摘要
在航空发动机的各式故障中,由振动引发的故障占有很大的比重。航空发动机的振动信号中蕴藏了大量的状态及故障信息,因此有必要寻找一种有效的特征提取和故障诊断方法。基于ICA和DHMM的理论方法,形成了ICA-DHMM故障诊断方法。其中ICA用于源信号分离以及特征提取;DHMM作为模式识别工具。通过与ICA-SVM故障诊断方法和传统的DHMM故障诊断方法进行比较,表明本方法有更好的识别效果。
Faults caused by engine vibration account for a large proportion in various faults of aero-engine.The mixed vibration signals of aero-engine contain abundant running information,it is necessary to seek for an efficient way for feature extraction and fault diagnosis.In this paper, a fault diagnosis approach ICA-DHMM is proposed.Independent component analysis（ICA） is used first for extracting feature and then discrete hidden Markov model（DHMM） is used as pattern recognition.Comparing with other fault diagnosis methods,experimental results show that the proposed method has higher recognized accuracy.
出处
《微处理机》
2011年第1期75-79,共5页
Microprocessors
关键词
航空发动机
故障诊断
特征提取
模式识别
Aero-engine
Fault diagnosis
Feature extraction
Pattern recognition
