Answers are grounded in this paper and cite their sources.
Rotating machines are widely used in industry, and occurrence of failures may cause severe economic losses and social impacts. Hence, this paper conducts an investigation on rotating machine fault diagnosis, and a novel method based on the structure of fast kurtogram is proposed. On one hand, a novel index is constructed that makes use of both the second order cyclostationary theory and the threshold of statistical test based on hypothesis of coloured noise. In order to reduce impact of outliers on the proposed method, information of fault characteristic frequency of rotating machine to be diagnosed is involved in the definition of index. On the other hand, the Meyer wavelet filters of empirical wavelet transform are employed to replace the filters of fast kurtogram, mainly considering the Meyer basis' properties of good time and frequency localization, good spectrum partitioning capability, and similarity with gear and bearing vibration signal. In the case study, an acoustic emission signal of planetary gearbox and a vibration signal of rolling bearing are used to validate the proposed method compared with the fast kurtogram and Gini-based method.
No takes yet. Share an insight, caveat, or question.
Luo et al. (2019) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: