Combined with full lifetime experiment of gearbox, state recognition based on Continue Hidden Markov Model is researched. The flow of State recognition based on Continue Hidden Markov Model using original vibration signal is founded. Virtue and defect of existing classification methods which classify state in full life cycle are analyzed. State number optimization model is established based on K means and cross validation. Gearbox’s operating state is determined by calculating the maximal log-likelihood. The results of recognition show that the method of state recognition based on Continue Hidden Markov Model using original vibration signal is feasible. The results prove that this kind of method is effective.
TENG Hong-zhi;ZHAO Jian-min;JIA Xi-sheng;ZHANG Xing-hui;WANG Zheng-jun .
Research on gearbox State recognition based on continue hidden markov model[J]. Journal of Vibration and Shock, 2012, 31(5): 92-96