机械

车轮多边形普遍存在于铁路车辆上，尤其是在高速情况下会对车辆和轨道产生强烈的周期性激励，影响车辆运行安全和乘客舒适性，因此研究车轮多边形的检测方法具有重要意义。采用轴箱垂向振动加速度信号作为检测数据，提出一种基于角域同步平均的高速列车车轮多边形检测方法。首先，结合一致相关系数，从滤波后的原始信号中提取相对稳定的短时时域信号。其次，将时域信号重采样到角域，再利用角域同步平均对角域信号进行去噪，得到特征向量。最后，根据特征向量计算反映车轮多边形状态的粗糙度水平和多边形阶次这两个参数，完成车轮多边形的准确估计。仿真分析和实例验证表明：该方法可以有效增强同步分量，去除速度因素以及轨道不平顺等偶然干扰因素的影响，完成高速列车车轮多边形的检测。

Wheel polygons generally exist on railway vehicles, especially in the case of high speed. It will produce strong periodic excitation to vehicles and tracks, affecting the safety of vehicle operation and passenger comfort. Therefore, it is of great significance to study the detection method of wheel polygons. Using the axle box vertical vibration acceleration signal as the detection data, a wheel polygon detection method of high-speed train based on angle domain synchronous averaging is proposed. First of all, combined with the concordance correlation coefficient, the relatively stable short-time domain signal is extracted from the filtered original signal.Secondly, the time domain signal is resampled to the angle-domain, and then the angle-domain synchronous averaging is used to de-noise the signal,so that the eigenvector is obtained.Finally,the roughness level and polygon order,which reflect the wheel polygon state,are calculated according to the eigenvector to complete the accurate estimation of the wheel polygon.Simulation analysis and case verification show that this method can effectively enhance the synchronous component,remove the influence of accidental interference factors such as speed factors and track irregularity,and complete the detection of wheel polygons of high-speed trains.