Intelligent monitoring of Synchronous Generators in Smart Grids using Deep Neural Network

Le résumé fourni par la source

Smart cities are on the rise and any component of such infrastructure would require reliable electrical power and hence reliable Synchronous generators. To reduce the down time of synchronous generators, it is necessary to implement an intelligent condition monitoring system. Stator inter turn fault is a critical fault as it is the beginning stage form many electrical faults in a Synchronous generator. Neural networks have proved their mettle in solving difficult classification problems. Here the stator interturn fault diagnosis problem has been cast as a multilabel, multiclass classification problem. It has been solved using neural networks using the Keras framework. The data for training the network was obtained from a Finite element model of a synchronous generator implemented in ANSYS Maxwell.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.