Khodja, Djalal EddineChetate, Boukhmis2015-06-292015-06-292008SPEEDAM 2008 - International Symposium on Power Electronics, Electrical Drives, Automation and Motion; Ischia; Italy; 11 June 2008 through 13 June 2008; Category number08EX2053; Code 73530978-142441664-6https://dspace.univ-boumerdes.dz123456789/2128In this work the strategy of the artificial intelligence (neural networks) is used to detect and localize the defects of the double stator asynchronous machine. In fact, several neural networks have been applied to the detection of defects. Then, we used a selector which allows activating only one network at a time. In this case, the selected network detects only defects corresponding to the torque developed by asynchronous machine. Finally, the simulation results were presented to show the effectiveness of artificial neural networks for automatic fault diagnosisenArtificial Neuron Networks (ANN)DetectionDouble stator asynchronous machineFailureRoot Mean Square (RMS)Artificial intelligenceBackpropagationElectric fault currentsIndustrial engineeringPower convertersPower electronicsStatorsArtificial neural networksAutomatic fault diagnosis;Electrical drivesInternational symposiumSimulation resultsNeural networksArticle