基于HMM-SVM的故障诊断模型及应用

针对直升机减速器故障诊断中机器学习方法存在的问题，根据隐马尔可夫模型（HMM）适合于处理连续动态信号与支持向量机（SVM）适合于模式分类的长处，提出了基于HMMSVM串联结构的故障诊断模型。通过从减速箱振动信号中有效提取AR特征，利用HMM汁算未知信号与减速器各状态的匹配程度，形成特征向量提供给SVM最后判别，实验结果表明该方法优于单纯的HMM或SVM诊断方法，能利用少量训练样本有效地完成直升机减速器的故障诊断。

Hybrid HMM and SVM Approach for Fault Diagnosis

Because of the problems of machine learning in fault diagnosing of the helicopter's gearbox and the merit of hidden Markov model(HMM) that have the ability to deal with continuous dynamic signals and the merit of support vector machine(SVM)with perfect classifying ability,HMM-SVM based diagnosing method is presented.With the features based on the reflection coefficients of AR model extracted from vibration signals,HMM was used to calculate the matching degree among the unknown signal and the gearbox's states,which formed the features for SVM to diagnosis.The result shows that this proposal method is better than HMM-based and SVM-based diagnosing methods in higher diagnostic accuracy with small training samples.