The invention discloses a train arrival forecasting method based on a distributed structure, and the method comprises the steps: employing a distributed organization structure, employing a specific method to independently divide a monitoring region for each station, and combining with a multi-train position relation recognition method. The method can effectively reduce the calculation amount of the ATS central server due to the continuous judgment of the position relation between all trains and stations, improve the accuracy of train arrival prediction, and especially guarantee the availability of functions in a train dense arrival scene and a complex station type scene.
本发明公开了一种基于分布式结构的列车进站预告方法,采用分布式组织结构,使用特定方法为每个站台单独划分出监控区域,结合多趟列车位置关系识别方法,可以有效减少ATS中央服务器因持续判断所有列车与站台位置关系而产生运算量的同时,提高列车进站预告的准确度,尤其保证了列车密集进站场景和复杂站型场景下功能的可用性。
Train arrival forecasting method based on distributed structure
基于分布式结构的列车进站预告方法
2021-04-02
Patent
Electronic Resource
Chinese
IPC: | B61L Leiten des Eisenbahnverkehrs , GUIDING RAILWAY TRAFFIC |
Forecasting Train Arrival Time Using Modular Artificial Neural Networks
British Library Conference Proceedings | 2002
|European Patent Office | 2020
|