate this anxiety. This is especially the case in urban
areas, where a large share of residents is dependent on public
charging infrastructure for home charging and the available space is
limited. To prevent wasting scarce resources from the general public,
it is necessary that each charging station is being placed on its best
suited location at the right time.
This PhD research therefore aims to improve the localisation of
public charging infrastructure in both space and time. This is done by
proposing a data-driven two-step methodology. The first step
consists of predicting charging demand based on user-, destinationand route- oriented data. This novel approach combines supervised
and unsupervised techniques to differentiate between opportunityand residential-oriented charging. The second step consists of
building a location model that strategically locates both fast and slow
charging stations according to the predicted charging demand. This
in turn allows to develop a charging infrastructure that meets the EV
users’ preferences while relying on data-driven decision making.
Date:1 Nov 2021 → Today
Keywords:Urban charging stations, Prediction methods, Spatial localization models
Disciplines:Transport economics, Mathematical methods, programming models, mathematical and simulation modelling, Geography of mobility and transportation