Fault diagnostic in power system using wavelet transforms and neural networks

This paper presents a new approach to Fault detection and diagnosis in power system. Discrete wavelet transformations (DWT) combined with neural networks (NN) have been applied to a typical three phase inverter. A set of faults have been examined, such as inverter IGBT open-circuit fault, leg open fault. The input signals of this algorithm are the three-phase stator currents. Identification and classification uses approximation and details at levels 6 of these currents. The results of simulation show that the proposed technique can accurately detect identify and classify effectively the faults of interest in the power system.