Abstract: Aiming at the problems such as low evaluation efficiency caused by incompleteness,insufficient digitization,lack of timeliness and accuracy and single evaluation mode in performance evaluation of subway train,a reference architecture for subway train performance evaluation system based on digital twin was proposed.From the perspective of virtual-reality mapping,the system was constructed from three progressive levels that were model mapping,behavior mapping and state mapping,which comprehensively expressed the train performance.Based on the time-space comparison,the abnormality was precisely positioning from the train level to the component level.A cloud-edge collaborative model was designed to solve the problem of low efficiency of train performance evaluation.The system was implemented to verify the effectiveness of the proposed method and theory and to provide a reference for the application and promotion of digital twin in subway.

Key words: digital twin, subway train, performance evaluation, virtual-reality mapping, time-space comparison, cloud-edge collaboration

摘要： 针对当前地铁列车性能评估中存在的评估不全面、数字化程度不足,评估缺乏时效性和准确性,评估模式单一导致的评估效率不高等问题,提出基于数字孪生的地铁列车性能评估系统参考架构,并从虚实映射的角度,通过模型映射、行为映射和状态映射3个递进层次构建性能评估系统,全面表达列车性能。进一步,基于时空环比实现从列车级到部件级的异常精确定位,设计云边协同的评估模式解决评估效率不高的问题。最后,对系统进行实现,验证了所提方法和理论的有效性,以期为数字孪生在地铁中的应用与推广提供一定的借鉴。

关键词: 数字孪生, 地铁列车, 性能评估, 虚实映射, 时空环比, 云边协同