Islanding Detection of Inverter based Distributed Generations using Wavelet Transform and KNN

In recent times, a significant level of advancement has been observed in Distributed Generation (DG) technologies and in its deployment. This scenario demands a mandatory requirement of effective islanding detection techniques to avoid possible damages to the system. This paper proposes an advanced islanding detection method for inverter based DGs using wavelet transform and machine learning. The method uses wavelet transform decomposition to extract the detailed coefficients and are then processed to identify typical transient variations during islanding. K-Nearest Neighbor (KNN) is used to classify the condition as islanding or non-islanding. The method will reduce the Non-Detection Zone (NDZ) and effective detection time. MATLAB/SIMULINK simulations are done on test system to validate the proposed method.