Signal Processing in Physical and Engineering Acoustics

2011, Acoustics Today

Abstract
AI

Signal processing is extensively applied in physical and engineering acoustics, playing a crucial role in various domains such as nondestructive evaluation, machinery condition monitoring, and localization of sound sources. This work presents three exemplars of signal processing techniques: leveraging reciprocal wave propagation for flaw detection in structures, employing statistical methods for assessing gearbox conditions through accelerometer data, and enhancing bullet localization using combined sound source information. Key methodologies highlighted include Time Reversal (TR) for source localization and Reverse Time Migration (RTM) for imaging scatterers, showcasing their effectiveness in nondestructive evaluation and geophysical applications.

Figures (10)

cemented wiedmenieietemnied eveenpueewene seater NEO fAppMeu CU LNUEE SERV AOE V YS NEUE BREW CORUER LV EBUREEUUE aS The non-invasive monitoring of the health of gearboxes has been a very active area of research over the past two decades. The capability to predict gear failures from accelerometer - based measurements of the gear meshing vibration signal is of great importance in industries such as the aerospace industry (e.g. helicopter gearboxes). If one can predict failure, then gears can be swapped during regular scheduled maintenance. This prevents accidents and saves money due to unscheduled down time. The main concept in gear health monitoring is that the meshing of the gear and pinion (Figure 3) gives rise to a vibration signal that propagates to the gearbox case where it is measured by accelerometers. The challenge is that, in real systems, many vibration signals arise due to vehicle motion, shaft imbalances, mode shape vibrations of the gearbox case, etc. All of these are also present at the measurement points necessitating the development of signal processing approaches that can isolate and analyze the gear mesh signal. Past condition monitoring techniques of gearboxes have used many different signal processing approaches such as synchronous time series averaging'’, amplitude and phase demodulation’, time-frequency distribution’? and wavelet analysis’’. The use of statistical signal processing approaches have also taken a hold in gear tooth failure detection*'. Non-linear adaptive algorithms for independent component analysis (ICA) have been shown to separate unknown, statistically independent sources that have been mixed in dynamic systems. This example illustrates the application of an information maximization based blind source separation algorithm (a type of independent component analysis (ICA)) to the prediction of gear failures. It is shown that ICA can be used to detect impulsive and random changes in the gear vibration data.

Figure 4. Crack formation and growth at the base of a gear tooth The crack formation at the base of the tooth begins to cause the tooth to flex when the pinion exerts force on the tooth. Because the tooth flexes, there is a slight delay in the rotation of the gear causing a modulation of the meshing vibration signal. It is this modulation that fault monitoring algorithms attempt to detect before the tooth actually fails.

wert Figure 7. Acoustic transient signals — the muzzle blast is generated at the point of fire and the ballistic shock wave originates from the detach point along the bullet’s trajectory. a, >> To an unalerted listener, the sound of gunfire represents an aural event in the soundscape that evokes a reaction of surprise tempered by curiosity. However, in engineering acoustics it represents a transient acoustic signal generated by the discharge of a firearm from which information can be extracted such as the location of the point of fire (/ocalization) and the type of firearm (classification). The sound pulse generated by the discharge of a bullet from a firearm is referred to as the muzzle blast or in military parlance — the report. The acoustic energy propagates at the speed of sound travel in air and expands as a spherical wavefront (of constant phase) centered on the point of fire. Since the propagation of the sound through the atmosphere is omnidirectional, it can be heard from any direction, even behind the firer. If the listener is positioned forward (towards the front) of the firer and the bullet travels at supersonic speed, hen a second transient signal, which is referred to as the (ballistic) shock wave is heard. To the istener, the origin of the shock wave occurs at a point P, (referred to as the detach point), which is located at a distance x, along the trajectory of the bullet — see Fig. 7. When the listener is near he trajectory, the sound pulse (or sonic boom) is extremely loud. Similar to the muzzle blast wave, the shock wavefront travels away from its point of emission at the speed of sound. But, unlike the muzzle blast wavefront, it expands as a conical surface with the trajectory and nose of the bullet defining the axis and apex of the cone respectively. The angle 0, (see Fig. 7) at the apex of the cone is referred to as the Mach angle, whose sine is equal to the reciprocal of the Mach number, which is defined as the ratio of the bullet’s speed V to the speed of sound c”? A casual listener will hear the shock wave before the muzzle blast and instinctively look in the direction of its origin (the detach point) and confuse it with the actual direction of the firer, which coincides with the direction of the muzzle blast that arrives later — see Fig 7.