Processing method of equipment vibration signal

Abstract

The invention relates to a processing method of equipment vibration signal, including the following steps: step (1), the equipment vibration signal is measured; step (2), the equipment vibration signal is processed, concrete steps include: step (2.1), the equipment vibration signal is decomposed into more than four independent frequency bands by second generation wavelet packet technology; step (2.2), energy of each decomposed frequency band is respectively calculated; step (2.3), the frequency band with energy maximum is selected, and autoregressive spectral processing is carried out on the corresponding second generation wavelet decomposed signal thereof, so as to obtain autoregressive spectrogram; step (3), equipment failure hidden trouble is judged according to the autoregressive spectrogram. In the processing method of the invention, after variable working condition equipment vibration signal is processed, a waveform graph beneficial to judgement can be obtained, thus creating condition for accurately judging equipment failure hidden trouble.

Description

technical field [0001] The invention relates to a method for detecting hidden dangers of equipment failures, in particular to a method for detecting hidden dangers of equipment failures using equipment vibration signals, in particular to a method for processing the measured equipment vibration signals. Background technique [0002] In order to detect potential equipment failures in advance, people have invented a detection method for potential equipment failures that uses vibration signals of mechanical equipment as state parameters. [0003] At present, the methods of using equipment vibration signals to detect potential equipment failures are generally traditional signal processing methods based on stationary processes, such as signal filtering, time-domain averaging, and adaptive noise reduction methods. The adoption of these processing methods makes the early detection of faults There have been many advances in diagnosis. However, the above processing methods also have ...

AI Technical Summary

Problems solved by technology

However, the above processing methods also have defects: for example, the signal filtering method generally cannot obtain information outside the filter passband at the same time; the time domain averaging method requires the signal to be stable and strictly sampled periodically, which is limited by the non-stationary operating conditions in practical applications. Difficult to implement; the adaptive denoising method requires the signal and noise to be stable and independent of each other. In the field, the signal and noise are not only non-stationary, but also have different degrees of correlation. Unsatisfactory handling affects the accuracy of diagnosing potential equipment failures

Images