Impact period automatic identification and extraction method and system based on envelope analysis

Abstract

The invention provides an impact period automatic identification and extraction method and system based on envelope analysis. A plurality of possible impact frequencies in an original signal are obtained by performing time domain peak envelope, envelope spectrum solving, automatic noise level obtaining, calculation of an effective peak point of the envelope spectrum, optimal characteristic frequency searching, obvious impact condition judgment and other algorithm steps, wherein the plurality of possible impact frequencies are arranged in a descending order according to the importance degree, and generally the former two main components are taken. According to the method, no complex operation is carried out in the calculation process, and the method is easy to transplant on embedded equipment and realize edge calculation; the time domain peak value is retained for envelope modulation, and the prior parameters of a resonance band are not needed; the impact period is automatically extracted through an algorithm, manual intervention is not needed, and the obtained impact frequency is directly compared with an equipment fault frequency library, so that corresponding fault information can be obtained, and the result is stable and reliable; the frequency extraction is realized in the frequency domain, so that when a certain noise exists in the signal and when obvious impact componentscannot be judged in the time domain, the impact period can still be effectively obtained; and the multi-cycle components in the signals can be extracted, and the method can be effectively applied tomulti-fault-type diagnosis occasions.

Description

technical field [0001] The present invention relates to the technical field of abnormalities of rotating machinery, such as abnormality detection of rotors, and in particular to a method and system for automatic identification and extraction of impact periods based on envelope analysis. Background technique [0002] There are widely used key components such as main shafts, gearboxes, rolling bearings, sliding bearings, couplings, etc. in rotating mechanical equipment. They often work under harsh conditions such as high and low speeds and heavy loads, and are particularly prone to damage, resulting in mechanical equipment shutdowns. Accidents that stop production. For this reason, effective fault diagnosis methods for rotating machinery have always been the goal of people's research. When these parts have local failures (peeling, wear, fracture, etc.), due to the rotation of the body, the damage point will hit the surface of other parts in contact with it during operation, r...

AI Technical Summary

Problems solved by technology

This method needs to filter the energy wave according to the specified conditions, and requires prior knowledge to determine the filtering parameters; two adjacent peaks or troughs of the energy wave are used to calculate the period of the energy wave change, only when the signal does not exist or exists very weak It is only applicable when there is an interfering signal, and the requirements for filtering results are very high; and this method will fail when there is broadband noise in the signal

Images