Snake state recognition method based on wheel set transverse movement amount

The invention, which belongs to the technical field of train operation monitoring, discloses a snaking state identification method based on the transverse movement amount of a wheel set, comprising the following steps: S1, constructing a vehicle operation data set; s2, wheel set transverse movement signals in the vehicle operation data set are analyzed and checked, and a snaking state classification result is obtained; s3, training a snaking state recognition model based on a corresponding relation among a sensor measuring point signal, a wheel set transverse movement signal and a snaking state classification result in the vehicle operation data set; s4, in the train running process, actual sensor measuring point signals are input into the snaking state recognition model, and a snaking state recognition result is obtained. 本发明公开了一种基于轮对横移量的蛇行状态识别方法，属于列车运行监测技术领域，包括：S1、构建车辆运行数据集；S2、对车辆运行数据集中的轮对横移信号进行分析及检验，得到蛇行状态分类结果；S3、基于车辆运行数据集中的传感器测点信号、轮对横移信号以及蛇行状态分类结果之间的对应关系，训练蛇行状态识别模型；S4、在列车运行过程中，将实际的传感器测点信号输入至蛇行状态识别模型中，得到蛇行