Electromagnetic signature analysis of inter turn short circuit and Broken rotor bar fault in induction Motor drives

Abstract

Fault Diagnosis of Induction Motor is gaining importance in industry because of the need to newlineincrease reliability and to decrease possible loss of production due to machine newlinebreakdown. Different faults occur in induction motor, owing to environmental stress or various others reasons. Many researchers proposed different techniques for fault detection and diagnosis. This thesis deals with electromagnetic signature analysis of broken rotor bar fault and stator inter-turn fault at the incipient stage in a closed loop speed controlled threephase Induction Motor. In recent years, inverter fed induction machines for variable speed drives are spreading in many industrial processes and applications and the detection of faults becomes a complex issue. The machine terminal quantities which are usually monitored for diagnostic purposes are inevitably influenced by the control technique used for the system. Also, the fault analysis techniques developed for mains fed induction motors and open loop drives cannot be used, since the control strategy of the drive changes the effects that the fault would introduce in terminal quantities, hence do not allow proper detection of faults. One of the major newlinechallenges in the field of induction motor fault analysis is the detection of fault at the incipient stage. The analytical study of rotor broken bar fault and stator inter-turn fault at the incipient stage in a closed loop inverter fed three phase induction motor is done in this work. In the case of broken rotor bar fault, the induction motor is modelled in ANSYS Maxwell using Finite Element Analysis (FEA) tool incorporating one, two and three broken bars. The newlineinduction machine in ANSYS Maxwell is co-simulated with ANSYS Simplorer for power newlinesupply and loading arrangements. It is fed from a 3 phase PWM inverter operating with newlineclosed loop V/f speed control strategy implemented using MATLAB Simulink and cosimulated with Simplorer.