Neighbor Machines Effects on the Efficiency of Electromagnetic Signatures-Based Diagnosis Techniques

Over the years, many researches have been focused on the suitable processing of the electromagnetic data for health assessment of the mechanical and electrical parts of electrical machines. These efforts have been devoted to trying to replace conventional vibration analysis techniques that are known as efficient yet expensive. In this paper, we discuss the efficiency of these alternate monitoring techniques when adopted in real industrial environments where there are many coexisting electrical machines. In that respect, we consider the probable effect of neighbor machines on the electromagnetic signature of an induction motor running under several load conditions. The experimental results obtained on a laboratory test-bench show that the call for electromagnetic signatures-based diagnosis techniques, requires special attention if adopted in industrial environment.