Identificação inteligente de falhas em máquinas elétricas utilizando informação mútua

Abstract

Three-phase induction motors are widely employed in industrial production processes due to their low acquisition cost and maintenance and adaptation in front of different conditions of mechanical loads and robustness for use in harsh environments. Diagnosing an incipient fault when these motors are in operation, even before significant faults may occur, should receive special attention from the industry. Thus, a proper diagnosis can reduce losses in the processes, time out of service, and the equipment breaks down and ensures consistent and reliable industrial systems operation. In this context, this paper presents an approach concerning the use of intelligent systems and signal processing tools in the classification of bearing, rotor, and stator fault diagnosis in induction motors, even when they are subject to various operating conditions such as variations in load torque independent of machine power supply. This work proposes the extraction of the fault characteristics based on delayed mutual information measurements between the stator line current signals, 𝑖𝑎, and 𝑖𝑏, in the time domain. To evaluate the classification accuracy under various fault severity levels, the performance of three different pattern recognition techniques were compared: artificial neural network type multilayer perceptron, k-nearest neighbors and support vector machine. The offline and online experimental results presented in this work validate the proposed approach’s robustness and effectiveness.