优化VMD在高速列车牵引电机轴承故障诊断中的应用

Application of optimized VMD in fault diagnosis of motor bearings of high-speed train

针对强噪声背景下高速列车电机轴承早期故障信息提取困难的问题,提出了优化VMD算法。该算法依据频谱趋势估计算法获取振动信号频谱趋势,从而确定共振频带和分解层数;根据各频带边界和带宽,依据取值公式获取惩罚因子初始矩阵,最后将已知参数代入到VMD算法中实现对信号的分解。通过仿真信号和试验信号验证表明,该方法能在低信噪比条件下较为准确地识别到共振频带,并对信号进行准确的分解,提高了VMD算法的准确性和自适应性。

Aiming at the difficulty in extracting early faults of high-speed train motor bearings under the background of strong noise, an optimized VMD algorithm was proposed. Firstly, the vibration signal spectrum trend was obtained according to the spectrum trend estimation algorithm, thereby the resonance frequency band and the number of decomposition layers were determined;then, according to the boundary and bandwidth of each frequency band, the initial matrix of the penalty factor was obtained according to the value formula;finally the known parameters were substituted into the VMD algorithm to realize the decomposition of the signals. Verification by simulation signals and experimental signals shows that the method can more accurately identify the resonance frequency band under the condition of low signal-to-noise ratio, and accurately decompose the signal, which improves the accuracy and adaptability of the VMD algorithm.