ANN Based Approach for Detection of Islanding Using HOG Features

In this paper, a novel approach for detecting islanding based upon image categorisation using Artificial Neural Network (ANN) is proposed. Images are used to derive the histogram of oriented gradient (HOG) characteristics so as to identify non-islanding as well as islanding events. In the suggestive method, spectrogram images are obtained from the time-series signal which are derived from the point of common coupling. These images are utilised to extract the features using histogram of oriented gradient, which is then utilised as a feature input for the ANN model for training as well as testing. It uses rate of change of negative sequence voltage parameter for deriving spectrogram image. The results from classification demonstrates that detection of islanding using ANN method and HOG features being used for image classification achieved excellent results with 97.3 % accuracy and mean square error of 0.0002.