abstract
- This paper presents a novel methodology for the detection and diagnosis of rotor faults in induction machines using signal estimation. The proposed approach is independent of the motor slip and relies on the isolation of main harmonics rather than on the investigation of signatures with traditional approaches such as the identification of fault-related sidebands. The method is applied on the stator line current capturing the transient nature of fault-related frequencies at the steady state. Thus, it enables a reliable diagnostic strategy by the isolated harmonics and their analysis over time and frequency. The method's effectiveness was explored with electromagnetic simulations of two induction machines of different geometry, manufacture, and power scale. Then, the method was validated experimentally on a 1.1 kW induction motor. © 1972-2012 IEEE.