Mother wavelet selection using signal energy for cracks detection in the rotation shafts.

Autor: Pricop, C., Pricop, M., Novac, G.
Předmět:
Zdroj: Scientific Bulletin 'Mircea cel Batran' Naval Academy; 2020, Vol. 23 Issue 2, Preceding p64-74, 12p
Abstrakt: The mother wavelet greatly influences the wavelet analysis of a non-stationary and nonlinear recorded signal. Choosing mother wavelet must be done to determine cracks in rotating shafts so as to take into account the nature and type of information signals to be extracted from the signal. The difficulty in optimum selection of the mother wavelet is determined by their complex properties that determine different selection criteria. In the paper, several families of functions (Haar, Daubechies, Symlets, Coiflet, BiorSplines) were used for analysis and the proposed selection criterion is the energy dissipated on the frequency bands. Signal recordings were made on a stand to determine the presence of cracks in rotating shafts and their classification. For discrete decomposition of recorded signals (DWT) and the calculation of energy dissipated on the frequency bands the Matlab wavelet instrument was used. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index