This paper proposes a diagnosis method that exploits the information delivered by external flux sensors placed in the vicinity of rotating electrical machines, in order to detect a stator interturn short circuit. This fault induces a dissymmetry in the external magnetic field that can be measured by the sensors. Sensitive harmonics are extracted from the signals delivered by a pair of sensors placed at 180° from each other around the machine, and data obtained for several sensor positions are analyzed by fusion techniques using the belief function theory. The diagnosis method is applied on induction and synchronous machines with artificial stator faults. It will be shown that one can obtain high probability to detect the fault using the proposed fusion technique: on various series of measurements, the proposed approach has obtained a 90% detection rate on a considered machine.

This paper presents a statistical methodology for detection of inter-turn short circuit fault in asynchronous and synchronous machines. This methodology uses a correlation coefficient obtained from external magnetic field measured in the machine vicinity. It is a noninvasive method which follows up the signals of two external flux sensors located symmetrically around the machine axis (180° spatially shifted). The principle is based on the calculation of the Pearson correlation coefficient between two signals delivered by two sensors S1 and S2 when the machine operates at different load conditions, which allows us to detect incipient faults in electrical induction and synchronous machines with a high probability of detection. Experimental tests are realized using two specific rewound machines to create inter-turn short circuit faults with different severity levels.