PRINCIPAL COMPONENT ANALYSIS IN THE PROCESSING OF MULTI-PARAMETER ACOUSTIC SIGNALS OF THE MIRROR-SHADOW METHOD OF BAR TESTING
O. V. Murav’eva1,2, V.A. Tenenev1, A.F. Brester1, K.Y. Belosludtsev1
1Kalashnikov Izhevsk State Technical University, Izhevsk, Russia
2Udmurt Federal Research Center, Ural Branch, Russian Academy of Sciences, Izhevsk, Russia
Keywords: acoustic mirror-shadow method, statistical parameters, defect, principal component analysis
Abstract
The paper provides a justification for the use of the principal component method for assessing the generalized characteristics of the defect in the processing of multi-parameter acoustic signals of the mirror-shadow multiple reflections method of bar stock testing. The method allows reducing the number of signal parameters in the formation of rejection criteria, developing a methodology for assessing the generalized characteristics of the defect, and forming a complex rejection criterion based on the unacceptable value of the generalized defect characteristic for objects made of any steel grades and of any diameters. The results of approbation of the proposed approach for evaluation of the generalized characteristic of natural defects using the regression dependence produced are satisfactorily consistent with the results of metallographic studies.
|
