%0 Journal Article %A CHENG Yanwei %A XIE Yongcheng %A LI Guangsheng %T Research on fault diagnosis of vehicle power system %D 2012 %R %J Computer Engineering and Applications %P 245-248 %V 48 %N 1 %X By analyzing the signal characteristics of vehicle power system, fault diagnosing method is proposed, which is combined with wavelet packet and hidden Markov model. With the features extracted from the various states of power system by wavelet packet decomposition, it selects the initial value of HMM by improved K means algorithm based on the simulated annealing, trains CHMM with the feature vector, which is used for condition monitoring and fault diagnosis of power system. The results show that it has accurate diagnosis through small training sample.
%U http://cea.ceaj.org/EN/abstract/article_27618.shtml