Research On Several Methods Of Rolling Bearing Fault Diagnosis

As the safety and stabilization of machinery is of great significance in improving economic efficiency and reducing casualties,it is necessary to develop condition monitoring and fault diagnosis technology for machinery.Focused on fault diagnosis of rolling bearing,some researches were carried out based on wayside acoustic signal and vibration signal to correct the Doppler distortion and enhance fault feature.Firstly,the vibration mechanism,the main failure modes and the calculation method of the characteristic frequency of the rolling bearing are briefly introduced.The experimental device and the experimental process are described.Then,two Doppler correction algorithms based on microphone array are applied to the wayside acoustic diagnosis of train bearings as there are some problems in the single microphone ones.Making full use of information provided by microphone array to acquire parameters rather than pre-measuring or searching,these two array methods are more efficient in computation and robust to background noise.Finally,an improved morphological filter is proposed to improve the fault feature.In order to enhance the pulse characteristic in the original signal,the plus operation is replaced by the inner product operation in the expansion operation.The flat structure operator is then applied for noise reduction in the corrosion operation.Compared with some of the traditional methods,this one shows a stronger fault feature extraction capability.