基于偏度的滚动轴承声信号故障分析方法

The Fault Diagnosis Method of Rolling Bearings based on Sound Signal Skewness

为了快速直接地对轴承故障进行诊断,且不受载荷和转速的影响,采用偏度作为时域分析的参数。首先,将声信号取绝对值,去掉信号的对称性,偏度便具有了诊断能力。其次,根据正态分布的特点,得到无故障轴承声信号偏度值的参考范围。最后,通过对仿真信号的研究得到了偏度与信噪比的拟合关系,从而初步判断轴承的故障程度。实验结果表明,偏度能准确地诊断出轴承是否有故障,而且在判断轴承故障程度上有良好表现。

In order to directly and rapidly diagnose bearing’s faults with avoiding the disturbance of the load and speed of the bearings, the sound signal skewness is chosen as the parameter for time domain analysis. First of all, the sound signal is changed into its absolute value to remove its symmetry so that the skewness can be used for fault diagnosis. Then,according to the feature of normal distribution, the reference range of the skewness values of healthy bearings is obtained.Finally, through the analysis of the simulation signal, the fitting curve between the skewness and the signal- to-noise ratio (SNR) is obtained and used for preliminary judgment of the severity of the bearing’s faults. The experimental results show that the skewness can be used to accurately diagnose the bearing’s faults and judge the severity of the bearing’s faults.