Active Islanding Detection Method Using Wavelet Fuzzy Neural Network

A novel active islanding detection method using d-axis disturbance signal injection with intelligent control is proposed in this study. The proposed active islanding detection method is based on injecting a disturbance signal into the system through the d-axis current which leads to a frequency deviation at the terminal of the RLC load when the grid is disconnected. The feasibility of the proposed method is evaluated under the UL1741 anti-islanding test configuration. The proposed d-axis disturbance signal injection method is intended to achieve a reliable detection with quasi zero non-detection zone (NDZ), minimum effects on power quality and easy implementation without additional sensing devices or equipments. Moreover, to further improve the performance of islanding detection method, a wavelet fuzzy neural network (WFNN) intelligent controller is proposed to replace the proportional-integral (PI) controller used in traditional injection method for islanding detection. Furthermore, the network structure and the on-line learning algorithm of the WFNN are introduced in detail. Finally, the feasibility and effectiveness of the proposed d-axis disturbance signal injection method is verified with experimental results.