A Reliable Protection Scheme for Smartgrid Based On Wavelet Transform and Ensemble Classifier

Abstract

The advent of the “Smartgrid” is a result of the new trend of managing the increasing power demand by harnessing the potential of distributed energy resources (DER) within the current low voltage distribution framework. Even while the distribution networks benefit greatly from the Smartgrid, the integration problems with renewable and synchronous DERs have rendered protection jobs inherently difficult. When compared to synchronous DERs, which can endure fault current up to 10 times, renewable DERs' dynamic producing behaviour, which is restricted to carrying fault current up to 2–3 times the rated current, creates a clear distinction between the two in terms of operating circumstances. As a result of these factors, the accompanying security challenges become even more difficult. Because the magnitude of fault current varies according to mode, there is a significant variation in the amount of fault current between islanded and grid-connected modes. As a result, it is necessary to develop an accurate and dependable approach that can operate in both modes and fulfil the Smartgrid's protection needs without being influenced by harmonics associated with power electronic interface devices and non-linear loads. In this context, a protection strategy for the Smartgrid has been presented that is based on the joint framework of discrete wavelet transform (DWT) and Bagged decision tree based data-mining approach.