Optimization method for spectrum harmonic averaging of rotating machinery

Abstract

An optimization method for spectrum harmonic averaging of rotating machinery comprises the following steps: first performing Fourier transform on an original vibration signal to obtain the spectrum ofthe vibration signal; then determining the spectrum calculation range according to the analysis needs, and performing frequency doubling averaging calculation on all the analyzing spectrum lines within the determined spectrum calculation range; and finally removing the spectral local minimum value mean and pseudo peak generated by the frequency doubling averaging to achieve the removal of the non-harmonic component interference noise. The method provided by the invention utilizes the feature that the characteristic component has the harmonic characteristic when a fault occurs, and uses a spectral harmonic averaging method to eliminate or weaken the non-harmonic low-frequency and co-channel interference in the spectrum and envelope spectrum to improve the signal-to-noise ratio, thereby realizing early identification and diagnosis of faults.

Description

technical field [0001] The invention belongs to the technical field of status monitoring and fault diagnosis of rotating machinery equipment, and in particular relates to a method for optimizing frequency spectrum harmonics of rotating machinery. Background technique [0002] At present, in my country's aviation, petroleum, chemical, metallurgical, electric power and other enterprises, rotating machinery and equipment account for about 80%. In order to improve work efficiency, many manufacturing processes are flow-line operations. In actual work, once a problem occurs in a certain link, the entire production process will be paralyzed, resulting in huge economic losses. For some equipment working at high speed, it may also be Serious accidents causing casualties. Therefore, it is of great significance to carry out fault monitoring and diagnosis on the above rotating mechanical equipment. [0003] The common faults of rotating machinery mainly include rotor faults, bearing f...

AI Technical Summary

Problems solved by technology

Spectrum analysis is an effective method for fault identification. The vibration acceleration signals of many faults usually contain fault characteristic information, which is represented as a specific fault frequency component on the spectrum. However, in the early stage of the fault, the energy of the characteristic information is small and affected by Path, sensor installation error, natural vibration of equipment components, etc., the signal-to-noise ratio of fault information is low, and it is difficult to directly use spectrum analysis for feature extraction

Taking rolling bearings as an example, at present, the method of demodulation spectrum analysis is usually used for early fault diagnosis. However, for faults such as inner rings, balls, and cages, because the fault location is far away from the sensor installation location, the transmission path is complicated, and the fault information has been greatly attenuated. Afterwards, it is easy to be submerged in low-frequency noise interference. When the fault develops in the middle stage, although the fault characteristics can be observed in the low-frequency area of ​​the spectrum, it is easily affected by power frequency and other co-frequency interference, which brings difficulties to the extraction of diagnostic information.

Images