In view of the problem that the traditional driving cycles are difficult to characterize the characteristics of electric vehicle, a typical large and medium-sized city, Xi’an, is taken as the research object to study the construction method of urban cycle condition of electric vehicle. The research work is mainly embodied in test route construction, test data processing and driving cycle construction. During the development of the test route, taking the map of Xi’an as the template, the total urban road length and the lengths of various roads of Xi’an are calculated at first. Then, the traffic flow data of various types of roads in Xi’an are collected. The length of the test route is determined by the repeated sampling hypothesis test method, the lengths and proportions of various road types in the test route are obtained by the analytic hierarchy process, and the test route is constructed based on the overall comprehensive characteristics of urban roads in Xi’an. In the data processing process, the original data are denoised by using wavelet decomposition and reconstruction. Then, the characteristic parameters of kinematics fragments are determined based on principal component analysis. Based on fuzzy C means (FCM) clustering analysis, the test data are divided into steady flow condition, congestion condition and smooth condition. In the driving cycles synthesis process, a complete driving cycle is synthesized by selecting the nearest kinematic fragment from the cluster center according to the proportion of the 3 kinds of kinematic segment. Finally, the characteristic parameters of the cycling condition of Xi’an is compared with the typical driving cycles of China, NEDC, UDDS, ECE15, EUDC, FTP-15 and soon, and a comparative study is carried out. The test result shows that the proposed urban driving cycle of electric vehicles can truly reflect the overall characteristics of Xi’an traffic, and the constructed driving cycle have higher authenticity.

CICTP