Limited driving range remains one of the barriers for widespread adoption of electricvehicles (EVs). To address the problem of range anxiety, this paper presents an energy consumptionprediction method for EVs, designed for energy-efficient routing. This data-driven methodologycombines real-world measured driving data with geographical and weather data to predict theconsumption over any given road in a road network. The driving data are linked to the roadnetwork using geographic information system software that allows to separate trips into segmentswith similar road characteristics. The energy consumption over road segments is estimated using amultiple linear regression (MLR) model that links the energy consumption with microscopic drivingparameters (such as speed and acceleration) and external parameters (such as temperature). A neuralnetwork (NN) is used to predict the unknown microscopic driving parameters over a segmentprior to departure, given the road segment characteristics and weather conditions. The completeproposed model predicts the energy consumption with a mean absolute error (MAE) of 12–14% ofthe average trip consumption, of which 7–9% is caused by the energy consumption estimation of theMLR model. This method allows for prediction of energy consumption over any route in the roadnetwork prior to departure, and enables cost-optimization algorithms to calculate energy efficientroutes. The data-driven approach has the advantage that the model can easily be updated over timewith changing conditions.
Original languageEnglish
Article number608
Number of pages18
Issue number5
StatePublished - 1 May 2017

ID: 31782700