基于小波包的2D-HMM旋转机械升速过程故障诊断

Fault Diagnosis in Speed-up Process of a Rotating Machine Based on Wavelet Packet and 2DHMM

利用小波包分解对信号进行精确细分的特点 ,构造出相应的能量谱作为旋转机械运行状态的特征向量 ,并以此作为 2 D HMM的输入进行训练 ,建立了基于 2 D HMM的旋转机械运行状态分类器 ,用以识别机组状态。最后通过 Bently-

The energy spectrum from the wavelet packet decomposition can be used as feature vectors of running state of a rotating machine to train 2DHMM. So the classified model based on 2DHMM was constructed. The new model was tested with the experimental data collected from a BentlyNevada rotor experimental system and the result demonstrated that the model was effective to classify classical faults in speed up process of a rotating machine.