КРАТНОМАСШТАБНЫЙ ВЕЙВЛЕТ-АНАЛИЗ ПРОФИЛОГРАММЫ

Abstract

The functional properties of the surface are significantly affected by its microgeometry, therefore it is necessary to optimize the parameters of the microgeometry of the surface of metal structures. The analysis and evaluation of the microgeometry of the surface allows an assessment of the fatigue strength of metal structures. In order to study the dependence of the durability of thin-sheet structures on the degree of corrosion destruction, the microgeometry of the surface was analyzed using a profilometer 130 in accordance with GOST 25142-82. Low-carbon cold-rolled steel 08kp was used as the test material. The surface roughness of the samples was Ra = 0.22 microns. The roughness parameters were calculated according to GOST 2789-73. A profilogram representing a discrete series of values of peaks and troughs of the relief of the metal plate surface was studied. To assess the microgeometry of the surface, a wavelet analysis was used, which allows processing signals that are nonstationary in time or inhomogeneous in space. The signal in the form of successive approximations has a trend, cyclic components and local features (fluctuations) around the components of the signal. The locality property of wavelets gives advantages over the trigonometric Fourier transform: the sines and cosines are defined on the entire numerical axis, and the wavelets have a compact carrier. For multiscale analysis, a fast cascade algorithm of calculations has been developed by analogy with the fast Fourier transform. In the study, the signal was decomposed to the 9th level. The reconstructed detailing coefficients represent high-frequency and low-frequency fluctuations. The decomposition of the signal to the 9th level made it possible to analyze the surface roughness of the 08kp steel using a multiple-scale wavelet analysis.