Feedforward control stochastic resonance-based bearing early-stage fault diagnosis method

Abstract

The invention discloses a feedforward control stochastic resonance-based bearing early-stage fault diagnosis method. The method includes the following steps that: (1) a sensor collects original bearing signals; (2) the original bearing signals are processed, so that square wave-like signals can be obtained and are adopted as feedforward control signals; (3) the original bearing signals and the feedforward control signals are jointly subjected to scale change so as to be converted into small-frequency signals with a feedforward control function; (4) two bistable systems are adopted to construct a coupled bistable system through a nonlinear coupling mode; (5) the small-frequency signals obtained in the step 3 are adopted as input signals, and a genetic algorithm is utilized to obtain the optimal parameters of the coupled bistable system; and (6) the small-frequency signals obtained in the step 3 are inputted into the optimal coupled bistable system, and diagnosis on a bearing early-stage fault can be realized through scale restoration. With the feedforward control stochastic resonance-based bearing early-stage fault diagnosis method of the invention adopted, the optimal system parameters of the coupled bistable system are searched through the genetic algorithm, and the feedforward control of stochastic resonance is combined with an improved potential function, so that an enhanced stochastic resonance effect can be generated.

Description

technical field [0001] The invention belongs to signal processing and detection technology, and particularly relates to a bearing early fault diagnosis method based on feedforward control stochastic resonance. Background technique [0002] People usually think that noise is a kind of harmful interference. Indeed, in the detection of useful signals, noise will affect many detection systems, resulting in abnormal detection. Therefore, people are looking for ways to eliminate noise. However, the emergence of stochastic resonance makes it surprising that noise can also be turned into a beneficial signal. Stochastic resonance is a phenomenon that reflects the positive effect of noise in nonlinear systems, that is, under certain nonlinear conditions, the nonlinear system enhances the periodic signal output caused by the synergy of weak periodic signals and noise (random interference). The phenomenon. However, the generation and enhancement of stochastic resonance requires condi...

AI Technical Summary

Problems solved by technology

In the process of real-time detection, there must be a contradiction between detection accuracy and detection speed.

Images