Adaptive neuro-fuzzy inference system (ANFIS) islanding detection based on wind turbine simulator

This paper presents a passive islanding detection method based on means of a neuro-fuzzy approach for wind turbines. Several methods based on passive and active detection scheme have been proposed. While passive schemes have a large Non detection zone (NDZ), concern has been raised on active method due to its degrading power quality effect. Reliably detecting this condition is regarded by many as an ongoing challenge as existing methods are not entirely satisfactory. The proposed method is based on voltage measurements and processing of the hybrid intelligent system called the Adaptive neuro-fuzzy inference system (ANFIS) for islanding detection. This new method based on passive methods will help to reduce the NDZ without any perturbation that deteriorates the output power quality opposite active methods. This method detects the islanding conditions with the analysis of these signals. The studies reported in this paper are based on an experimental system (wind turbine simulator). The results showed that, the ANFIS-based algorithm detects islanding situation accurate than other islanding detection algorithms. Moreover, for those regions which are in need of a better visualization, the proposed approach would serve as an efficient aid such that the mains power disconnection can be better distinguished.