A comprehensive evaluation of intelligent classifiers for fault identification in three-phase induction motors

• Present a comprehensive evaluation of intelligent classifiers to identify stator, rotor, and bearing faults in three-phase induction motors.
• Proposed methodology uses the current signal in time domain as the inputs of the pattern classifiers for fault diagnosis.
• Experimental results gathered from three-phase induction motors operating with different load conditions and fed under unbalance voltage are provided.
• Six different intelligent methods are presented and compared for each proposed fault condition.

Three-phase induction motors are the key elements of electromechanical energy conversion for a variety of industrial sectors. The ability to identify motor faults before they occur can reduce the risks in decisions regarding machine maintenance, lower costs, and increase process availability. This article proposes a comprehensive evaluation of pattern classification methods for fault identification in induction motors. The methods discussed in this work are: Naive Bayes, k-Nearest Neighbor, Support Vector Machine (Sequential Minimal Optimization), Artificial Neural Network (Multilayer Perceptron), Repeated Incremental Pruning to Produce Error Reduction, and C4.5 Decision Tree. By analyzing the amplitudes of current signals in the time domain, experimental results with bearing, stator, and rotor faults are tested using different pattern classification methods under varied power supply and mechanical loading conditions.

Figure optionsDownload as PowerPoint slide