versión On-line ISSN 0718-0764
VILLADA, Fernando y CADAVID, Diego R. Fault Diagnosis in Induction Motors Using Artificial Neuronal Networks. Inf. tecnol. [online]. 2007, vol.18, n.2, pp. 105-112. ISSN 0718-0764. doi: 10.4067/S0718-07642007000200016.
A new algorithm to diagnose inter-turn faults in induction motors based on Artificial Neural Networks (ANN) is presented in this work. A machine model able to simulate internal faults under different load conditions and voltage unbalance was implemented and tested, in order to generate the training patterns of the ANN. An electrical network analyzer and a digital signal processor (DSP) are used to show the implementation of the method. Experimental results in a 2 Hp and 3 Hp induction motors show the robustness of the algorithm allowing detect incipient faults and its implementation feasibility at industrial plants.
Palabras clave : Artificial neural networks; induction motors; stator faults; diagnosis.