Application of Hybrid Neural Network to Detection of Induction Motor Electrical Faults

Abstract

Induction motors (IMs) play a key role in industrial drives systems. During motors normal operation, some unexpected damages may occur, resulting in economic losses. Stator windings degradation and rotor broken bars are the most common sources of faults in induction machines. The electrical winding faults, namely the stator inter-turns short circuits and rotor bar damages constitutes around 40% of all faults of the induction motors. Nowadays, faults early detection systems play an essential role in IMs drive control systems. In the aim of faults detection process automation, diagnostic systems are increasingly based on artificial intelligence methods. This paper presents the results of experimental research on the application of axial flux symptoms of the converter-fed induction motor drive to the electrical fault detection and classifications using hybrid neural networks.