Generic structural model of machinery vibroacoustic signal

Abstract

The paper deals with the challenges of the machinery vibroacoustic signal modeling. The purpose of this paper is to form a phenomenological model for the structure of vibroacoustic oscillations and signals appeared in case of malfunctions and defects. It presents particular structural models for vibroacoustic signals of some reciprocating machine malfunctions. Mathematical description of the vibroacoustic signal reveals the informative parameters of the malfunction diagnostic signs. The model enables us to define in detail the informative vibroacoustic signal parameters. The model also allows us to select the diagnostic signs of the given malfunctions and defects in certain machine parts and components. At the same time, it is considered that the signal is received in certain points on the machine taking into account the excitation sources, conditioning channel and propagation of the vibroacoustic oscillations. The model has been constructed considering the presence and interaction of the three main sources of excitation of the asset vibroacoustic oscillations, which are identified by the types of excited oscillations. It is assumed that there are damped natural, undamped induced, random broadband and narrowband vibrations. The parameters of the vibroacoustic signal structural model also take into account the properties of converting the vibroacoustic oscillation to the vibroacoustic signal by means of a piezoelectric sensor. The model shows that the vibroacoustic signal has not only the components from each source, but their mutually modulated components. We developed partial models for mechanisms of structuring the vibroacoustic signal appeared in case of malfunctions and defects in rotating parts and couplings, machine shaft and drive shaft misalignment, increased clearances and other defects and failures of the rotating and reciprocating parts and components. The proposed phenomenological structural models of the vibroacoustic signals appeared in case of malfunctions of reciprocating machines show that for the diagnostic purposes it is necessary to analyze vibroacoustic signals either in the time domain according to the cyclogram of the reciprocating machine operation or in the envelope spectrum at characteristic frequencies or the noise component. In the event of defects and malfunctions in the centrifugal rotating parts and components, there are excited the vibroacoustic signals, which are advisable to analyze in the direct spectrum or in the spectrum of the noise component envelope. The proposed models enable us to select and form scientifically grounded systems of the machine failure diagnostic signs.