Projects / Programmes
Blind source separation and location of acoustic emission sources on constructions with rotary machines
Code |
Science |
Field |
Subfield |
2.05.00 |
Engineering sciences and technologies |
Mechanics |
|
Code |
Science |
Field |
T121 |
Technological sciences |
Signal processing |
acoustic emission, blind source separation, non-destructive testing , waveguides, adaptive systems, rotary machines
Researchers (1)
no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
1. |
15696 |
PhD Tadej Kosel |
Engineering sciences and technologies |
Head |
2002 - 2004 |
441 |
Organisations (1)
Abstract
Research field of this project is acoustic emission analysis and empirical modeling of sound. Signals of acoustic emission which are generated and propagated on constructions with mounted rotary machines can be composed of different independent components. The main problem in such case is how to find out a number of independent acoustic emission sources and how to locate and characterize such independent sources.
Acoustic emission analysis is one non-destructive testing method and is used for location and characterization of sound sources. Acoustic emission signals are usualy composed of more independent source signals which are generated in different locations on the construction and are result of different source mechanisms ie. crack growth, fatigue, friction etc. One of the most difficult problems of acoustic emission analysis is the separation, location and characterization of the independent acoustic emission sources without knowing source signals and the mixing function. The method for solving this problem in acoustics is known as blind source separation.
In our research we will mainly focuse to the constructions which have mounted rotary machines. The examples of such constructions are compresors of refrigirators and machine tools for grinding. On such constructions more instantaneous continuous and discrete acoustic emission sources can be generated. For analysis of acoustic emission sources we will use blind source separation algorithms and algorithms fo acoustic emission analysis and we will try to separate independent sources. The result of this project will be the measuring system, which will be mounted on the construction and will monitor dynamic of the acoustic emission sources. Sources can be generated by parts of the rotary machines, crack growth, fatigu etc. This data can be used for monitoring of life period of the constructions, detection of the defects on the constructions and rotary machines and on-line machine monitoring and quality control.