PSI - Issue 50

R.S. Akhmetkhanov et al. / Procedia Structural Integrity 50 (2023) 11–16 R.S.Akhmetkhanov / Structural Integrity Procedia 00 (2019) 000 – 000

13

According to the data of Table 1, variants of mixing at t=90 and t=120 minutes are close by statistical data, but the preferable variant is at t=90 minutes, because statistical characteristics are less (mean radius, standard, interquartile range). Let's consider these variants from the position of fractal theory, let's compare multifractal spectra of images of obtained composite material (see fig.3). Multifractal spectrum f (  ) reflects scale ordering of image structure - grain size distribution, in our case - filler grains. The greater the width of the spectrum, the greater the number of significant scale elements in the set under consideration (Schroeder M., 2001, .Pavlov A.N., at etc., 2007.). All three spectra are multifractal; the initial spectrum contains fewer significant scale elements; during mixing, new significant scale elements (grains) arise. And the latter variants are close in the number of significant scale elements (grain sizes and their number) in the material structure.

Fig. 2. Distribution of the number of filler grains N by their size R m (average radius of the equivalent disk) at wave mixing technology of composite material: a - t=1 minute; b - t=15 minutes; c - t=60 minutes; d - t=90 minutes; e - t=120 minutes.

Multifractal spectra for materials at 90 min and 120 min treatment differ in the length of the tails of the function f(  ); at 120 min treatment time, structural significant elements (grain size) of large and small sizes appear. Two