Recognition of Acoustic Signals of Induction Motors with the Use of MSAF10 and Bayes Classfier

Abstract

Condition monitoring of deterioration in the metallurgical equipment is essential for faultless operation of the metallurgical processes. These processes use various metallurgical equipment, such as induction motors or industrial furnaces. These devices operate continuously. Correct diagnosis and early detection of incipient faults allow to avoid accidents and help reducing financial loss. This paper deals with monitoring of rotor electrical faults of induction motor. A technique of recognition of acoustic signals of induction motors is presented. Three states of induction motor were analyzed. Studies were carried out for methods of data processing: Method of Selection of Amplitudes of Frequencies (MSAF10) and Bayes classifier. Condition monitoring is helpful to protect induction motors and metallurgical equipment. Further researches will allow to analyze other metallurgical equipment.