基于小波分析的高速列车车体运行状态估计

基于监测数据评估高速列车空气弹簧和横向减振器等关键部件的运行状态, 针对车体垂向加速度振动信号, 提出了小波包能量矩的列车状态估计方法。首先分析车体垂向振动特征, 对不同工况和不同速度下的信号进行小波包分解, 并重构能量较大的频带信号, 再计算各频带的小波包能量矩特征, 不同频带信号的小波包能量矩变化反映了列车运行状态的改变。将不同频带的小波包能量矩组成特征向量, 最后用支持向量机进行故障识别。实验数据仿真分析表明, 列车空簧失气故障和横向减振器失效故障识别率为100%, 说明该方法能很好地估计出高速列车的故障状态。

Running state estimation of high-speed train based on wavelet analysis

Based on monitoring data, this paper estimated the running state of key components about air springs and lateral damper of high-speed train. Aiming at vertical acceleration for high-speed train vibration signal, this paper proposed a state estimation method of wavelet packet energy moment. Firstly, it analyzed the feature of vertical vibration, and collected the vibration signals under different running conditions and different speeds. It transformed the experiment data by wavelet packet, and selected the larger energy frequency bands signal to reconstruct. Then, it calculated the value of wavelet packet energy moment. The energy moment of different frequency bands reflected different high-speed train's status. Finally, it extracted the wavelet packet energy moment of different bands to make feature vectors, and used SVM method to identify faults. Experimental results show that the recognition rate of air spring lose fault and lateral damper fault are 100%. So it is proved that the method is effective for estimating accurately different faults of high-speed train.