PSI - Issue 32

I.O. Glot et al. / Procedia Structural Integrity 32 (2021) 216–223 Shestakov A.P./ Structural Integrity Procedia 00 (2021) 000 – 000

222

7

looks like Fig. 9a. Thin lines existing over the entire time interval reflect electrical noise. To eliminate them, it is necessary to average the spectrogram over a time interval without technological operations. Then subtract the resulting function from the entire spectrogram. The result is a spectrogram (Fig. 9b) without electrical noise constant in time. 7. Vibration distribution along the structure during dropping of ore Distribution of vibrations over the structure during dropping of ore is shown in Figure 10. It shows spectrograms for sensors evenly distributed along the height of the structure. These sensors are marked on the scheme with green circles with the corresponding numbers (Fig. 10a). According to the figures, the maximum vibrations are observed in the vicinity of the HFS5 and HFS7 sensors. These sensors are closest to the place of dropping of ore. They record vibrations in the entire frequency range available for the used measuring system. As the distance from the vibration source, high-frequency components are rapidly damped signal. On the HFS4 sensor, vibrations at frequencies above 1 kHz are poorly represented, and on the next-farther sensors HFS3 and HFS1 they are absent. These sensors are not very informative for constructing diagnostic criteria, since the dominant part of the spectrum is not represented in them. One of the options for increasing the information content of sensors remote from natural sources of vibration is to install on the structure of actuators. Actuators will allow you to determine the spectral characteristics of the structure at all measuring points and at the entire measuring range. 8. Discussion The main reason for the destruction of metal structures is metal corrosion. Under the influence of corrosion, structural elements are destroyed, as well as their bolted and welded joints. As a rule, the destruction is local in nature. In such conditions, the use of vibration diagnostics methods is associated with a number of problems.

Fig. 10. (a) sensor locations; (b) spectrograms of signals from sensors HFS1, HFS3, HFS4, HFS5, HFS7, HFS8 during dropping of ore.