OPTIMIZED NEURAL NETWORK RELAYING FOR EFFICIENT POWER TRANSFORMER PROTECTION

Abstract

One of the most important devices in a power system is a Power transformer. However, the main issues in power transformers are stability and dependability. The automation system’s transformer relaying protection is very important for operating transformers safely. So, it is imperative that continuity in the operation of transformers is required to ensure dependability of power. Any repair work that is not as per schedule more so in cases of faulty transformers being replaced it becomes time and money consuming. A milestone in the development of Fourier transform which is the Wavelet Transform (WT) has now gained a lot of popularity and is being successfully applied in the various fields. In solving engineering modeling problems, the Neural Network method (NN) is being successfully applied. When searching for parameters in any function for citing an optimal solution such a search is known as optimization in which all possible values and solutions are considered among which the one with the extreme value is accepted as the optimal solution. Firefly Algorithm (FA) is considered very effective and is producing good optimal solutions. In this work, the FA is duly hybridized along with hill climbing algorithm for improving the efficiency of NN for improved power transformed protection