基于时频消噪TFPF和时频分布MBD的轴承早期故障诊断

噪声是影响轴承、齿轮等机械设备早期微弱故障特征正确提取的主要因素,利用新颖的时频峰值滤波技术TFPT有力的噪声消减特性,将PTFT与改进的时频分布MBD相结合,提出了时频峰值滤波TFPT-时频分布MBD的故障识别新方法,即应用TFPF消减振动信号的随机噪声作为时频分析的前置处理,对消噪的故障信号作MBD时频分析来识别故障特征,给出了时频峰值滤波时频分布的故障诊断模型。诊断实例的分析结果表明了与传统的MBD的故障特征提取相比,提出的改进方法更易提取出强噪声背景下的轴承早期的微弱故障,具有明显的可诊断性和实用性。

Rolling Bearing Incipient Fault Diagnosis Based on TFPT and Modified B-distribution

Noise is the biggest obstacle that makes the incipient fault diagnosis results of gear and rolling element bearing uncorrected,an approach to the extraction of weak fault features from vibration noise based on a fresh TFPF（time frequency peak filter） and MBD（modified B distribution） has been proposed,firstly,the weak fault information features are picked up from the vibration noise using the de-noising characteristic of TFPT as the preprocessing of the MBD analysis,the de-noised vibration is analyzed by MBD to distinguish fault features,the diagnosis model is presented based ob TPFT and TFD.The simulation signal and diagnosing example analysis results show that the proposed method is more effective than the method of direct MBD analysis in extracting weak fault against the background of strong noise.