PSI - Issue 65

I. Shardakov et al. / Procedia Structural Integrity 65 (2024) 233–240 I. Shardakov, R. Tsvetkov, I. Glot, A. Shestakov, G. Gusev, V. Yepin / Structural Integrity Procedia 00 (2024) 000–000

236

4

3. Results

During the loading process, synchronous recording of AE signals was performed by all acoustic sensors. A threshold value was set to select signals for further analysis. If the signal value was below the threshold, the signal was ignored. At the first stage of mathematical processing of the obtained data, an estimate of the frequency spectrum interval of the AE signal was made using the Fourier transform. The frequencies are located in the interval from 20 kHz to 160 kHz. When processing digital AE data, the time of the event was recorded in the table, with the data file position, index in the array, as well as a unique timestamp that allows synchronization of signals from different boards. Next, based on the results of measurements on all acoustic sensors, a number of generally accepted parameters of AE signals were calculated; their interpretation can be found, for example, in the work of Ohno (2010). These parameters are: double maximum amplitude (signal range), integral-amplitude parameter of the signal (E), dominant frequency (corresponding to the maximum modulus of the Fourier coefficient), average frequency AF (by the number of pulses in the signal), arrival time of the signal of one event, signal shape parameter WI, signal rise parameter RA. In addition to this traditional analysis, signal processing using Wavelet analysis was performed. The main relationships for calculating the parameters E (arbitrary units), WI (dimensionless) and RA (s/V), characterizing the AE pulse are the following:

T

1

end f    s t T st

(1)

2

E

U

t

max T T T T   ( end

)

(2)

WI



st

(

)

max

max ( ) st T T 

(3)

RA



U

max

Here U t – voltage on the sensor, V ; T st – time of AE signal start; T end – time of AE signal end; T max – time corresponding to the maximum value of acoustic impulse U max .

Fig. 3. Number of AE pulses on sensor 1 at different loading stages (black bars); time dependence of force under cyclic loading (green curve); the cycle number is shown in the circle