Rolling bearing fault feature extraction method based on improved quantum evolution algorithm

Abstract

The invention discloses a rolling bearing fault feature extraction method based on an improved quantum evolution algorithm, comprising steps of S1, acquiring the vibration signal f of a rolling bearing; S2, establishing a Gabor atom library; S3, establishing a quantum population [psi]; S4, performing quantum encoding on Gabor atoms by using a quantum probability amplitude; S5, subjecting the rolling bearing vibration signal f to a sparse decomposition on the Gabor atom library after the quantum probability amplitude encoding to select the optimal encoded atom g[gamma]*(t), the optimal population individual [psi]* and the optimal quantum phase [theta]*; S6, evolving the quantum population; S7, mutating the quantum population; S8, calculating the kurtosis value of a sparse reconstruction signal; and S9, repeating the step S4 to the step S8 until the maximum kurtosis value of the sparse reconstruction signal, wherein the sparse reconstruction signal at the maximum kurtosis value is the extracted rolling bearing fault feature component. The fault feature component extracted by the method has obvious periodicity, and the noise contained therein is obviously reduced. The method has certain advantages in terms of rapidity and adaptability.

Description

technical field [0001] The invention belongs to the technical field of signal analysis and fault diagnosis, and in particular relates to a rolling bearing fault feature extraction method based on an improved quantum evolutionary algorithm. Background technique [0002] Rolling bearings are an important part of rotating machinery, and their fault monitoring is of great significance to ensure the safe operation of mechanical equipment. When a rolling bearing fails, its vibration signal is a complex nonlinear and non-stationary signal. How to accurately and quickly extract signal components that reflect fault characteristics from complex vibration signals is the key to fault diagnosis of rolling bearings. [0003] In recent years, artificial intelligence algorithms have been rapidly applied to the feature extraction of rolling bearing faults, such as particle swarm optimization, convolutional neural network, and shape optimization filtering. As an artificial intelligence algo...

AI Technical Summary

Problems solved by technology

However, since the existing quantum evolutionary algorithms based on probability amplitude mostly use fixed rotation phase angles or fitness gradients to update qubit phase angles, and use NOT gates or other gate operations as mutation strategies, this makes quantum evolutionary algorithms accurate and effective in searching There is a lack of adaptability in extracting the characteristic components of rolling bearing faults

Images