An artificial immune system approach for fault detection in the stator and rotor circuits of induction machines

In this paper, an artificial immune system approach to the detection and diagnosis of faults in the stator and rotor circuits of an induction machine is presented. The proposed technique requires the measurement of two stator currents to compute their alpha beta representation before and after a fault condition. It is verified that for different faults, different patterns are generated by the vector alpha beta representation, helping to construct a characteristic image of the operating condition of the induction machine. A pattern recognition algorithm inspired by how the immune system operates throughout the body is proposed to identify and classify the fault condition. According to the proposed methodology, there is no need to know the details of machine operation in a certain regime and all phenomena and effects resulting from the machine operating in this regime are taken into account. Several experimental results obtained on 2.2 kW and 3.2 kW three-phase induction machines are presented and discussed to validate the methodology, verifying its good performance in preventive fault detection. (C) 2010 Elsevier B.V. All rights reserved.