A new approach to broken rotor bar detection in squirrel cage induction motor via model-based parameter estimation
Academic Article
Overview
Additional Document Info
View All
Overview
abstract
This paper studies a novel approach for the detection of broken rotor bars in induction motors. The hypothesis on which the detection is based is the model-based parameter estimation technique. A high order transfer function is obtained for the induction motor. This transfer function is Introduced between the stator voltages as inputs and stator currents as outputs. Then to detect the broken bars, stator voltages and stator currents have been measured at nominal torque and speed. This measurement processed by a near-least-square-error estimation and Auto Regressive Moving Average with Auxiliary input (ARMAX) to produce estimated motor state and parameter. In particular, estimated coefficient of appropriate motor transfer function has been compared with its nominal value during healthy operation. Estimation of the coefficient is presented experimentally for three different motor cases; healthy, two and three broken bars by milling into the 3-hp squirrel cage rotor. Simulation and experimental results show the effectiveness of the proposed method for broken rotor bar detection in squirrel cage induction motors.