Hojatolah Makvandi - Department of Electrical Engineering, Dezful Branch, Islamic Azad University, Dezful, Iran
Mahmood Joorabian - Department of Electrical Engineering, Faculty of Engineering, Shahid Chamran University of Ahvaz, Ahvaz, Iran
Hassan Barati - Department of Electrical Engineering, Dezful Branch, Islamic Azad University, Dezful, Iran

The present study introduces a new extensive-area ANFIS (Adaptive Neuro-Fuzzy Interface System)-based method to detect wide area instability and control the time of controlled islanding execution within power systems. The ANFIS parameters are optimized by the PSO method to increase the method’s accuracy at various disturbances and loading circumstances. In addition, to take various stability margins within the areas into account, a novel parallel ANFIS network (P-ANFIS) is implemented in which a distinct ANFIS is allocated for every nearby area. Extended off-line studies are performed to train ANFIS to respond in real-time accurately based on the selected wide area input signals. These parameters are monitored continuously through a wide area measurement system (WAMS) and the proposed P-ANFIS starts to assess the stability between related areas in real-time in the case of potentially unstable oscillations. Once an unstable oscillation is detected, the islanding command is transmitted to perform the controlled islanding scheme. The suggested technique is used in an IEEE ۳۹ bus power system and its performance is demonstrated at different disturbances in terms of both speed and accuracy. It is found that the suggested ANFIS-based technique can determine islanding requirement and its time of execution properly at different disturbances.

WAMS, islanding detection, ANFIS, PSO