Stochastic resonance weak impact feature enhancement extraction method on the basis of sliding window

Abstract

The invention discloses a stochastic resonance weak impact feature enhancement extraction method on the basis of a sliding window. The method comprises the following steps: firstly, according to the rotation frequency of a tested object, constructing the sliding window in a time domain, and utilizing the sliding window to carry out time domain interception on a signal; secondly, successively taking each intercepted sub-signal as the input of a variable-scale random resonance system, selecting an optimal system parameter according to the maximization of a resonance response measurement index, and substituting the optimal system parameter into the resonance system to obtain an optimal detection result of the corresponding sub-signal; and finally, carrying out normalization processing on the optimal detection results of all sub-signals according to a uniform standard to eliminate amplitude differences, and successively carrying out end-to-end connection to obtain a final detection result. The method utilizes the sliding window to partition the signal which contains a plurality of impact components into a plurality of sub-signals which contain a single impact component, solves the problem that weak impact features are difficult in effectively extracting, realizes the enhancement extraction of the weak impact features of different amplitudes, and exhibits an important meaning on the early weak fault diagnosis of mechanical equipment.

Description

technical field [0001] The invention belongs to the technical field of signal processing, and relates to a random resonance weak shock feature enhancement extraction method based on a sliding window, which can realize effective extraction of different amplitude shock features in a signal. Background technique [0002] With the development of modern production and the advancement of science and technology, the structure of mechanical equipment is becoming more and more complex, the functions are becoming more and more perfect, and the degree of automation is getting higher and higher, which also makes the failure rate of mechanical equipment increase, and it is difficult to Various damages will inevitably occur, resulting in reduction or loss of intended functions, and even serious and even catastrophic accidents. Therefore, the monitoring and fault diagnosis of mechanical equipment operation status, especially for the monitoring and diagnosis of early weak faults, has import...

AI Technical Summary

Problems solved by technology

Aiming at the problem of shock feature extraction under strong background noise, the method uses sliding windows to realize data segmentation of multi-shock component signals, and uses the correlation margin index to realize the optimal selection of stochastic resonance system parameters, and then realizes the effective extraction of weak shock features

Images