An islanding detection algorithm for distributed generation based on Hilbert-Huang transform and extreme learning machine

This study presents a novel method to detect an islanding condition in a distribution system with distributed generations (DGs). The proposed approach is based on Hilbert-Huang transform (HHT) and Extreme learning machine (ELM). The system taken for testing of the proposed method consists of different types of DGs like hydro turbine generator with synchronous machine and wind turbine generator with asynchronous machine. The analysis starts with extracting the non-stationary three phase voltage signals at the target DG end and decomposed into mono component signals, called intrinsic mode function (IMF), by the empirical mode decomposition (EMD) method. In the next step, the amplitude, phase angle and frequency of the components are computed by applying the HHT to each IMF. Then, the different distinguish features are calculated such as, energy, standard deviation of phase and amplitude to track the islanding condition from different non-islanding conditions like single line to ground fault, line to line fault, three phase fault, voltage sag, voltage swell, sudden load change, capacitor switching and other DG tripping etc. To test the accuracy of proposed method, a modified ELM classifier is developed based on the feature index. It has been found that the proposed HHT-ELM technique is highly successful to discriminate islanding events under a wide range of operating conditions from the other type of disturbances in the power distribution network. The proposed scheme is simulated by the MATLAB/SIMULINK environment.