Enhancement of energy consumption estimation for electric vehicles by using machine learning

Adnane Cabani, Peiwen Zhang, Redouane Khemmar, Jin Xu

Abstract


Three main classes are considered of significant influence factors when predicting the energy consumption rate of electric vehicles (EV): environment, driver behaviour, and vehicle. These classes take into account constant or variable parameters which influences the energy consumption of the EV. In this paper, we develop a new model taking into account the three classes as well as the interaction between them in order to improve the quality of EV energy consumption. The model depends on a new approach based on machine learning and especially k-NN algorithm in order to estimate the EV energy consumption. Following a lazy learning paradigm, this approach allows better estimation performance. The advantage of our proposal, in regards to mathematical approach, is taking into account the real situation of the ecosystem on the basis of historical data. In fact, the behavior of the driver (driving style, heating usage, air conditioner usage, battery state, etc.) impacts directly the EV energy consumption. The obtained results show that we can reach up to 96.5% of accuracy about the estimated of energy-consumption. The proposed method is used in order to find the optimal path between two points (departure-destination) in terms of energy consumption.


Keywords


Electric vehicle; Energy consumption; Intelligent transportation systems; K-NN, Machine learning

Full Text:

PDF


DOI: http://doi.org/10.11591/ijai.v10.i1.pp215-223

Refbacks

  • There are currently no refbacks.


View IJAI Stats

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.