Designing of Islanding Detection Relay for Synchronous Generator using Zones Analysis of Two-dimensional Wavelet Signal Energy

This paper presents a novel time-frequency analysis based technique for detection of synchronous generator islanding conditions. In the proposed method the output signals of Distributed Generation are measured locally and analyzed in time-frequency domain. In order to detection of islanding condition, the proposed method is used of energy variations of the rate of change of active power wavelet coefficients signal in a two-dimensional form. By considering the place of signals in any time, the islanding condition could be detected. At first, all possible linear and nonlinear load switching, motor starting, capacitor bank switching, and islanding conditions, are simulated and the required detection parameters measured. Using the discrete wavelet theory, the energy of any decomposition level of all mother wavelet for parameters detection is calculated. From of these signals, the best of them are selected for ANFIS training for islanding. Simulation results confirm the performance of the proposed detection algorithm in comparison with existing methods in islanding and switching conditions.