An accurate prediction of the electric vehicles (EVs) energy consumption is the crucial requirement to deliver the promise of the green energy solution for relieving the concerns from fossil energy depletion and vehicle emissions. To solve the problem, the most substantial facing challenges are laying down on heterogeneous data insight, modelling the non-linear problem, and the lack of supporting technologies to provide the primary data to model the problem. In this paper, the latent pattern of the heterogeneous data is extracted and presented as a new set of metadata. Statistical features extracted from crowdsourced EVs profiles for the energy prediction since they consider the varying impact factors of the individual impact factors. Supervised learning approach is followed to address the non-linear problem and unseen correlation in the heterogeneous data. The results show an improvement of the energy prediction if the statistical features are considered.
|Titolo:||Energy Consumption Prediction of Electric Vehicles Based on Big Data Approach|
|Data di pubblicazione:||2018|
|Appare nelle tipologie:||04.1 Contributo in Atti di convegno|
File in questo prodotto:
|08494573.pdf||PDF editoriale||Accesso riservato|