Mechanical fault detection method and device

The invention discloses a mechanical fault detection method and device, and the method comprises the steps: carrying out the feature extraction of a preset mechanical vibration signal through employing a preset wavelet packet transformation method, and obtaining an initial feature vector which compr...

Full description

Saved in:
Bibliographic Details
Main Authors GUO WEIYING, JIANG LIN, LIU ZHIHU, FANG XIAOZHAO, JIANG KAIHANG, TENG SHAOHUA
Format Patent
LanguageChinese
English
Published 24.11.2020
Subjects
CALCULATING
COMPUTING
COUNTING
HANDLING RECORD CARRIERS
MEASURING
PHYSICS
PRESENTATION OF DATA
RECOGNITION OF DATA
RECORD CARRIERS
TESTING
TESTING STATIC OR DYNAMIC BALANCE OF MACHINES ORSTRUCTURES
TESTING STRUCTURES OR APPARATUS NOT OTHERWISE PROVIDED FOR
Online AccessGet full text

Cover

More Information
Summary:The invention discloses a mechanical fault detection method and device, and the method comprises the steps: carrying out the feature extraction of a preset mechanical vibration signal through employing a preset wavelet packet transformation method, and obtaining an initial feature vector which comprises an energy distribution feature; performing dictionary learning on the initial feature vector through a preset K-SVD algorithm to obtain an over-complete dictionary; performing sparse solution on the over-complete dictionary according to a preset CoSaMp algorithm to obtain sparse representation,wherein the sparse representation comprises sparse parameters; and classifying the sparse representation through a preset SVM model to obtain a detection result. According to the invention, the technical problem of low fault detection accuracy due to the fact that effective and reliable fault features cannot be obtained under the condition that mechanical part parameters or models are unknown inthe prior art can be solve
Bibliography:Application Number: CN202010831504