PRACTICAL APPLICATION OF THE METHOD OF AUTOMATIC ADAPTATION OF VIBRATION DIAGNOSTICS SIGNALS TO THRESHOLD VALUES

Abstract

The aim of the article is to test the developed methods, which are part of the complex method of vibration diagnostics of the main power plant of water transport means. This goal is achieved by in-depth analysis of the theoretical foundations of vibration diagnostics of control objects, as set out in modern sources of information, as well as direct application of the developed theoretical methods for conducting a comprehensive vibration diagnostic of the main power plant of water transport. It has been proved that the simplest way is to analyse the general levels of standard frequency bands when determining the vibration state of equipment. In this case, the general levels of standard frequency bands are compared with the threshold values set for different types of machines. The close relationship between these values and the thresholds given in the annexes of the standards for the respective types of equipment and obtained during operation is emphasised. In the process of monitoring the general technical condition of the main power plant of waterborne transport means, a set of octave and third-octave filters with a constant relative width, which form a broadband spectrum, is most often used. When analysing the spectra of sub-octave bands, the level of each band is compared with threshold values, which requires additional processing. The most significant result of the study is the establishment of the fact that the developed method allows detecting a change in state either at the same time or earlier than the standard approach. It is confirmed that the result is especially effective for signals without transients and for high degradation rates that exceed the rate of maximum fluctuations, which is important for operational control for the purpose of managing critical equipment. The proposed method of constructing thresholds allows to increase the speed of response to growth associated with accelerated degradation. At the same time, the advantage of approximation on the interval corresponding to the latest trend is proved.