Application Of Neural Network For Transformer Protection

Abstract

The demand for a reliable supply of electrical energy for the exigency of modern world in each and every field has increased considerably requiring nearly a no-fault operation of power systems. The crucial objective is to mitigate the frequency and duration of unwanted outages related to power transformer puts a high pointed demand on power transformer protective relays to operate immaculately and capriciously. The high pointed demand includes the requirements of dependability associated with no false tripping, and operating speed with short fault detection and clearing time. The second harmonic restrain principle is widely used in industrial application for many years, which uses discrete Fourier transform (DFT) often encounters some problems such as long restrain time and inability to discriminate internal fault from magnetizing inrush condition. Hence, artificial neural network (ANN), a powerful tool for artificial intelligence (AI), which has the ability to mimic and automate the knowledge, has been proposed for detection and classification of faults from normal and inrush condition. The wavelet transform(WT) which has the ability to extract information from transient signals in both time and frequency domain simultaneously is used for the analysis of power transformer transient phenomena in various conditions. All the above mentioned conditions of power transformer to be analysed in a power system are modelled in MATLAB/SIMULINK environment. Secondly the WT is applied to decompose the different current signals of the power transformer into a series of detailed wavelet components. The statistical features of the wavelet components are calculated and are used to train a multilayer feed forward neural network designed using back propagation algorithm to discriminate various conditions. The best suitable architecture of ANN is selected having least mean square error during training. The ANN model is implemented in LabVIEW environment.