Identificação de falhas de curto-circuito de estator em motores de indução trifásicos utilizando evolução diferencial

Abstract

The squirrel cage three-phase induction motor is the main electromechanical conversion machine due to its robustness and easy maintenance, therefore it is indispensable in the industrial production processes. Due to its great importance, surveys related to its fault identification are conducted constantly, in order to reduce the corrective maintenance rates and allow a scheduled maintenance, reducing the losses in the production process, due to unexpected stops. With this purpose, this work proposes the study and development of an alternative methodology based on Differential Evolution algorithm to identify stator short-circuit failures in induction motors connected directly on the electric grid, through its voltage and current signals acquired in the time domain. This Differential Evolution algorithm is used to estimate the induction motor electric parameters based on its equivalent electric circuit model and the fault identification occurs in result of the calculation of the estimated magnetization inductance percentage variation considering a healthy motor. The database used for this work was obtained through laboratory experiments performed with two different types of 1 CV motor and a 2 CV motor under different conditions of torque variation and unbalanced voltages.