Dynamic driving cycle analyses using electric vehicle time-series data

Abstract

Dynamic analyses of time-series data collected from real-world driving-cycle field testing of electric vehicles is providing evidence that certain driving-cycle conditions can significantly impact vehicle performance. In addition, vehicle performance results derived from time-series data show relationships that help to characterize driving cycles. Such findings confirm the advantages of time-series data over statistical data, in allowing correlation of vehicle performance characteristics with driving cycles. The driving-cycle vehicle performance analyses were performed using time-series data collected at the Electric and Hybrid Vehicle (EHV) National Data Center (NDC). A total of 71 EHVs are registered in the NDC and over 4000 trips files have already been uploaded into the NDC database, as of May 1997. Numerous EHVs on multiple trips have been analyzed over the past two years. This paper presents the results of time-series data collected and analyzed for two specific vehicles of the overall program, to illustrate the value of time-series data. We examined specific parameters such as average vehicle speed, number of stops during a trip, average distance traveled between stops, vehicle acceleration, and average DC kWh consumed per kilometer. Correlation among various parameters is presented in relationship to three driving cycles (highway, suburban, and urban), along with suggested ranges of parametric values defining the regimes of the different cycles.