Online estimation of driving range for battery electric vehicles based on SOC-segmented actual driving cycle

Abstract

The transport sector is one of the most polluting sectors globally, battery electric vehicles (BEVs) are deemed as one alternative to improve the environmental efficiency for the transport sector. However, limited driving range of BEVs leads to the range anxiety of drivers. Therefore, accurate estimation of the remaining driving range (RDR) is a vital issue for BEVs. In this paper, by using the real-world data collected from one BEV operating on actual roads in Beijing, China, an online estimation model for RDR is established based on the classification of typical urban driving cycle. The driving cycle data is divided into segments according to the state-of-charge (SOC) first, and then the actual driving cycle is classified into four categories by combining the principal component analysis (PCA) with fuzzy c-means (FCM) clustering. Model verification results confirm the accuracy of the model. Moreover, the influencing factors on energy consumption rate (ECR) of BEVs are analyzed, and an econometric model of ECR is established to explore the economical driving speed for the BEV. The results indicate that the overall average ECR of the BEV is 0.151 kW⋅h/km, and a 39% increase in ECR and a 30% decrease in driving range in winter compared to summer. The optimal driving speeds range from 48 to 62 km/h, and the economical driving speed is 58 km/h from an energy efficiency perspective. The findings from this study could inspire the automobile manufacturers and consumers to be informed more clearly about the actual utilization of their BEVs.