Skip to Main content Skip to Navigation

L'imagerie acoustique au service de la surveillance et de la détection des défauts mécaniques

Abstract : Vibration analysis is mainly used in condition monitoring and fault detection of rotating machine domain. The success of the diagnosis is strongly related to the position of the accelerometers. However, the machine geometry sometimes prevents the sensors to be placed close enough to the faulted part causing the diagnostic failure. The sound emitted by a mechanism and its condition are related. Using microphones to optimize condition monitoring is then justified. Acoustic imaging techniques (acoustic holography, beamforming, etc…) are mainly used as a source localization and quantification tool but they can be turned into a powerful diagnosis tool. Several strategies based on the beamforming algorithm are developed in this work. Firstly, diagnosis features commonly used in condition monitoring of rotating machinery are mapped as a function of space. Kurtosis allows localizing impulsive sources which eventually can be related to a mechanism failure. New features based on the squared envelope spectrum of the focused signals are also introduced. They aim toward the detection of inner and outer race fault in roller element bearings. On the other hand, angular synchronous average is used to extract the acoustic field synchronous with one component rotation. The sources related to a fault are localized in the residual field mappings. Finally, a new imaging technique based on the vibroacoustic transfer functions between a few accelerometers placed on the machine and the microphone array is developed. It allows obtaining the mappings of the radiated pressure on the machine surface only thanks to the accelerometers. It is tested as a fault detection tool on a test bench
Complete list of metadatas

Cited literature [100 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Monday, March 11, 2019 - 10:23:09 AM
Last modification on : Wednesday, July 8, 2020 - 12:42:27 PM
Long-term archiving on: : Wednesday, June 12, 2019 - 1:13:01 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02063337, version 1


Edouard Cardenas Cabada. L'imagerie acoustique au service de la surveillance et de la détection des défauts mécaniques. Acoustique [physics.class-ph]. Université de Lyon, 2017. Français. ⟨NNT : 2017LYSEI124⟩. ⟨tel-02063337⟩



Record views


Files downloads