Abstract
This paper provides an intelligent diagnosis method for gear damage based on techniques of empirical mode decomposition (EMD) and support vector machines (SVMs). By the data processing of EMD, the original signal is decomposed into a finite set of intrinsic mode functions (IMFs). The characteristic energy ratios of IMFs are acquired as representative parameters of the signal. Furthermore, statistical parameters such as standard deviation, root mean square value, maximum value, mean value, crest factor and shape factor are extracted from the original signal in time-domain. The characteristic energy ratios and statistical parameters are combined as failure feature vectors to be input to the multiple classifiers developed from SVMs for gear damage diagnosis. The validity of the presented method is confirmed by the application of monitoring the gear conditions during the pitting experiments, and the diagnostic accuracy is 72.7%.
