Plug-in Electric Vehicle (PEV) users tend to plugin and leave their vehicles for an extended periods of time in PEV specified areas of public parking lots. The extended amount of idle time concerns other PEV users who need to charge their vehicles. Several well-known, supervised machine-learning methods were applied by using data from 27,481 charging sessions. The data was obtained from existing public charging stations within Nebraska, predicting idle time calculations to be able to help state leaders minimize prolonged charging sessions. XGboost outperforms the other methods in the results with Root Mean Square Error equivalent to 0.9552 and R2 equating to 40.80%. Additionally, this study considers the relative importance of the input variable. By using the proposed data-driven strategy to predict the idle time at public charging stations, PEV users PEV can decide to wait or consider a different charging station.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    PEVs Idle Time Prediction at Public Charging Stations Using Machine-Learning Methods


    Contributors:


    Publication date :

    2021-06-21


    Size :

    867539 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English