动车组牵引系统故障导致的停车延迟或事故问题始终存在，针对中国2022年某型号动车组牵引系统故障数据，分析主要故障设备与故障频次分布，统计各类故障问题造成的影响时长，作为该型动车组牵引系统故障的时空特征；选择分布模型对比各设备故障影响时长的分布特征，并通过Kolmogorov-Smirnov检验对各分布拟合进行检验。结果表明：受电弓、牵引变流器与车顶高压电缆的故障影响时长最适于用逻辑斯蒂分布模型拟合，而对数正态分布最适于拟合牵引变压器，主断路器与牵引电机的故障影响时长。动车组牵引系统故障发生时判断影响时长，指明系统设备维修优化的重点方向，对提高动车组运行安全性具有重大意义。

According to the fault data of a certain type EMU traction system in China in 2022, the location and frequency distribution of key faulty equipment were analyzed, and the impact duration caused by various failure problems was counted. These two are combined as the spatio-temporal characteristics of EMU traction system faults. The distribution model was selected to compare the duration distribution characteristics of various faults, and the K-S test method was used to analyze the fitting effects. The results show that the fault influence time of pantograph, traction converter and roof high-voltage cable is the most suitable for the logistic distribution model fitting, while the lognormal distribution is most suitable for the traction transformer, main circuit breaker and traction motor. It is of great significance to predict the failure time of EMU traction system, point out the optimization direction of system equipment maintenance, and improve the efficiency of train operation and scheduling.