Abstract: The gearboxes are very important to the transmission system of a helicopter, so it is necessary to monitor and diagnose their conditions and faults. Because of the merit of hidden Markov model (HMM) that has the ability to deal with continuous dynamic signals and the merit of support vector machine (SVM) with perfect classify ability, the HMM-SVM based diagnostic method is presented. With the features extracted from vibration signals by wavelet packet decomposition, the HMM-SVM diagnostic model is trained and used to monitor and diagnose the gearbox’s conditions and faults. The results show that this proposal method is better than HMM-based and SVM-based diagnosing methods in high diagnostic accuracy with small training samples.

Key words: hidden Markov model, support vector machine, wavelet packet, fault diagnosis, gearbox