Acoustic particle velocity for fault detection of rotating machinery using tachless order analysis
Most conventional acoustic techniques for fault diagnosis are based primarily on the analysis of the sound pressure emitted by certain device or machine. Reverberant measurement environments with high levels of background noise are typical scenarios for undertaking the evaluation tests. However, these conventional conditions strongly limit the performance of non-contact solutions which require the use of pressure microphones. In contrast, acoustic particle velocity sensors provide a better signal to noise ratio when the measurement is performed close to a radiating surface due to its vector nature, directivity and intrinsic dependency upon surface displacement. This paper introduces acoustic particle velocity transducers for fault detection and classification applications, implementing tachless order analysis and Gaussian mixture models. The experimental results achieved provide experimental evidence of the potential of particle velocity-based solutions for end of line control applications.
Carrillo Pousa, G., Fernandez Comesana, D. and Wild, J., 2015, August. Acoustic particle velocity for fault detection of rotating machinery using tachless order analysis. In INTER-NOISE and NOISE-CON Congress and Conference Proceedings (Vol. 250, No. 1, pp. 6074-6081). Institute of Noise Control Engineering.