基于DCAE-CNN的高铁轮轨关系自动化检测方法

Automatic detection method of high-speed rail wheel-rail relationship based on DCAE-CNN

在高噪声环境下,高铁轮轨关系分析准确率较低,难以达到用户要求,对此,基于DCAE-CNN提出了一种新的高铁轮轨关系自动化检测方法,同时检测轮轨高频振动数据、弹条共振数据、轮重满载率、直尖轨受力数据,来确定高铁轮轨关系,分析高铁轮轨之间的作用力。基于DACE-CNN结构建立神经网络,在高速条件下进行轨道数据采集、轮轨关系检测、轨道几何尺寸测量、轨道线路质量评估、受电弓数据集成、信号源处理等,实现高速列车轮轨关系的匹配。实验结果表明,基于DCAE-CNN的高铁轮轨关系自动化检测方法的数据集检测准确率超过98%,检测能力较强,保证高铁轮轨关系符合要求。

In the high noise environment,the accuracy of wheel rail relationship analysis of high-speed railway is low and it is difficult to meet the user’s requirements. Therefore,a new automatic detection method of wheel rail relationship of high-speed railway is proposed based on dcae-cnn. At the same time,the high-frequency vibration data of wheel rail,elastic strip resonance data,wheel load ratio and force data of straight switch rail are detected to determine the wheel rail relationship and analyze the force between wheel rail of high-speed railway. The neural network is established based on dace-cnn structure to carry out track data acquisition, wheel rail relationship detection, track geometric dimension measurement,track line quality evaluation,pantograph data integration and signal source processing under high-speed conditions,so as to realize the matching of wheel rail relationship of high-speed train.The experimental results show that the data set detection accuracy of the automatic detection method of high-speed rail wheel rail relationship based on dcae-cnn is more than 98%,and the detection ability is strong to ensure that the high-speed rail wheel rail relationship meets the requirements.