Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox

Heyns, THeyns, PSDe Villiers, JP2013-01-022013-01-022012-10Heyns, T, De Villiers, J.P and Heyns, P.S. 2012. Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox. Mechanical Systems and Signal Processing, Vol. 32, pp 200-215.0888-3270http://www.sciencedirect.com/science/article/pii/S0888327012002221http://hdl.handle.net/10204/6408Copyright: 2012 Elsevier. This is the Post-Print version of the work. The definitive version is published in Mechanical Systems and Signal Processing, Vol. 32, pp 200-215This paper investigates how Gaussian mixture models (GMMs) may be used to detect and trend fault induced vibration signal irregularities, such as those which might be indicative of the onset of gear damage. The negative log likelihood (NLL) of signal segments are computed and used as measure of the extent to which a signal segment deviates from a reference density distribution which represents the healthy gearbox. The NLL discrepancy signal is subsequently synchronous averaged so that an intuitive, yet sensitive and robust, representation may be obtained which offers insight into the nature and extent to which a gear is damaged. The methodology is applicable to non-linear, non-stationary machine response signals.enGaussian mixture modelsGMMsGearbox monitoringCombining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearboxArticleHeyns, T., Heyns, P., & De Villiers, J. (2012). Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox. http://hdl.handle.net/10204/6408Heyns, T, PS Heyns, and JP De Villiers "Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox." (2012) http://hdl.handle.net/10204/6408Heyns T, Heyns P, De Villiers J. Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox. 2012; http://hdl.handle.net/10204/6408.TY - Article AU - Heyns, T AU - Heyns, PS AU - De Villiers, JP AB - This paper investigates how Gaussian mixture models (GMMs) may be used to detect and trend fault induced vibration signal irregularities, such as those which might be indicative of the onset of gear damage. The negative log likelihood (NLL) of signal segments are computed and used as measure of the extent to which a signal segment deviates from a reference density distribution which represents the healthy gearbox. The NLL discrepancy signal is subsequently synchronous averaged so that an intuitive, yet sensitive and robust, representation may be obtained which offers insight into the nature and extent to which a gear is damaged. The methodology is applicable to non-linear, non-stationary machine response signals. DA - 2012-10 DB - ResearchSpace DP - CSIR KW - Gaussian mixture models KW - GMMs KW - Gearbox monitoring LK - https://researchspace.csir.co.za PY - 2012 SM - 0888-3270 T1 - Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox TI - Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox UR - http://hdl.handle.net/10204/6408 ER -