Kurtosis-index-based rotating machine fault diagnosis method

Abstract

The invention provides a kurtosis-index-based rotating machine fault diagnosis method. A kurtosis index is constructed based on a signal separation idea, so that defects of the existing kurtosis index diagnosis technology can be overcome and the diagnosis precision and reliability are improved. According to a standard vibration signal, real-time collected vibration signals are classified into a non-fault vibration signal and a mixed signal; and a kurtosis index is constructed by using the mixed signal and the non-fault vibration signal, so that the fault diagnosis becomes sensitive and a weak fault feature signal can be detected by the signal change. When the rotating machine equipment works normally or a fault occurs at the equipment, the overlapped part of valuing ranges of the kurtosis index is small and the kurtosis indexes in different states change obviously; the sensitivity to the eccentric shaft type fault is high; and distinguishing of the fault and other faults can be carried out well. And thus the precision and reliability of the rotating machine fault diagnosis are high.

Description

technical field [0001] The invention relates to a method for diagnosing a fault of a rotating machine, in particular to a method for diagnosing a fault of a rotating machine based on a kurtosis index, and belongs to the technical field of fault diagnosis and signal processing and analysis. Background technique [0002] Large-scale rotating machinery equipment (such as steam turbines, rotating bearings, compressors, etc.) is the key equipment in important engineering fields such as petroleum, chemical industry, machinery manufacturing, and aerospace. Rotating machinery equipment is constantly developing in the direction of large-scale, automation, and precision , its composition and structure are becoming more and more complex, and the probability of failure is also increasing, so the requirements for the safety and reliability of large rotating machinery equipment are also getting higher and higher. [0003] However, when large-scale rotating machinery fails, vibration monit...

AI Technical Summary

Problems solved by technology

[0007] However, the prior art using kurtosis indicators to diagnose the faults of rotating machinery mainly has the following problems: First, the idea of ​​signal separation is not used to construct kurtosis indicators, and the standard vibration signal is not separated, which cannot better reflect the characteristics of rotating machinery. The fault is not sensitive to the diagnosis of the fault; the second is that when the kurtosis index is used for analysis, the kurtosis index is unstable for fault judgment. , the kurtosis index drops rapidly, so the kurtosis index is not applicable to the late fault of the bearing, and the reliability of the fault diagnosis of the rotating machinery is poor; the third is to use the vibration signal of the running-in period as the standard vibration signal, and does not consider the initial operation of the rotating machinery. The noise in the vibration signal is serious, it is not suitable as a standard vibration signal, and the fault diagnosis is not accurate

Images