Fault diagnosis of inter-turn short-circuit in rotor windings based on artificial intelligence

The inter turn short-circuit in rotor windings take the induced electromotive force, which is detected by detecting coil, as a study object. And a method of fault diagnosis based on Wavelet analysis and neural network is presented. The induced electromotive force is analyzed by wavelet packet, which can decompose and construct the energy eigenvectors. Then set up the neural network and use the energy eigenvectors as the input vector of neural network. The method can correctly locate singularity that appears on the measured potential signal to diagnose faulting slot correspondingly. The simulated experimental results show that the artificial intelligence method combining detection coil can detect the inter-turn shorted-circuit fault.