Rolling bearing fault identification and trend prediction method

Abstract

The invention discloses a rolling bearing fault identification and trend prediction method, belonging to the field of rotating machinery fault diagnosis. The method comprises a step of obtaining the vibration signals of a bearing in normal and fault conditions as original signals, a step of using EMD to decompose the original signals into a finite number of IMF components, a step of selecting typical IMF components according to correlation analysis and summing the components to obtain a recombination signal, a step of using an FK algorithm to process the recombination signal, automatically obtaining an optimal center frequency and bandwidth for envelope analysis, extracting a fault characteristic frequency, and thus realizing fault type identification, and a step of selecting the energy percentage of an IMF component with a largest correlation coefficient with the original signals as a fault degree evaluation index. The method is simple and effective, the SNR can be effectively improved, the accurate identification of a fault type is helped, the fault degree evaluation index selection is reasonable, the fault development trend can be effectively reflected, and the method has a large application value for bearing state monitoring and fault evaluation.

Description

technical field [0001] The invention belongs to the field of fault diagnosis of rotating machinery, in particular to a rolling bearing fault diagnosis method based on typical IMF component selection and fast kurtosis diagram. Background technique [0002] Rolling bearings are key components of rotating machinery and are also one of the most common sources of equipment failure. Their operating status directly affects the safety and reliability of the entire machine. Therefore, it is one of the most important topics in the field of mechanical fault diagnosis to grasp the working state of the bearing in real time and understand the formation and development of faults. At present, the most commonly used method for condition monitoring and diagnosis of rolling bearings is based on the analysis and processing of vibration signals. However, due to the weak early fault characteristics and low signal-to-noise ratio, useful characteristic signals are often submerged in background nois...

AI Technical Summary

Problems solved by technology

At present, the most commonly used method for rolling bearing condition monitoring and diagnosis is based on the analysis and processing of vibration signals, but because the early fault features are very weak and the signal-to-noise ratio is low, useful feature signals are often submerged in the background noise

Images