a new technique for bearing fault detection in the time-frequency domain

This paper presents a new fast kurtogram method in the time-frequency domain using novel types of statistical features instead of the kurtosis. for this study, the problem of four classes for bearing fault detection is investigated using various statistical features. this research is conducted in four stages. at first, the stability of each feature for each fault mode is investigated. then, resistance to load changes as well as failure growth is studied. in the end, the resolution and fault detection for each feature using comparison with a determined pattern and the coherence rate is calculated. from the above results, the best feature that is both resistant and repeatable to different variations, as well as having suitable accuracy of detection and resolution, is selected. it is found that kurtosis feature is not in a good condition in comparison with other statistical features such as harmmean and median. this approach increases the fault identification accuracy significantly.