Study On Energy Consumption Estimation Of Electric Vehicle Based On Travelling Observations

Modelling a high accurate energy consumption estimation model is the best way to alleviate the driver's mileage anxiety in a short period,which is an important foundation for the time-space distribution prediction of the electric vehicle's electricity consumption demand.It is a vital part for reasonable planning and layout of charging stations and other infrastructureBased on the large-scale long-term actual travel data in Aichi Prefecture,Japan,this paper has overcome the difficulties in the low frequency of data acquisition,incomplete information,potential heterogeneity,lack of temperature and other driving environment information.The present work takes trip as object of study,matching the elevation information,exploring the potential heterogeneity,matching the ambient temperature for each trip and fusing the information with different particle size,different dimensions and different types from multisource.Finally,taking into account the common characteristics and specialty of electric vehicles,the present study builds an energy consumption estimation model.As a result,the model proposed in the paper can effectively improve the estimation accuracy and reliability of energy consumption based on sparse observations.The results show that there are significant differences in travel behavior and activity characteristics between the corporate vehicle and the private vehicle,and the heterogeneity of energy consumption characteristics between different vehicles is significant.The ambient temperature plays an important role in the energy consumption of electric vehicles.The variance of driving habit leads to the difference of energy consumption characteristics,and the random coefficient setting of the relevant variables can effectively explain the heterogeneity of the energy consumption characteristics.The results of the estimation of various models show that the model proposed can significantly improve the estimation accuracy of energy consumption.The multi-level mixed effects regression model,which is good at heterogeneous effect modeling,has the best effect on the calibration of the model proposed in the research.The usage of air conditioner and heater has a considerable proportion of impact on the energy consumption of electric vehicles.And the energy consumed by the heating per minute is more than that of the air conditioning,which is distinct from the internal engine combustion vehicles.The effect of the ambient temperature on the travel energy consumption is the cubic polynomial of temperature.The marginal effect of different slopes on the energy consumption doesn't change as a straight line.Previous works which only take the work that vehicle tractive force overcomes the gravity as climbing power consumption can't really reflect the climbing energy consumption.From the interaction effect analysis,it is concluded that the ambient temperature has a "regulating effect" on the efficiency of air conditioning and heating.The effect of the interaction should not be neglected when modeling.An air conditioning and heating control strategy is proposed.When the ambient temperature is below 17.7 ?,air conditioning is prohibited.When the ambient temperature is higher than 17.7 ?,it is forbidden to open the electric heating.The control strategy can effectively reduce the energy consumption caused by the irrational use of air conditioner and heater,and save 10% of the energy consumption averagely.