eduzhai > Applied Sciences > Computer Science >

Predicting popularity of EV charging infrastructure from GIS data

  • KanKan
  • (0) Download
  • 20210425
  • Save

... pages left unread,continue reading

Document pages: 47 pages

Abstract: The availability of charging infrastructure is essential for large-scaleadoption of electric vehicles (EV). Charging patterns and the utilization ofinfrastructure have consequences not only for the energy demand, loading localpower grids but influence the economic returns, parking policies and furtheradoption of EVs. We develop a data-driven approach that is exploitingpredictors compiled from GIS data describing the urban context and urbanactivities near charging infrastructure to explore correlations with acomprehensive set of indicators measuring the performance of charginginfrastructure. The best fit was identified for the size of the unique group ofvisitors (popularity) attracted by the charging infrastructure. Consecutively,charging infrastructure is ranked by popularity. The question of whether or nota given charging spot belongs to the top tier is posed as a binaryclassification problem and predictive performance of logistic regressionregularized with an l-1 penalty, random forests and gradient boosted regressiontrees is evaluated. Obtained results indicate that the collected predictorscontain information that can be used to predict the popularity of charginginfrastructure. The significance of predictors and how they are linked with thepopularity are explored as well. The proposed methodology can be used to informcharging infrastructure deployment strategies.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...