Neuro-wavelet based islanding detection technique

Abstract

Connecting distributed generators to the normal radial distribution system improve the power quality and increase the capacity of the electric grid. However, they disturb the radial nature of the network and thus give rise to many problems. Unintentional islanding is one of the encountered problems. In this paper a neuro-wavelet islanding detection technique has been developed. The method is based on the transient voltage signals generated during the islanding event. Discrete wavelet transform is adopted to extract feature vectors which will then be fed to a trained artificial neural network classifier to classify the transients generated as islanding or non-islanding events. The trained classifier was then tested using novel voltage signals. The test results indicate that this approach can detect islanding events with a good degree of accuracy.