Implementation of neural network and genetic algorithms for novelty filters for fault detection

In this paper a method of detecting shorted turns in rotating machines using computational intelligence techniques (neural network and genetic algorithm) is presented. The methods of signal processing and detection of faults in operating machines is discussed. The use of novelty filters for the detection of shorted turns and mechanical failures in operating machines is described. Genetic algorithm have been used to train the neural network to enhance the capabilities of the novelty detector neural network. The proposed techniques have been applied on an induction machine and the simulation results have been presented to show the effectiveness of the proposed technique.