Dissertation abstract:

Based on the advantages of non-contact acoustic diagnosis technology, an acoustic image pattern recognizing technique with support vector machine is developed and extended into the fault diagnosis. First, the array microphones are employed to collect mechanical noise signals, and the noise source recognition and location is carried out through the acoustic imaging algorithm. Next, the texture features are extracted through the acoustic images, which are used for fault diagnosis by support vector machine. The good diagnosis results demonstrate that the technology combined feature extraction with support vector machine can be applied to machinery fault diagnosis and provide a novel attacking approach for the application of acoustic images in fault diagnosis.