Abstract
The transients in power system cause serious disturbances in the reliability, safety and economy of the system. The transient signals possess the nonstationary characteristics in which the frequency as well as varying time information is compulsory for the analysis. Hence, it is vital, first to detect and classify the type of transient fault and then to mitigate them.
This article proposes time-frequency and FFNN (Feedforward Neural Network) approach for the classification of power system transients problems. In this work it is suggested that all the major categories of transients are simulated, de-noised, and decomposed with DWT (Discrete Wavelet) and MRA (Multiresolution Analysis) algorithm and then distinctive features are extracted to get optimal vector as input for training of PNN (Probabilistic Neural Network) classifier.
The simulation results of proposed approach prove their simplicity, accurateness and effectiveness for the automatic detection and classification of PST (Power System Transient) types.