Diagnostics of synchronous motor based on analysis of acoustic signals with application of MFCC and Nearest Mean classifier

The paper presents method of diagnostics of imminent failure conditions of synchronous motor. This method is based on a study of acoustic signals generated by synchronous motor. Sound recognition system is based on algorithms of data processing, such as MFCC and Nearest Mean classifier with cosine distance. Software to recognize the sounds of synchronous motor was implemented. The studies were carried out for four imminent failure conditions of synchronous motor. The results confirm that the system can be useful for detecting damage and protect the engines.