基于模糊聚类的西安市区轿车行驶工况构建

Construction of Sedan＇s Driving Cycle in Xi＇an City Based on Fuzzy Clustering

在行驶工况构建过程中，实验数据的解析与处理方法直接影响构建的代表性行驶工况的精度。由于城市道路交通结构复杂，车辆运行方式特殊，因此，引入模糊聚类方法对西安市区道路车辆运动状态数据进行处理，以建立能够反映交通流状况且与车辆行驶状态紧密相关的行驶工况。通过对平均速度和行驶速度标准偏差模糊聚类，将西安市区道路轿车行驶状态分为拥堵状态、稳定流动状态和畅通状态三类，在此基础上采用分类法和短行程法相结合构建了西安市区轿车代表性行驶工况。验证发现：利用模糊聚类方法构建的西安工况与实验数据相对误差较小，能够反映西安市区轿车真实的运行状态。

In the process of driving cycle construction, the analysis and processing in experimental data will di- rectly affect the accuracy of driving cycle. Due to the complexity of urban road traffic structure and vehicle＇s special operating mode, fuzzy clustering method is introduced to analyze vehicle＇s typical motion data of Xi＇an urban road traffic in order to establish driving cycle closely related to vehicle＇s running state and road traffic flow. After average speed and speed standard deviation fuzzy clustering,the results are as following： sedan＇s driving states on Xi ＇an ur- ban road are divided into three categories congestion running state, steady flow running state and free flow run- ning state. Furthermore, combining the classification method with short - stroke method, the sedan＇ s representative driving cycle in Xi ＇an city is built and verified. The result show that the relative error between Xi ＇an representative driving cycle constructed by fuzzy clustering method and experimental data is a bit of small, which would reflect the vehicle＇s actual running state.